aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--.clang-format24
-rw-r--r--include/serd/serd.h281
-rw-r--r--src/attributes.h4
-rw-r--r--src/base64.c94
-rw-r--r--src/base64.h2
-rw-r--r--src/byte_sink.h95
-rw-r--r--src/byte_source.c107
-rw-r--r--src/byte_source.h108
-rw-r--r--src/env.c333
-rw-r--r--src/n3.c2668
-rw-r--r--src/node.c493
-rw-r--r--src/node.h19
-rw-r--r--src/reader.c429
-rw-r--r--src/reader.h186
-rw-r--r--src/serd_internal.h16
-rw-r--r--src/serdi.c552
-rw-r--r--src/stack.h86
-rw-r--r--src/string.c223
-rw-r--r--src/string_utils.h109
-rw-r--r--src/system.c41
-rw-r--r--src/system.h12
-rw-r--r--src/uri.c766
-rw-r--r--src/uri_utils.h81
-rw-r--r--src/writer.c1565
-rw-r--r--test/test_env.c94
-rw-r--r--test/test_free_null.c12
-rw-r--r--test/test_node.c239
-rw-r--r--test/test_read_chunk.c117
-rw-r--r--test/test_reader_writer.c529
-rw-r--r--test/test_string.c42
-rw-r--r--test/test_uri.c216
31 files changed, 4918 insertions, 4625 deletions
diff --git a/.clang-format b/.clang-format
new file mode 100644
index 00000000..06635478
--- /dev/null
+++ b/.clang-format
@@ -0,0 +1,24 @@
+---
+AlignConsecutiveAssignments: true
+AlignConsecutiveDeclarations: true
+AlignEscapedNewlinesLeft: true
+BasedOnStyle: Mozilla
+BraceWrapping:
+ AfterNamespace: false
+ AfterClass: true
+ AfterEnum: false
+ AfterExternBlock: false
+ AfterFunction: true
+ AfterStruct: false
+ SplitEmptyFunction: false
+ SplitEmptyRecord: false
+BreakBeforeBraces: Custom
+Cpp11BracedListStyle: true
+IndentCaseLabels: false
+IndentPPDirectives: AfterHash
+KeepEmptyLinesAtTheStartOfBlocks: false
+SpacesInContainerLiterals: false
+StatementMacros:
+ - SERD_API
+ - SERD_DEPRECATED_BY
+...
diff --git a/include/serd/serd.h b/include/serd/serd.h
index 6dc33441..32a60f29 100644
--- a/include/serd/serd.h
+++ b/include/serd/serd.h
@@ -26,54 +26,59 @@
#include <stdio.h>
#if defined(SERD_SHARED) && defined(SERD_INTERNAL) && defined(_WIN32)
-# define SERD_API __declspec(dllexport)
+# define SERD_API __declspec(dllexport)
#elif defined(SERD_SHARED) && defined(_WIN32)
-# define SERD_API __declspec(dllimport)
+# define SERD_API __declspec(dllimport)
#elif defined(SERD_SHARED) && defined(__GNUC__)
-# define SERD_API __attribute__((visibility("default")))
+# define SERD_API __attribute__((visibility("default")))
#else
-# define SERD_API
+# define SERD_API
#endif
#ifdef __GNUC__
-# define SERD_PURE_FUNC __attribute__((pure))
-# define SERD_CONST_FUNC __attribute__((const))
+# define SERD_PURE_FUNC __attribute__((pure))
+# define SERD_CONST_FUNC __attribute__((const))
#else
-# define SERD_PURE_FUNC
-# define SERD_CONST_FUNC
+# define SERD_PURE_FUNC
+# define SERD_CONST_FUNC
#endif
#ifdef __clang__
-# define SERD_NONNULL _Nonnull
-# define SERD_NULLABLE _Nullable
-# define SERD_ALLOCATED _Null_unspecified
+# define SERD_NONNULL _Nonnull
+# define SERD_NULLABLE _Nullable
+# define SERD_ALLOCATED _Null_unspecified
#else
-# define SERD_NONNULL
-# define SERD_NULLABLE
-# define SERD_ALLOCATED
+# define SERD_NONNULL
+# define SERD_NULLABLE
+# define SERD_ALLOCATED
#endif
-#define SERD_PURE_API SERD_API SERD_PURE_FUNC
-#define SERD_CONST_API SERD_API SERD_CONST_FUNC
+#define SERD_PURE_API \
+ SERD_API \
+ SERD_PURE_FUNC
+
+#define SERD_CONST_API \
+ SERD_API \
+ SERD_CONST_FUNC
#ifndef SERD_DISABLE_DEPRECATED
-# if defined(__clang__)
-# define SERD_DEPRECATED_BY(rep) __attribute__((deprecated("", rep)))
-# elif defined(__GNUC__) && __GNUC__ > 4
-# define SERD_DEPRECATED_BY(rep) __attribute__((deprecated("Use " rep)))
-# elif defined(__GNUC__)
-# define SERD_DEPRECATED_BY(rep) __attribute__((deprecated))
-# else
-# define SERD_DEPRECATED_BY(rep)
-# endif
+# if defined(__clang__)
+# define SERD_DEPRECATED_BY(rep) __attribute__((deprecated("", rep)))
+# elif defined(__GNUC__) && __GNUC__ > 4
+# define SERD_DEPRECATED_BY(rep) __attribute__((deprecated("Use " rep)))
+# elif defined(__GNUC__)
+# define SERD_DEPRECATED_BY(rep) __attribute__((deprecated))
+# else
+# define SERD_DEPRECATED_BY(rep)
+# endif
#endif
#ifdef __cplusplus
extern "C" {
-# if defined(__GNUC__)
-# pragma GCC diagnostic push
-# pragma GCC diagnostic ignored "-Wzero-as-null-pointer-constant"
-# endif
+# if defined(__GNUC__)
+# pragma GCC diagnostic push
+# pragma GCC diagnostic ignored "-Wzero-as-null-pointer-constant"
+# endif
#endif
/**
@@ -93,35 +98,35 @@ typedef struct SerdWriterImpl SerdWriter;
/// Return status code
typedef enum {
- SERD_SUCCESS, ///< No error
- SERD_FAILURE, ///< Non-fatal failure
- SERD_ERR_UNKNOWN, ///< Unknown error
- SERD_ERR_BAD_SYNTAX, ///< Invalid syntax
- SERD_ERR_BAD_ARG, ///< Invalid argument
- SERD_ERR_NOT_FOUND, ///< Not found
- SERD_ERR_ID_CLASH, ///< Encountered clashing blank node IDs
- SERD_ERR_BAD_CURIE, ///< Invalid CURIE (e.g. prefix does not exist)
- SERD_ERR_INTERNAL ///< Unexpected internal error (should not happen)
+ SERD_SUCCESS, ///< No error
+ SERD_FAILURE, ///< Non-fatal failure
+ SERD_ERR_UNKNOWN, ///< Unknown error
+ SERD_ERR_BAD_SYNTAX, ///< Invalid syntax
+ SERD_ERR_BAD_ARG, ///< Invalid argument
+ SERD_ERR_NOT_FOUND, ///< Not found
+ SERD_ERR_ID_CLASH, ///< Encountered clashing blank node IDs
+ SERD_ERR_BAD_CURIE, ///< Invalid CURIE (e.g. prefix does not exist)
+ SERD_ERR_INTERNAL ///< Unexpected internal error (should not happen)
} SerdStatus;
/// RDF syntax type
typedef enum {
- SERD_TURTLE = 1, ///< Terse triples http://www.w3.org/TR/turtle
- SERD_NTRIPLES = 2, ///< Line-based triples http://www.w3.org/TR/n-triples/
- SERD_NQUADS = 3, ///< Line-based quads http://www.w3.org/TR/n-quads/
- SERD_TRIG = 4 ///< Terse quads http://www.w3.org/TR/trig/
+ SERD_TURTLE = 1, ///< Terse triples http://www.w3.org/TR/turtle
+ SERD_NTRIPLES = 2, ///< Line-based triples http://www.w3.org/TR/n-triples/
+ SERD_NQUADS = 3, ///< Line-based quads http://www.w3.org/TR/n-quads/
+ SERD_TRIG = 4 ///< Terse quads http://www.w3.org/TR/trig/
} SerdSyntax;
/// Flags indicating inline abbreviation information for a statement
typedef enum {
- SERD_EMPTY_S = 1u << 1u, ///< Empty blank node subject
- SERD_EMPTY_O = 1u << 2u, ///< Empty blank node object
- SERD_ANON_S_BEGIN = 1u << 3u, ///< Start of anonymous subject
- SERD_ANON_O_BEGIN = 1u << 4u, ///< Start of anonymous object
- SERD_ANON_CONT = 1u << 5u, ///< Continuation of anonymous node
- SERD_LIST_S_BEGIN = 1u << 6u, ///< Start of list subject
- SERD_LIST_O_BEGIN = 1u << 7u, ///< Start of list object
- SERD_LIST_CONT = 1u << 8u ///< Continuation of list
+ SERD_EMPTY_S = 1u << 1u, ///< Empty blank node subject
+ SERD_EMPTY_O = 1u << 2u, ///< Empty blank node object
+ SERD_ANON_S_BEGIN = 1u << 3u, ///< Start of anonymous subject
+ SERD_ANON_O_BEGIN = 1u << 4u, ///< Start of anonymous object
+ SERD_ANON_CONT = 1u << 5u, ///< Continuation of anonymous node
+ SERD_LIST_S_BEGIN = 1u << 6u, ///< Start of list subject
+ SERD_LIST_O_BEGIN = 1u << 7u, ///< Start of list object
+ SERD_LIST_CONT = 1u << 8u ///< Continuation of list
} SerdStatementFlag;
/// Bitwise OR of SerdStatementFlag values
@@ -139,52 +144,52 @@ typedef uint32_t SerdStatementFlags;
node types.
*/
typedef enum {
- /**
- The type of a nonexistent node.
-
- This type is useful as a sentinel, but is never emitted by the reader.
- */
- SERD_NOTHING = 0,
-
- /**
- Literal value.
-
- A literal optionally has either a language, or a datatype (not both).
- */
- SERD_LITERAL = 1,
-
- /**
- URI (absolute or relative).
-
- Value is an unquoted URI string, which is either a relative reference
- with respect to the current base URI (e.g. "foo/bar"), or an absolute
- URI (e.g. "http://example.org/foo").
- @see [RFC3986](http://tools.ietf.org/html/rfc3986)
- */
- SERD_URI = 2,
-
- /**
- CURIE, a shortened URI.
-
- Value is an unquoted CURIE string relative to the current environment,
- e.g. "rdf:type". @see [CURIE Syntax 1.0](http://www.w3.org/TR/curie)
- */
- SERD_CURIE = 3,
-
- /**
- A blank node.
-
- Value is a blank node ID without any syntactic prefix, like "id3", which
- is meaningful only within this serialisation. @see [RDF 1.1
- Turtle](http://www.w3.org/TR/turtle/#grammar-production-BLANK_NODE_LABEL)
- */
- SERD_BLANK = 4
+ /**
+ The type of a nonexistent node.
+
+ This type is useful as a sentinel, but is never emitted by the reader.
+ */
+ SERD_NOTHING = 0,
+
+ /**
+ Literal value.
+
+ A literal optionally has either a language, or a datatype (not both).
+ */
+ SERD_LITERAL = 1,
+
+ /**
+ URI (absolute or relative).
+
+ Value is an unquoted URI string, which is either a relative reference
+ with respect to the current base URI (e.g. "foo/bar"), or an absolute
+ URI (e.g. "http://example.org/foo").
+ @see [RFC3986](http://tools.ietf.org/html/rfc3986)
+ */
+ SERD_URI = 2,
+
+ /**
+ CURIE, a shortened URI.
+
+ Value is an unquoted CURIE string relative to the current environment,
+ e.g. "rdf:type". @see [CURIE Syntax 1.0](http://www.w3.org/TR/curie)
+ */
+ SERD_CURIE = 3,
+
+ /**
+ A blank node.
+
+ Value is a blank node ID without any syntactic prefix, like "id3", which
+ is meaningful only within this serialisation. @see [RDF 1.1
+ Turtle](http://www.w3.org/TR/turtle/#grammar-production-BLANK_NODE_LABEL)
+ */
+ SERD_BLANK = 4
} SerdType;
/// Flags indicating certain string properties relevant to serialisation
typedef enum {
- SERD_HAS_NEWLINE = 1u << 0u, ///< Contains line breaks ('\\n' or '\\r')
- SERD_HAS_QUOTE = 1u << 1u ///< Contains quotes ('"')
+ SERD_HAS_NEWLINE = 1u << 0u, ///< Contains line breaks ('\\n' or '\\r')
+ SERD_HAS_QUOTE = 1u << 1u ///< Contains quotes ('"')
} SerdNodeFlag;
/// Bitwise OR of SerdNodeFlag values
@@ -192,27 +197,27 @@ typedef uint32_t SerdNodeFlags;
/// A syntactic RDF node
typedef struct {
- const uint8_t* SERD_NULLABLE buf; ///< Value string
- size_t n_bytes; ///< Size in bytes (excluding null)
- size_t n_chars; ///< String length (excluding null)
- SerdNodeFlags flags; ///< Node flags (string properties)
- SerdType type; ///< Node type
+ const uint8_t* SERD_NULLABLE buf; ///< Value string
+ size_t n_bytes; ///< Size in bytes (excluding null)
+ size_t n_chars; ///< String length (excluding null)
+ SerdNodeFlags flags; ///< Node flags (string properties)
+ SerdType type; ///< Node type
} SerdNode;
/// An unterminated string fragment
typedef struct {
- const uint8_t* SERD_NULLABLE buf; ///< Start of chunk
- size_t len; ///< Length of chunk in bytes
+ const uint8_t* SERD_NULLABLE buf; ///< Start of chunk
+ size_t len; ///< Length of chunk in bytes
} SerdChunk;
/// An error description
typedef struct {
- SerdStatus status; ///< Error code
- const uint8_t* SERD_NULLABLE filename; ///< File with error
- unsigned line; ///< Line in file with error or 0
- unsigned col; ///< Column in file with error
- const char* SERD_NONNULL fmt; ///< Printf-style format string
- va_list* SERD_NONNULL args; ///< Arguments for fmt
+ SerdStatus status; ///< Error code
+ const uint8_t* SERD_NULLABLE filename; ///< File with error
+ unsigned line; ///< Line in file with error or 0
+ unsigned col; ///< Column in file with error
+ const char* SERD_NONNULL fmt; ///< Printf-style format string
+ va_list* SERD_NONNULL args; ///< Arguments for fmt
} SerdError;
/**
@@ -223,12 +228,12 @@ typedef struct {
in-place without allocating memory.
*/
typedef struct {
- SerdChunk scheme; ///< Scheme
- SerdChunk authority; ///< Authority
- SerdChunk path_base; ///< Path prefix if relative
- SerdChunk path; ///< Path suffix
- SerdChunk query; ///< Query
- SerdChunk fragment; ///< Fragment
+ SerdChunk scheme; ///< Scheme
+ SerdChunk authority; ///< Authority
+ SerdChunk path_base; ///< Path prefix if relative
+ SerdChunk path; ///< Path suffix
+ SerdChunk query; ///< Query
+ SerdChunk fragment; ///< Fragment
} SerdURI;
/**
@@ -239,11 +244,11 @@ typedef struct {
does not support abbreviation and is always ASCII.
*/
typedef enum {
- SERD_STYLE_ABBREVIATED = 1u << 0u, ///< Abbreviate triples when possible.
- SERD_STYLE_ASCII = 1u << 1u, ///< Escape all non-ASCII characters.
- SERD_STYLE_RESOLVED = 1u << 2u, ///< Resolve URIs against base URI.
- SERD_STYLE_CURIED = 1u << 3u, ///< Shorten URIs into CURIEs.
- SERD_STYLE_BULK = 1u << 4u, ///< Write output in pages.
+ SERD_STYLE_ABBREVIATED = 1u << 0u, ///< Abbreviate triples when possible.
+ SERD_STYLE_ASCII = 1u << 1u, ///< Escape all non-ASCII characters.
+ SERD_STYLE_RESOLVED = 1u << 2u, ///< Resolve URIs against base URI.
+ SERD_STYLE_CURIED = 1u << 3u, ///< Shorten URIs into CURIEs.
+ SERD_STYLE_BULK = 1u << 4u, ///< Write output in pages.
} SerdStyle;
/**
@@ -353,9 +358,8 @@ typedef size_t (*SerdSink)(const void* SERD_NONNULL buf,
@{
*/
-static const SerdURI SERD_URI_NULL = {
- {NULL, 0}, {NULL, 0}, {NULL, 0}, {NULL, 0}, {NULL, 0}, {NULL, 0}
-};
+static const SerdURI SERD_URI_NULL =
+ {{NULL, 0}, {NULL, 0}, {NULL, 0}, {NULL, 0}, {NULL, 0}, {NULL, 0}};
/**
Return the local path for `uri`, or NULL if `uri` is not a file URI.
@@ -364,7 +368,8 @@ static const SerdURI SERD_URI_NULL = {
encoding and other issues are not handled, to properly convert a file URI to
a path, use serd_file_uri_parse().
*/
-SERD_API SERD_DEPRECATED_BY("serd_file_uri_parse")
+SERD_API
+SERD_DEPRECATED_BY("serd_file_uri_parse")
const uint8_t* SERD_NULLABLE
serd_uri_to_path(const uint8_t* SERD_NONNULL uri);
@@ -431,7 +436,7 @@ serd_uri_serialise_relative(const SerdURI* SERD_NONNULL uri,
@{
*/
-static const SerdNode SERD_NODE_NULL = { NULL, 0, 0, 0, SERD_NOTHING };
+static const SerdNode SERD_NODE_NULL = {NULL, 0, 0, 0, SERD_NOTHING};
/**
Make a (shallow) node from `str`.
@@ -450,16 +455,16 @@ serd_node_from_string(SerdType type, const uint8_t* SERD_NULLABLE str);
*/
SERD_API
SerdNode
-serd_node_from_substring(SerdType type,
+serd_node_from_substring(SerdType type,
const uint8_t* SERD_NULLABLE str,
- size_t len);
+ size_t len);
/// Simple wrapper for serd_node_new_uri() to resolve a URI node
SERD_API
SerdNode
serd_node_new_uri_from_node(const SerdNode* SERD_NONNULL uri_node,
- const SerdURI* SERD_NULLABLE base,
- SerdURI* SERD_NULLABLE out);
+ const SerdURI* SERD_NULLABLE base,
+ SerdURI* SERD_NULLABLE out);
/// Simple wrapper for serd_node_new_uri() to resolve a URI string
SERD_API
@@ -621,14 +626,14 @@ typedef SerdStatus (*SerdPrefixSink)(void* SERD_NULLABLE handle,
Called for every RDF statement in the serialisation.
*/
typedef SerdStatus (*SerdStatementSink)(
- void* SERD_NULLABLE handle,
- SerdStatementFlags flags,
- const SerdNode* SERD_NULLABLE graph,
- const SerdNode* SERD_NONNULL subject,
- const SerdNode* SERD_NONNULL predicate,
- const SerdNode* SERD_NONNULL object,
- const SerdNode* SERD_NULLABLE object_datatype,
- const SerdNode* SERD_NULLABLE object_lang);
+ void* SERD_NULLABLE handle,
+ SerdStatementFlags flags,
+ const SerdNode* SERD_NULLABLE graph,
+ const SerdNode* SERD_NONNULL subject,
+ const SerdNode* SERD_NONNULL predicate,
+ const SerdNode* SERD_NONNULL object,
+ const SerdNode* SERD_NULLABLE object_datatype,
+ const SerdNode* SERD_NULLABLE object_lang);
/**
Sink (callback) for anonymous node end markers
@@ -1041,10 +1046,10 @@ serd_writer_finish(SerdWriter* SERD_NONNULL writer);
*/
#ifdef __cplusplus
-# if defined(__GNUC__)
-# pragma GCC diagnostic pop
-# endif
-} /* extern "C" */
+# if defined(__GNUC__)
+# pragma GCC diagnostic pop
+# endif
+} /* extern "C" */
#endif
-#endif /* SERD_SERD_H */
+#endif /* SERD_SERD_H */
diff --git a/src/attributes.h b/src/attributes.h
index 3ca797a4..fa60b862 100644
--- a/src/attributes.h
+++ b/src/attributes.h
@@ -18,9 +18,9 @@
#define SERD_ATTRIBUTES_H
#ifdef __GNUC__
-# define SERD_MALLOC_FUNC __attribute__((malloc))
+# define SERD_MALLOC_FUNC __attribute__((malloc))
#else
-# define SERD_MALLOC_FUNC
+# define SERD_MALLOC_FUNC
#endif
#endif // SERD_ATTRIBUTES_H
diff --git a/src/base64.c b/src/base64.c
index 763c2d2e..6f292d02 100644
--- a/src/base64.c
+++ b/src/base64.c
@@ -32,7 +32,7 @@
@see <a href="http://tools.ietf.org/html/rfc3548#section-3">RFC3548 S3</a>.
*/
static const uint8_t b64_map[] =
- "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
+ "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
/**
Base64 decoding table.
@@ -42,27 +42,28 @@ static const uint8_t b64_map[] =
A '$' is a placeholder for characters not in the base64 alphabet.
*/
static const char b64_unmap[] =
- "$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$m$$$ncdefghijkl$$$$$$"
- "$/0123456789:;<=>?@ABCDEFGH$$$$$$IJKLMNOPQRSTUVWXYZ[\\]^_`ab$$$$"
- "$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$"
- "$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$";
+ "$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$m$$$ncdefghijkl$$$$$$"
+ "$/0123456789:;<=>?@ABCDEFGH$$$$$$IJKLMNOPQRSTUVWXYZ[\\]^_`ab$$$$"
+ "$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$"
+ "$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$";
/** Encode 3 raw bytes to 4 base64 characters. */
static inline void
encode_chunk(uint8_t out[4], const uint8_t in[3], size_t n_in)
{
- out[0] = b64_map[in[0] >> 2];
- out[1] = b64_map[((in[0] & 0x03) << 4) | ((in[1] & 0xF0) >> 4)];
- out[2] = ((n_in > 1)
- ? (b64_map[((in[1] & 0x0F) << 2) | ((in[2] & 0xC0) >> 6)])
- : (uint8_t)'=');
- out[3] = ((n_in > 2) ? b64_map[in[2] & 0x3F] : (uint8_t)'=');
+ out[0] = b64_map[in[0] >> 2];
+ out[1] = b64_map[((in[0] & 0x03) << 4) | ((in[1] & 0xF0) >> 4)];
+
+ out[2] = (n_in > 1) ? (b64_map[((in[1] & 0x0F) << 2) | ((in[2] & 0xC0) >> 6)])
+ : (uint8_t)'=';
+
+ out[3] = ((n_in > 2) ? b64_map[in[2] & 0x3F] : (uint8_t)'=');
}
size_t
serd_base64_get_length(const size_t size, const bool wrap_lines)
{
- return (size + 2) / 3 * 4 + (wrap_lines * ((size - 1) / 57));
+ return (size + 2) / 3 * 4 + (wrap_lines * ((size - 1) / 57));
}
bool
@@ -71,54 +72,61 @@ serd_base64_encode(uint8_t* const str,
const size_t size,
const bool wrap_lines)
{
- bool has_newline = false;
- for (size_t i = 0, j = 0; i < size; i += 3, j += 4) {
- uint8_t in[4] = { 0, 0, 0, 0 };
- size_t n_in = MIN(3, size - i);
- memcpy(in, (const uint8_t*)buf + i, n_in);
+ bool has_newline = false;
+
+ for (size_t i = 0, j = 0; i < size; i += 3, j += 4) {
+ uint8_t in[4] = {0, 0, 0, 0};
+ size_t n_in = MIN(3, size - i);
+ memcpy(in, (const uint8_t*)buf + i, n_in);
- if (wrap_lines && i > 0 && (i % 57) == 0) {
- str[j++] = '\n';
- has_newline = true;
- }
+ if (wrap_lines && i > 0 && (i % 57) == 0) {
+ str[j++] = '\n';
+ has_newline = true;
+ }
- encode_chunk(str + j, in, n_in);
- }
+ encode_chunk(str + j, in, n_in);
+ }
- return has_newline;
+ return has_newline;
}
static inline uint8_t
unmap(const uint8_t in)
{
- return (uint8_t)(b64_unmap[in] - 47);
+ return (uint8_t)(b64_unmap[in] - 47);
}
/** Decode 4 base64 characters to 3 raw bytes. */
static inline size_t
decode_chunk(const uint8_t in[4], uint8_t out[3])
{
- out[0] = (uint8_t)(((unmap(in[0]) << 2)) | unmap(in[1]) >> 4);
- out[1] = (uint8_t)(((unmap(in[1]) << 4) & 0xF0) | unmap(in[2]) >> 2);
- out[2] = (uint8_t)(((unmap(in[2]) << 6) & 0xC0) | unmap(in[3]));
- return 1 + (in[2] != '=') + ((in[2] != '=') && (in[3] != '='));
+ out[0] = (uint8_t)(((unmap(in[0]) << 2)) | unmap(in[1]) >> 4);
+ out[1] = (uint8_t)(((unmap(in[1]) << 4) & 0xF0) | unmap(in[2]) >> 2);
+ out[2] = (uint8_t)(((unmap(in[2]) << 6) & 0xC0) | unmap(in[3]));
+ return 1 + (in[2] != '=') + ((in[2] != '=') && (in[3] != '='));
}
void*
serd_base64_decode(const uint8_t* str, size_t len, size_t* size)
{
- void* buf = malloc((len * 3) / 4 + 2);
- *size = 0;
- for (size_t i = 0, j = 0; i < len; j += 3) {
- uint8_t in[] = "====";
- size_t n_in = 0;
- for (; i < len && n_in < 4; ++n_in) {
- for (; i < len && !is_base64(str[i]); ++i) {} // Skip junk
- in[n_in] = str[i++];
- }
- if (n_in > 1) {
- *size += decode_chunk(in, (uint8_t*)buf + j);
- }
- }
- return buf;
+ void* buf = malloc((len * 3) / 4 + 2);
+
+ *size = 0;
+ for (size_t i = 0, j = 0; i < len; j += 3) {
+ uint8_t in[] = "====";
+ size_t n_in = 0;
+ for (; i < len && n_in < 4; ++n_in) {
+ for (; i < len && !is_base64(str[i]); ++i) {
+ // Skip junk
+ }
+
+ in[n_in] = str[i++];
+ }
+
+ if (n_in > 1) {
+ *size += decode_chunk(in, (uint8_t*)buf + j);
+ }
+ }
+
+ return buf;
}
diff --git a/src/base64.h b/src/base64.h
index 7fd29fef..0969b2d3 100644
--- a/src/base64.h
+++ b/src/base64.h
@@ -45,4 +45,4 @@ serd_base64_get_length(size_t size, bool wrap_lines);
bool
serd_base64_encode(uint8_t* str, const void* buf, size_t size, bool wrap_lines);
-#endif // SERD_BASE64_H
+#endif // SERD_BASE64_H
diff --git a/src/byte_sink.h b/src/byte_sink.h
index 71d525ce..b58d5d39 100644
--- a/src/byte_sink.h
+++ b/src/byte_sink.h
@@ -27,73 +27,72 @@
#include <string.h>
typedef struct SerdByteSinkImpl {
- SerdSink sink;
- void* stream;
- uint8_t* buf;
- size_t size;
- size_t block_size;
+ SerdSink sink;
+ void* stream;
+ uint8_t* buf;
+ size_t size;
+ size_t block_size;
} SerdByteSink;
static inline SerdByteSink
serd_byte_sink_new(SerdSink sink, void* stream, size_t block_size)
{
- SerdByteSink bsink;
- bsink.sink = sink;
- bsink.stream = stream;
- bsink.size = 0;
- bsink.block_size = block_size;
- bsink.buf = ((block_size > 1)
- ? (uint8_t*)serd_allocate_buffer(block_size)
- : NULL);
- return bsink;
+ SerdByteSink bsink = {sink, stream, NULL, 0, block_size};
+
+ if (block_size > 1) {
+ bsink.buf = (uint8_t*)serd_allocate_buffer(block_size);
+ }
+
+ return bsink;
}
static inline void
serd_byte_sink_flush(SerdByteSink* bsink)
{
- if (bsink->block_size > 1 && bsink->size > 0) {
- bsink->sink(bsink->buf, bsink->size, bsink->stream);
- bsink->size = 0;
- }
+ if (bsink->block_size > 1 && bsink->size > 0) {
+ bsink->sink(bsink->buf, bsink->size, bsink->stream);
+ bsink->size = 0;
+ }
}
static inline void
serd_byte_sink_free(SerdByteSink* bsink)
{
- serd_byte_sink_flush(bsink);
- serd_free_aligned(bsink->buf);
- bsink->buf = NULL;
+ serd_byte_sink_flush(bsink);
+ serd_free_aligned(bsink->buf);
+ bsink->buf = NULL;
}
static inline size_t
serd_byte_sink_write(const void* buf, size_t len, SerdByteSink* bsink)
{
- if (len == 0) {
- return 0;
- }
-
- if (bsink->block_size == 1) {
- return bsink->sink(buf, len, bsink->stream);
- }
-
- const size_t orig_len = len;
- while (len) {
- const size_t space = bsink->block_size - bsink->size;
- const size_t n = MIN(space, len);
-
- // Write as much as possible into the remaining buffer space
- memcpy(bsink->buf + bsink->size, buf, n);
- bsink->size += n;
- buf = (const uint8_t*)buf + n;
- len -= n;
-
- // Flush page if buffer is full
- if (bsink->size == bsink->block_size) {
- bsink->sink(bsink->buf, bsink->block_size, bsink->stream);
- bsink->size = 0;
- }
- }
- return orig_len;
+ if (len == 0) {
+ return 0;
+ }
+
+ if (bsink->block_size == 1) {
+ return bsink->sink(buf, len, bsink->stream);
+ }
+
+ const size_t orig_len = len;
+ while (len) {
+ const size_t space = bsink->block_size - bsink->size;
+ const size_t n = MIN(space, len);
+
+ // Write as much as possible into the remaining buffer space
+ memcpy(bsink->buf + bsink->size, buf, n);
+ bsink->size += n;
+ buf = (const uint8_t*)buf + n;
+ len -= n;
+
+ // Flush page if buffer is full
+ if (bsink->size == bsink->block_size) {
+ bsink->sink(bsink->buf, bsink->block_size, bsink->stream);
+ bsink->size = 0;
+ }
+ }
+
+ return orig_len;
}
-#endif // SERD_BYTE_SINK_H
+#endif // SERD_BYTE_SINK_H
diff --git a/src/byte_source.c b/src/byte_source.c
index a62f7a47..1b461fd0 100644
--- a/src/byte_source.c
+++ b/src/byte_source.c
@@ -27,23 +27,23 @@
SerdStatus
serd_byte_source_page(SerdByteSource* source)
{
- source->read_head = 0;
- const size_t n_read = source->read_func(
- source->file_buf, 1, source->page_size, source->stream);
-
- if (n_read == 0) {
- source->file_buf[0] = '\0';
- source->eof = true;
- return (source->error_func(source->stream)
- ? SERD_ERR_UNKNOWN : SERD_FAILURE);
- }
-
- if (n_read < source->page_size) {
- source->file_buf[n_read] = '\0';
- source->buf_size = n_read;
- }
-
- return SERD_SUCCESS;
+ source->read_head = 0;
+ const size_t n_read =
+ source->read_func(source->file_buf, 1, source->page_size, source->stream);
+
+ if (n_read == 0) {
+ source->file_buf[0] = '\0';
+ source->eof = true;
+ return (source->error_func(source->stream) ? SERD_ERR_UNKNOWN
+ : SERD_FAILURE);
+ }
+
+ if (n_read < source->page_size) {
+ source->file_buf[n_read] = '\0';
+ source->buf_size = n_read;
+ }
+
+ return SERD_SUCCESS;
}
SerdStatus
@@ -54,58 +54,59 @@ serd_byte_source_open_source(SerdByteSource* source,
const uint8_t* name,
size_t page_size)
{
- const Cursor cur = { name, 1, 1 };
-
- memset(source, '\0', sizeof(*source));
- source->stream = stream;
- source->from_stream = true;
- source->page_size = page_size;
- source->buf_size = page_size;
- source->cur = cur;
- source->error_func = error_func;
- source->read_func = read_func;
-
- if (page_size > 1) {
- source->file_buf = (uint8_t*)serd_allocate_buffer(page_size);
- source->read_buf = source->file_buf;
- memset(source->file_buf, '\0', page_size);
- } else {
- source->read_buf = &source->read_byte;
- }
-
- return SERD_SUCCESS;
+ const Cursor cur = {name, 1, 1};
+
+ memset(source, '\0', sizeof(*source));
+ source->stream = stream;
+ source->from_stream = true;
+ source->page_size = page_size;
+ source->buf_size = page_size;
+ source->cur = cur;
+ source->error_func = error_func;
+ source->read_func = read_func;
+
+ if (page_size > 1) {
+ source->file_buf = (uint8_t*)serd_allocate_buffer(page_size);
+ source->read_buf = source->file_buf;
+ memset(source->file_buf, '\0', page_size);
+ } else {
+ source->read_buf = &source->read_byte;
+ }
+
+ return SERD_SUCCESS;
}
SerdStatus
serd_byte_source_prepare(SerdByteSource* source)
{
- source->prepared = true;
+ source->prepared = true;
- if (source->from_stream) {
- return (source->page_size > 1 ? serd_byte_source_page(source)
- : serd_byte_source_advance(source));
- }
+ if (source->from_stream) {
+ return (source->page_size > 1 ? serd_byte_source_page(source)
+ : serd_byte_source_advance(source));
+ }
- return SERD_SUCCESS;
+ return SERD_SUCCESS;
}
SerdStatus
serd_byte_source_open_string(SerdByteSource* source, const uint8_t* utf8)
{
- const Cursor cur = { (const uint8_t*)"(string)", 1, 1 };
+ const Cursor cur = {(const uint8_t*)"(string)", 1, 1};
- memset(source, '\0', sizeof(*source));
- source->cur = cur;
- source->read_buf = utf8;
- return SERD_SUCCESS;
+ memset(source, '\0', sizeof(*source));
+ source->cur = cur;
+ source->read_buf = utf8;
+ return SERD_SUCCESS;
}
SerdStatus
serd_byte_source_close(SerdByteSource* source)
{
- if (source->page_size > 1) {
- serd_free_aligned(source->file_buf);
- }
- memset(source, '\0', sizeof(*source));
- return SERD_SUCCESS;
+ if (source->page_size > 1) {
+ serd_free_aligned(source->file_buf);
+ }
+
+ memset(source, '\0', sizeof(*source));
+ return SERD_SUCCESS;
}
diff --git a/src/byte_source.h b/src/byte_source.h
index a0dfa140..90958f41 100644
--- a/src/byte_source.h
+++ b/src/byte_source.h
@@ -26,31 +26,29 @@
#include <stdio.h>
typedef struct {
- const uint8_t* filename;
- unsigned line;
- unsigned col;
+ const uint8_t* filename;
+ unsigned line;
+ unsigned col;
} Cursor;
typedef struct {
- SerdSource read_func; ///< Read function (e.g. fread)
- SerdStreamErrorFunc error_func; ///< Error function (e.g. ferror)
- void* stream; ///< Stream (e.g. FILE)
- size_t page_size; ///< Number of bytes to read at a time
- size_t buf_size; ///< Number of bytes in file_buf
- Cursor cur; ///< Cursor for error reporting
- uint8_t* file_buf; ///< Buffer iff reading pages from a file
- const uint8_t* read_buf; ///< Pointer to file_buf or read_byte
- size_t read_head; ///< Offset into read_buf
- uint8_t read_byte; ///< 1-byte 'buffer' used when not paging
- bool from_stream; ///< True iff reading from `stream`
- bool prepared; ///< True iff prepared for reading
- bool eof; ///< True iff end of file reached
+ SerdSource read_func; ///< Read function (e.g. fread)
+ SerdStreamErrorFunc error_func; ///< Error function (e.g. ferror)
+ void* stream; ///< Stream (e.g. FILE)
+ size_t page_size; ///< Number of bytes to read at a time
+ size_t buf_size; ///< Number of bytes in file_buf
+ Cursor cur; ///< Cursor for error reporting
+ uint8_t* file_buf; ///< Buffer iff reading pages from a file
+ const uint8_t* read_buf; ///< Pointer to file_buf or read_byte
+ size_t read_head; ///< Offset into read_buf
+ uint8_t read_byte; ///< 1-byte 'buffer' used when not paging
+ bool from_stream; ///< True iff reading from `stream`
+ bool prepared; ///< True iff prepared for reading
+ bool eof; ///< True iff end of file reached
} SerdByteSource;
SerdStatus
-serd_byte_source_open_file(SerdByteSource* source,
- FILE* file,
- bool bulk);
+serd_byte_source_open_file(SerdByteSource* source, FILE* file, bool bulk);
SerdStatus
serd_byte_source_open_string(SerdByteSource* source, const uint8_t* utf8);
@@ -75,44 +73,48 @@ serd_byte_source_page(SerdByteSource* source);
static inline uint8_t
serd_byte_source_peek(SerdByteSource* source)
{
- assert(source->prepared);
- return source->read_buf[source->read_head];
+ assert(source->prepared);
+ return source->read_buf[source->read_head];
}
static inline SerdStatus
serd_byte_source_advance(SerdByteSource* source)
{
- SerdStatus st = SERD_SUCCESS;
-
- switch (serd_byte_source_peek(source)) {
- case '\n': ++source->cur.line; source->cur.col = 0; break;
- default: ++source->cur.col;
- }
-
- const bool was_eof = source->eof;
- if (source->from_stream) {
- source->eof = false;
- if (source->page_size > 1) {
- if (++source->read_head == source->page_size) {
- st = serd_byte_source_page(source);
- } else if (source->read_head == source->buf_size) {
- source->eof = true;
- }
- } else {
- if (!source->read_func(&source->read_byte, 1, 1, source->stream)) {
- source->eof = true;
- st = source->error_func(source->stream) ? SERD_ERR_UNKNOWN
- : SERD_FAILURE;
- }
- }
- } else if (!source->eof) {
- ++source->read_head; // Move to next character in string
- if (source->read_buf[source->read_head] == '\0') {
- source->eof = true;
- }
- }
-
- return (was_eof && source->eof) ? SERD_FAILURE : st;
+ SerdStatus st = SERD_SUCCESS;
+
+ switch (serd_byte_source_peek(source)) {
+ case '\n':
+ ++source->cur.line;
+ source->cur.col = 0;
+ break;
+ default:
+ ++source->cur.col;
+ }
+
+ const bool was_eof = source->eof;
+ if (source->from_stream) {
+ source->eof = false;
+ if (source->page_size > 1) {
+ if (++source->read_head == source->page_size) {
+ st = serd_byte_source_page(source);
+ } else if (source->read_head == source->buf_size) {
+ source->eof = true;
+ }
+ } else {
+ if (!source->read_func(&source->read_byte, 1, 1, source->stream)) {
+ source->eof = true;
+ st =
+ source->error_func(source->stream) ? SERD_ERR_UNKNOWN : SERD_FAILURE;
+ }
+ }
+ } else if (!source->eof) {
+ ++source->read_head; // Move to next character in string
+ if (source->read_buf[source->read_head] == '\0') {
+ source->eof = true;
+ }
+ }
+
+ return (was_eof && source->eof) ? SERD_FAILURE : st;
}
-#endif // SERD_BYTE_SOURCE_H
+#endif // SERD_BYTE_SOURCE_H
diff --git a/src/env.c b/src/env.c
index d11553b4..d2f7a3f5 100644
--- a/src/env.c
+++ b/src/env.c
@@ -23,141 +23,137 @@
#include <string.h>
typedef struct {
- SerdNode name;
- SerdNode uri;
+ SerdNode name;
+ SerdNode uri;
} SerdPrefix;
struct SerdEnvImpl {
- SerdPrefix* prefixes;
- size_t n_prefixes;
- SerdNode base_uri_node;
- SerdURI base_uri;
+ SerdPrefix* prefixes;
+ size_t n_prefixes;
+ SerdNode base_uri_node;
+ SerdURI base_uri;
};
SerdEnv*
serd_env_new(const SerdNode* base_uri)
{
- SerdEnv* env = (SerdEnv*)calloc(1, sizeof(struct SerdEnvImpl));
- if (env && base_uri) {
- serd_env_set_base_uri(env, base_uri);
- }
- return env;
+ SerdEnv* env = (SerdEnv*)calloc(1, sizeof(struct SerdEnvImpl));
+ if (env && base_uri) {
+ serd_env_set_base_uri(env, base_uri);
+ }
+
+ return env;
}
void
serd_env_free(SerdEnv* env)
{
- if (!env) {
- return;
- }
-
- for (size_t i = 0; i < env->n_prefixes; ++i) {
- serd_node_free(&env->prefixes[i].name);
- serd_node_free(&env->prefixes[i].uri);
- }
- free(env->prefixes);
- serd_node_free(&env->base_uri_node);
- free(env);
+ if (!env) {
+ return;
+ }
+
+ for (size_t i = 0; i < env->n_prefixes; ++i) {
+ serd_node_free(&env->prefixes[i].name);
+ serd_node_free(&env->prefixes[i].uri);
+ }
+
+ free(env->prefixes);
+ serd_node_free(&env->base_uri_node);
+ free(env);
}
const SerdNode*
-serd_env_get_base_uri(const SerdEnv* env,
- SerdURI* out)
+serd_env_get_base_uri(const SerdEnv* env, SerdURI* out)
{
- if (out) {
- *out = env->base_uri;
- }
- return &env->base_uri_node;
+ if (out) {
+ *out = env->base_uri;
+ }
+
+ return &env->base_uri_node;
}
SerdStatus
-serd_env_set_base_uri(SerdEnv* env,
- const SerdNode* uri)
+serd_env_set_base_uri(SerdEnv* env, const SerdNode* uri)
{
- if (!env || (uri && uri->type != SERD_URI)) {
- return SERD_ERR_BAD_ARG;
- }
-
- if (!uri || !uri->buf) {
- serd_node_free(&env->base_uri_node);
- env->base_uri_node = SERD_NODE_NULL;
- env->base_uri = SERD_URI_NULL;
- return SERD_SUCCESS;
- }
-
- // Resolve base URI and create a new node and URI for it
- SerdURI base_uri;
- SerdNode base_uri_node = serd_node_new_uri_from_node(
- uri, &env->base_uri, &base_uri);
-
- // Replace the current base URI
- serd_node_free(&env->base_uri_node);
- env->base_uri_node = base_uri_node;
- env->base_uri = base_uri;
-
- return SERD_SUCCESS;
+ if (!env || (uri && uri->type != SERD_URI)) {
+ return SERD_ERR_BAD_ARG;
+ }
+
+ if (!uri || !uri->buf) {
+ serd_node_free(&env->base_uri_node);
+ env->base_uri_node = SERD_NODE_NULL;
+ env->base_uri = SERD_URI_NULL;
+ return SERD_SUCCESS;
+ }
+
+ // Resolve base URI and create a new node and URI for it
+ SerdURI base_uri;
+ SerdNode base_uri_node =
+ serd_node_new_uri_from_node(uri, &env->base_uri, &base_uri);
+
+ // Replace the current base URI
+ serd_node_free(&env->base_uri_node);
+ env->base_uri_node = base_uri_node;
+ env->base_uri = base_uri;
+
+ return SERD_SUCCESS;
}
static inline SERD_PURE_FUNC SerdPrefix*
-serd_env_find(const SerdEnv* env,
- const uint8_t* name,
- size_t name_len)
+serd_env_find(const SerdEnv* env, const uint8_t* name, size_t name_len)
{
- for (size_t i = 0; i < env->n_prefixes; ++i) {
- const SerdNode* const prefix_name = &env->prefixes[i].name;
- if (prefix_name->n_bytes == name_len) {
- if (!memcmp(prefix_name->buf, name, name_len)) {
- return &env->prefixes[i];
- }
- }
- }
- return NULL;
+ for (size_t i = 0; i < env->n_prefixes; ++i) {
+ const SerdNode* const prefix_name = &env->prefixes[i].name;
+ if (prefix_name->n_bytes == name_len) {
+ if (!memcmp(prefix_name->buf, name, name_len)) {
+ return &env->prefixes[i];
+ }
+ }
+ }
+
+ return NULL;
}
static void
-serd_env_add(SerdEnv* env,
- const SerdNode* name,
- const SerdNode* uri)
+serd_env_add(SerdEnv* env, const SerdNode* name, const SerdNode* uri)
{
- SerdPrefix* const prefix = serd_env_find(env, name->buf, name->n_bytes);
- if (prefix) {
- if (!serd_node_equals(&prefix->uri, uri)) {
- SerdNode old_prefix_uri = prefix->uri;
- prefix->uri = serd_node_copy(uri);
- serd_node_free(&old_prefix_uri);
- }
- } else {
- env->prefixes = (SerdPrefix*)realloc(
- env->prefixes, (++env->n_prefixes) * sizeof(SerdPrefix));
- env->prefixes[env->n_prefixes - 1].name = serd_node_copy(name);
- env->prefixes[env->n_prefixes - 1].uri = serd_node_copy(uri);
- }
+ SerdPrefix* const prefix = serd_env_find(env, name->buf, name->n_bytes);
+ if (prefix) {
+ if (!serd_node_equals(&prefix->uri, uri)) {
+ SerdNode old_prefix_uri = prefix->uri;
+ prefix->uri = serd_node_copy(uri);
+ serd_node_free(&old_prefix_uri);
+ }
+ } else {
+ env->prefixes = (SerdPrefix*)realloc(
+ env->prefixes, (++env->n_prefixes) * sizeof(SerdPrefix));
+ env->prefixes[env->n_prefixes - 1].name = serd_node_copy(name);
+ env->prefixes[env->n_prefixes - 1].uri = serd_node_copy(uri);
+ }
}
SerdStatus
-serd_env_set_prefix(SerdEnv* env,
- const SerdNode* name,
- const SerdNode* uri)
+serd_env_set_prefix(SerdEnv* env, const SerdNode* name, const SerdNode* uri)
{
- if (!name->buf || uri->type != SERD_URI) {
- return SERD_ERR_BAD_ARG;
- }
-
- if (serd_uri_string_has_scheme(uri->buf)) {
- // Set prefix to absolute URI
- serd_env_add(env, name, uri);
- } else {
- // Resolve relative URI and create a new node and URI for it
- SerdURI abs_uri;
- SerdNode abs_uri_node = serd_node_new_uri_from_node(
- uri, &env->base_uri, &abs_uri);
-
- // Set prefix to resolved (absolute) URI
- serd_env_add(env, name, &abs_uri_node);
- serd_node_free(&abs_uri_node);
- }
-
- return SERD_SUCCESS;
+ if (!name->buf || uri->type != SERD_URI) {
+ return SERD_ERR_BAD_ARG;
+ }
+
+ if (serd_uri_string_has_scheme(uri->buf)) {
+ // Set prefix to absolute URI
+ serd_env_add(env, name, uri);
+ } else {
+ // Resolve relative URI and create a new node and URI for it
+ SerdURI abs_uri;
+ SerdNode abs_uri_node =
+ serd_node_new_uri_from_node(uri, &env->base_uri, &abs_uri);
+
+ // Set prefix to resolved (absolute) URI
+ serd_env_add(env, name, &abs_uri_node);
+ serd_node_free(&abs_uri_node);
+ }
+
+ return SERD_SUCCESS;
}
SerdStatus
@@ -165,10 +161,10 @@ serd_env_set_prefix_from_strings(SerdEnv* env,
const uint8_t* name,
const uint8_t* uri)
{
- const SerdNode name_node = serd_node_from_string(SERD_LITERAL, name);
- const SerdNode uri_node = serd_node_from_string(SERD_URI, uri);
+ const SerdNode name_node = serd_node_from_string(SERD_LITERAL, name);
+ const SerdNode uri_node = serd_node_from_string(SERD_URI, uri);
- return serd_env_set_prefix(env, &name_node, &uri_node);
+ return serd_env_set_prefix(env, &name_node, &uri_node);
}
bool
@@ -177,20 +173,20 @@ serd_env_qualify(const SerdEnv* env,
SerdNode* prefix,
SerdChunk* suffix)
{
- for (size_t i = 0; i < env->n_prefixes; ++i) {
- const SerdNode* const prefix_uri = &env->prefixes[i].uri;
- if (uri->n_bytes >= prefix_uri->n_bytes) {
- if (!strncmp((const char*)uri->buf,
- (const char*)prefix_uri->buf,
- prefix_uri->n_bytes)) {
- *prefix = env->prefixes[i].name;
- suffix->buf = uri->buf + prefix_uri->n_bytes;
- suffix->len = uri->n_bytes - prefix_uri->n_bytes;
- return true;
- }
- }
- }
- return false;
+ for (size_t i = 0; i < env->n_prefixes; ++i) {
+ const SerdNode* const prefix_uri = &env->prefixes[i].uri;
+ if (uri->n_bytes >= prefix_uri->n_bytes) {
+ if (!strncmp((const char*)uri->buf,
+ (const char*)prefix_uri->buf,
+ prefix_uri->n_bytes)) {
+ *prefix = env->prefixes[i].name;
+ suffix->buf = uri->buf + prefix_uri->n_bytes;
+ suffix->len = uri->n_bytes - prefix_uri->n_bytes;
+ return true;
+ }
+ }
+ }
+ return false;
}
SerdStatus
@@ -199,61 +195,62 @@ serd_env_expand(const SerdEnv* env,
SerdChunk* uri_prefix,
SerdChunk* uri_suffix)
{
- const uint8_t* const colon = (const uint8_t*)memchr(
- curie->buf, ':', curie->n_bytes + 1);
- if (curie->type != SERD_CURIE || !colon) {
- return SERD_ERR_BAD_ARG;
- }
-
- const size_t name_len = (size_t)(colon - curie->buf);
- const SerdPrefix* const prefix = serd_env_find(env, curie->buf, name_len);
- if (prefix) {
- uri_prefix->buf = prefix->uri.buf;
- uri_prefix->len = prefix->uri.n_bytes;
- uri_suffix->buf = colon + 1;
- uri_suffix->len = curie->n_bytes - name_len - 1;
- return SERD_SUCCESS;
- }
- return SERD_ERR_BAD_CURIE;
+ const uint8_t* const colon =
+ (const uint8_t*)memchr(curie->buf, ':', curie->n_bytes + 1);
+ if (curie->type != SERD_CURIE || !colon) {
+ return SERD_ERR_BAD_ARG;
+ }
+
+ const size_t name_len = (size_t)(colon - curie->buf);
+ const SerdPrefix* const prefix = serd_env_find(env, curie->buf, name_len);
+ if (prefix) {
+ uri_prefix->buf = prefix->uri.buf;
+ uri_prefix->len = prefix->uri.n_bytes;
+ uri_suffix->buf = colon + 1;
+ uri_suffix->len = curie->n_bytes - name_len - 1;
+ return SERD_SUCCESS;
+ }
+ return SERD_ERR_BAD_CURIE;
}
SerdNode
-serd_env_expand_node(const SerdEnv* env,
- const SerdNode* node)
+serd_env_expand_node(const SerdEnv* env, const SerdNode* node)
{
- switch (node->type) {
- case SERD_NOTHING:
- case SERD_LITERAL:
- break;
- case SERD_URI: {
- SerdURI ignored;
- return serd_node_new_uri_from_node(node, &env->base_uri, &ignored);
- }
- case SERD_CURIE: {
- SerdChunk prefix;
- SerdChunk suffix;
- if (serd_env_expand(env, node, &prefix, &suffix)) {
- return SERD_NODE_NULL;
- }
- const size_t len = prefix.len + suffix.len;
- uint8_t* buf = (uint8_t*)malloc(len + 1);
- SerdNode ret = { buf, len, 0, 0, SERD_URI };
- snprintf((char*)buf, len + 1, "%s%s", prefix.buf, suffix.buf);
- ret.n_chars = serd_strlen(buf, NULL, NULL);
- return ret;
- }
- case SERD_BLANK:
- break;
- }
- return SERD_NODE_NULL;
+ switch (node->type) {
+ case SERD_NOTHING:
+ case SERD_LITERAL:
+ break;
+
+ case SERD_URI: {
+ SerdURI ignored;
+ return serd_node_new_uri_from_node(node, &env->base_uri, &ignored);
+ }
+
+ case SERD_CURIE: {
+ SerdChunk prefix;
+ SerdChunk suffix;
+ if (serd_env_expand(env, node, &prefix, &suffix)) {
+ return SERD_NODE_NULL;
+ }
+ const size_t len = prefix.len + suffix.len;
+ uint8_t* buf = (uint8_t*)malloc(len + 1);
+ SerdNode ret = {buf, len, 0, 0, SERD_URI};
+ snprintf((char*)buf, len + 1, "%s%s", prefix.buf, suffix.buf);
+ ret.n_chars = serd_strlen(buf, NULL, NULL);
+ return ret;
+ }
+
+ case SERD_BLANK:
+ break;
+ }
+
+ return SERD_NODE_NULL;
}
void
-serd_env_foreach(const SerdEnv* env,
- SerdPrefixSink func,
- void* handle)
+serd_env_foreach(const SerdEnv* env, SerdPrefixSink func, void* handle)
{
- for (size_t i = 0; i < env->n_prefixes; ++i) {
- func(handle, &env->prefixes[i].name, &env->prefixes[i].uri);
- }
+ for (size_t i = 0; i < env->n_prefixes; ++i) {
+ func(handle, &env->prefixes[i].name, &env->prefixes[i].uri);
+ }
}
diff --git a/src/n3.c b/src/n3.c
index c12b1ff0..e5fa4f94 100644
--- a/src/n3.c
+++ b/src/n3.c
@@ -30,12 +30,17 @@
#include <stdlib.h>
#include <string.h>
-#define TRY(st, exp) do { if (((st) = (exp))) { return (st); } } while (0)
+#define TRY(st, exp) \
+ do { \
+ if (((st) = (exp))) { \
+ return (st); \
+ } \
+ } while (0)
static inline bool
fancy_syntax(const SerdReader* reader)
{
- return reader->syntax == SERD_TURTLE || reader->syntax == SERD_TRIG;
+ return reader->syntax == SERD_TURTLE || reader->syntax == SERD_TRIG;
}
static SerdStatus
@@ -47,192 +52,197 @@ read_predicateObjectList(SerdReader* reader, ReadContext ctx, bool* ate_dot);
static inline uint8_t
read_HEX(SerdReader* reader)
{
- const int c = peek_byte(reader);
- if (is_xdigit(c)) {
- return (uint8_t)eat_byte_safe(reader, c);
- }
+ const int c = peek_byte(reader);
+ if (is_xdigit(c)) {
+ return (uint8_t)eat_byte_safe(reader, c);
+ }
- r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid hexadecimal digit `%c'\n", c);
- return 0;
+ r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid hexadecimal digit `%c'\n", c);
+ return 0;
}
// Read UCHAR escape, initial \ is already eaten by caller
static inline SerdStatus
read_UCHAR(SerdReader* reader, Ref dest, uint32_t* char_code)
{
- const int b = peek_byte(reader);
- unsigned length = 0;
- switch (b) {
- case 'U':
- length = 8;
- break;
- case 'u':
- length = 4;
- break;
- default:
- return SERD_ERR_BAD_SYNTAX;
- }
- eat_byte_safe(reader, b);
-
- uint8_t buf[9] = { 0, 0, 0, 0, 0, 0, 0, 0, 0 };
- for (unsigned i = 0; i < length; ++i) {
- if (!(buf[i] = read_HEX(reader))) {
- return SERD_ERR_BAD_SYNTAX;
- }
- }
-
- char* endptr = NULL;
- const uint32_t code = (uint32_t)strtoul((const char*)buf, &endptr, 16);
- assert(endptr == (char*)buf + length);
-
- unsigned size = 0;
- if (code < 0x00000080) {
- size = 1;
- } else if (code < 0x00000800) {
- size = 2;
- } else if (code < 0x00010000) {
- size = 3;
- } else if (code < 0x00110000) {
- size = 4;
- } else {
- r_err(reader, SERD_ERR_BAD_SYNTAX,
- "unicode character 0x%X out of range\n", code);
- push_bytes(reader, dest, replacement_char, 3);
- *char_code = 0xFFFD;
- return SERD_SUCCESS;
- }
-
- // Build output in buf
- // (Note # of bytes = # of leading 1 bits in first byte)
- uint32_t c = code;
- switch (size) {
- case 4:
- buf[3] = (uint8_t)(0x80u | (c & 0x3Fu));
- c >>= 6;
- c |= (16 << 12); // set bit 4
- // fallthru
- case 3:
- buf[2] = (uint8_t)(0x80u | (c & 0x3Fu));
- c >>= 6;
- c |= (32 << 6); // set bit 5
- // fallthru
- case 2:
- buf[1] = (uint8_t)(0x80u | (c & 0x3Fu));
- c >>= 6;
- c |= 0xC0; // set bits 6 and 7
- // fallthru
- case 1:
- buf[0] = (uint8_t)c;
- // fallthru
- default:
- break;
- }
-
- push_bytes(reader, dest, buf, size);
- *char_code = code;
- return SERD_SUCCESS;
+ const int b = peek_byte(reader);
+ unsigned length = 0;
+ switch (b) {
+ case 'U':
+ length = 8;
+ break;
+ case 'u':
+ length = 4;
+ break;
+ default:
+ return SERD_ERR_BAD_SYNTAX;
+ }
+
+ eat_byte_safe(reader, b);
+
+ uint8_t buf[9] = {0, 0, 0, 0, 0, 0, 0, 0, 0};
+ for (unsigned i = 0; i < length; ++i) {
+ if (!(buf[i] = read_HEX(reader))) {
+ return SERD_ERR_BAD_SYNTAX;
+ }
+ }
+
+ char* endptr = NULL;
+ const uint32_t code = (uint32_t)strtoul((const char*)buf, &endptr, 16);
+ assert(endptr == (char*)buf + length);
+
+ unsigned size = 0;
+ if (code < 0x00000080) {
+ size = 1;
+ } else if (code < 0x00000800) {
+ size = 2;
+ } else if (code < 0x00010000) {
+ size = 3;
+ } else if (code < 0x00110000) {
+ size = 4;
+ } else {
+ r_err(reader,
+ SERD_ERR_BAD_SYNTAX,
+ "unicode character 0x%X out of range\n",
+ code);
+ push_bytes(reader, dest, replacement_char, 3);
+ *char_code = 0xFFFD;
+ return SERD_SUCCESS;
+ }
+
+ // Build output in buf
+ // (Note # of bytes = # of leading 1 bits in first byte)
+ uint32_t c = code;
+ switch (size) {
+ case 4:
+ buf[3] = (uint8_t)(0x80u | (c & 0x3Fu));
+ c >>= 6;
+ c |= (16 << 12); // set bit 4
+ /* fallthru */
+ case 3:
+ buf[2] = (uint8_t)(0x80u | (c & 0x3Fu));
+ c >>= 6;
+ c |= (32 << 6); // set bit 5
+ /* fallthru */
+ case 2:
+ buf[1] = (uint8_t)(0x80u | (c & 0x3Fu));
+ c >>= 6;
+ c |= 0xC0; // set bits 6 and 7
+ /* fallthru */
+ case 1:
+ buf[0] = (uint8_t)c;
+ /* fallthru */
+ default:
+ break;
+ }
+
+ push_bytes(reader, dest, buf, size);
+ *char_code = code;
+ return SERD_SUCCESS;
}
// Read ECHAR escape, initial \ is already eaten by caller
static inline SerdStatus
read_ECHAR(SerdReader* reader, Ref dest, SerdNodeFlags* flags)
{
- const int c = peek_byte(reader);
- switch (c) {
- case 't':
- eat_byte_safe(reader, 't');
- push_byte(reader, dest, '\t');
- return SERD_SUCCESS;
- case 'b':
- eat_byte_safe(reader, 'b');
- push_byte(reader, dest, '\b');
- return SERD_SUCCESS;
- case 'n':
- *flags |= SERD_HAS_NEWLINE;
- eat_byte_safe(reader, 'n');
- push_byte(reader, dest, '\n');
- return SERD_SUCCESS;
- case 'r':
- *flags |= SERD_HAS_NEWLINE;
- eat_byte_safe(reader, 'r');
- push_byte(reader, dest, '\r');
- return SERD_SUCCESS;
- case 'f':
- eat_byte_safe(reader, 'f');
- push_byte(reader, dest, '\f');
- return SERD_SUCCESS;
- case '\\': case '"': case '\'':
- push_byte(reader, dest, eat_byte_safe(reader, c));
- return SERD_SUCCESS;
- default:
- return SERD_ERR_BAD_SYNTAX;
- }
+ const int c = peek_byte(reader);
+ switch (c) {
+ case 't':
+ eat_byte_safe(reader, 't');
+ push_byte(reader, dest, '\t');
+ return SERD_SUCCESS;
+ case 'b':
+ eat_byte_safe(reader, 'b');
+ push_byte(reader, dest, '\b');
+ return SERD_SUCCESS;
+ case 'n':
+ *flags |= SERD_HAS_NEWLINE;
+ eat_byte_safe(reader, 'n');
+ push_byte(reader, dest, '\n');
+ return SERD_SUCCESS;
+ case 'r':
+ *flags |= SERD_HAS_NEWLINE;
+ eat_byte_safe(reader, 'r');
+ push_byte(reader, dest, '\r');
+ return SERD_SUCCESS;
+ case 'f':
+ eat_byte_safe(reader, 'f');
+ push_byte(reader, dest, '\f');
+ return SERD_SUCCESS;
+ case '\\':
+ case '"':
+ case '\'':
+ push_byte(reader, dest, eat_byte_safe(reader, c));
+ return SERD_SUCCESS;
+ default:
+ return SERD_ERR_BAD_SYNTAX;
+ }
}
static inline SerdStatus
bad_char(SerdReader* reader, const char* fmt, uint8_t c)
{
- // Skip bytes until the next start byte
- for (int b = peek_byte(reader); b != EOF && ((uint8_t)b & 0x80);) {
- eat_byte_safe(reader, b);
- b = peek_byte(reader);
- }
-
- r_err(reader, SERD_ERR_BAD_SYNTAX, fmt, c);
- return reader->strict ? SERD_ERR_BAD_SYNTAX : SERD_FAILURE;
+ // Skip bytes until the next start byte
+ for (int b = peek_byte(reader); b != EOF && ((uint8_t)b & 0x80);) {
+ eat_byte_safe(reader, b);
+ b = peek_byte(reader);
+ }
+
+ r_err(reader, SERD_ERR_BAD_SYNTAX, fmt, c);
+ return reader->strict ? SERD_ERR_BAD_SYNTAX : SERD_FAILURE;
}
static SerdStatus
read_utf8_bytes(SerdReader* reader, uint8_t bytes[4], uint32_t* size, uint8_t c)
{
- *size = utf8_num_bytes(c);
- if (*size <= 1 || *size > 4) {
- return bad_char(reader, "invalid UTF-8 start 0x%X\n", c);
- }
-
- bytes[0] = c;
- for (unsigned i = 1; i < *size; ++i) {
- const int b = peek_byte(reader);
- if (b == EOF || ((uint8_t)b & 0x80) == 0) {
- return bad_char(reader, "invalid UTF-8 continuation 0x%X\n",
- (uint8_t)b);
- }
-
- eat_byte_safe(reader, b);
- bytes[i] = (uint8_t)b;
- }
-
- return SERD_SUCCESS;
+ *size = utf8_num_bytes(c);
+ if (*size <= 1 || *size > 4) {
+ return bad_char(reader, "invalid UTF-8 start 0x%X\n", c);
+ }
+
+ bytes[0] = c;
+ for (unsigned i = 1; i < *size; ++i) {
+ const int b = peek_byte(reader);
+ if (b == EOF || ((uint8_t)b & 0x80) == 0) {
+ return bad_char(reader, "invalid UTF-8 continuation 0x%X\n", (uint8_t)b);
+ }
+
+ eat_byte_safe(reader, b);
+ bytes[i] = (uint8_t)b;
+ }
+
+ return SERD_SUCCESS;
}
static SerdStatus
read_utf8_character(SerdReader* reader, Ref dest, uint8_t c)
{
- uint32_t size = 0;
- uint8_t bytes[4] = {0, 0, 0, 0};
- SerdStatus st = read_utf8_bytes(reader, bytes, &size, c);
- if (st) {
- push_bytes(reader, dest, replacement_char, 3);
- } else {
- push_bytes(reader, dest, bytes, size);
- }
- return st;
+ uint32_t size = 0;
+ uint8_t bytes[4] = {0, 0, 0, 0};
+ SerdStatus st = read_utf8_bytes(reader, bytes, &size, c);
+ if (st) {
+ push_bytes(reader, dest, replacement_char, 3);
+ } else {
+ push_bytes(reader, dest, bytes, size);
+ }
+
+ return st;
}
static SerdStatus
read_utf8_code(SerdReader* reader, Ref dest, uint32_t* code, uint8_t c)
{
- uint32_t size = 0;
- uint8_t bytes[4] = {0, 0, 0, 0};
- SerdStatus st = read_utf8_bytes(reader, bytes, &size, c);
- if (st) {
- push_bytes(reader, dest, replacement_char, 3);
- return st;
- }
-
- push_bytes(reader, dest, bytes, size);
- *code = parse_counted_utf8_char(bytes, size);
- return st;
+ uint32_t size = 0;
+ uint8_t bytes[4] = {0, 0, 0, 0};
+ SerdStatus st = read_utf8_bytes(reader, bytes, &size, c);
+ if (st) {
+ push_bytes(reader, dest, replacement_char, 3);
+ return st;
+ }
+
+ push_bytes(reader, dest, bytes, size);
+ *code = parse_counted_utf8_char(bytes, size);
+ return st;
}
// Read one character (possibly multi-byte)
@@ -240,72 +250,81 @@ read_utf8_code(SerdReader* reader, Ref dest, uint32_t* code, uint8_t c)
static inline SerdStatus
read_character(SerdReader* reader, Ref dest, SerdNodeFlags* flags, uint8_t c)
{
- if (!(c & 0x80)) {
- switch (c) {
- case 0xA: case 0xD:
- *flags |= SERD_HAS_NEWLINE;
- break;
- case '"': case '\'':
- *flags |= SERD_HAS_QUOTE;
- break;
- default:
- break;
- }
- return push_byte(reader, dest, c);
- }
- return read_utf8_character(reader, dest, c);
+ if (!(c & 0x80)) {
+ switch (c) {
+ case 0xA:
+ case 0xD:
+ *flags |= SERD_HAS_NEWLINE;
+ break;
+ case '"':
+ case '\'':
+ *flags |= SERD_HAS_QUOTE;
+ break;
+ default:
+ break;
+ }
+ return push_byte(reader, dest, c);
+ }
+
+ return read_utf8_character(reader, dest, c);
}
// [10] comment ::= '#' ( [^#xA #xD] )*
static void
read_comment(SerdReader* reader)
{
- eat_byte_safe(reader, '#');
- int c = 0;
- while (((c = peek_byte(reader)) != 0xA) && c != 0xD && c != EOF && c) {
- eat_byte_safe(reader, c);
- }
+ eat_byte_safe(reader, '#');
+ int c = 0;
+ while (((c = peek_byte(reader)) != 0xA) && c != 0xD && c != EOF && c) {
+ eat_byte_safe(reader, c);
+ }
}
// [24] ws ::= #x9 | #xA | #xD | #x20 | comment
static inline bool
read_ws(SerdReader* reader)
{
- const int c = peek_byte(reader);
- switch (c) {
- case 0x9: case 0xA: case 0xD: case 0x20:
- eat_byte_safe(reader, c);
- return true;
- case '#':
- read_comment(reader);
- return true;
- default:
- return false;
- }
+ const int c = peek_byte(reader);
+ switch (c) {
+ case 0x9:
+ case 0xA:
+ case 0xD:
+ case 0x20:
+ eat_byte_safe(reader, c);
+ return true;
+ case '#':
+ read_comment(reader);
+ return true;
+ default:
+ return false;
+ }
}
static inline bool
read_ws_star(SerdReader* reader)
{
- while (read_ws(reader)) {}
- return true;
+ while (read_ws(reader)) {
+ }
+
+ return true;
}
static inline bool
peek_delim(SerdReader* reader, const char delim)
{
- read_ws_star(reader);
- return peek_byte(reader) == delim;
+ read_ws_star(reader);
+ return peek_byte(reader) == delim;
}
static inline bool
eat_delim(SerdReader* reader, const char delim)
{
- if (peek_delim(reader, delim)) {
- eat_byte_safe(reader, delim);
- return read_ws_star(reader);
- }
- return false;
+ if (peek_delim(reader, delim)) {
+ eat_byte_safe(reader, delim);
+ return read_ws_star(reader);
+ }
+
+ return false;
}
// STRING_LITERAL_LONG_QUOTE and STRING_LITERAL_LONG_SINGLE_QUOTE
@@ -316,39 +335,37 @@ read_STRING_LITERAL_LONG(SerdReader* reader,
SerdNodeFlags* flags,
uint8_t q)
{
- SerdStatus st = SERD_SUCCESS;
-
- while (!(st && reader->strict)) {
- const int c = peek_byte(reader);
- if (c == '\\') {
- eat_byte_safe(reader, c);
- uint32_t code = 0;
- if ((st = read_ECHAR(reader, ref, flags)) &&
- (st = read_UCHAR(reader, ref, &code))) {
- return r_err(reader, st,
- "invalid escape `\\%c'\n", peek_byte(reader));
- }
- } else if (c == q) {
- eat_byte_safe(reader, q);
- const int q2 = eat_byte_safe(reader, peek_byte(reader));
- const int q3 = peek_byte(reader);
- if (q2 == q && q3 == q) { // End of string
- eat_byte_safe(reader, q3);
- break;
- }
- *flags |= SERD_HAS_QUOTE;
- push_byte(reader, ref, c);
- st = read_character(reader, ref, flags, (uint8_t)q2);
- } else if (c == EOF) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "end of file in long string\n");
- } else {
- st = read_character(
- reader, ref, flags, (uint8_t)eat_byte_safe(reader, c));
- }
- }
-
- return (st && reader->strict) ? st : SERD_SUCCESS;
+ SerdStatus st = SERD_SUCCESS;
+
+ while (!(st && reader->strict)) {
+ const int c = peek_byte(reader);
+ if (c == '\\') {
+ eat_byte_safe(reader, c);
+ uint32_t code = 0;
+ if ((st = read_ECHAR(reader, ref, flags)) &&
+ (st = read_UCHAR(reader, ref, &code))) {
+ return r_err(reader, st, "invalid escape `\\%c'\n", peek_byte(reader));
+ }
+ } else if (c == q) {
+ eat_byte_safe(reader, q);
+ const int q2 = eat_byte_safe(reader, peek_byte(reader));
+ const int q3 = peek_byte(reader);
+ if (q2 == q && q3 == q) { // End of string
+ eat_byte_safe(reader, q3);
+ break;
+ }
+ *flags |= SERD_HAS_QUOTE;
+ push_byte(reader, ref, c);
+ st = read_character(reader, ref, flags, (uint8_t)q2);
+ } else if (c == EOF) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "end of file in long string\n");
+ } else {
+ st =
+ read_character(reader, ref, flags, (uint8_t)eat_byte_safe(reader, c));
+ }
+ }
+
+ return (st && reader->strict) ? st : SERD_SUCCESS;
}
// STRING_LITERAL_QUOTE and STRING_LITERAL_SINGLE_QUOTE
@@ -359,689 +376,722 @@ read_STRING_LITERAL(SerdReader* reader,
SerdNodeFlags* flags,
uint8_t q)
{
- SerdStatus st = SERD_SUCCESS;
-
- while (!(st && reader->strict)) {
- const int c = peek_byte(reader);
- uint32_t code = 0;
- switch (c) {
- case EOF:
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "end of file in short string\n");
- case '\n': case '\r':
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "line end in short string\n");
- case '\\':
- eat_byte_safe(reader, c);
- if ((st = read_ECHAR(reader, ref, flags)) &&
- (st = read_UCHAR(reader, ref, &code))) {
- return r_err(reader, st,
- "invalid escape `\\%c'\n", peek_byte(reader));
- }
- break;
- default:
- if (c == q) {
- eat_byte_check(reader, q);
- return SERD_SUCCESS;
- } else {
- st = read_character(
- reader, ref, flags, (uint8_t)eat_byte_safe(reader, c));
- }
- }
- }
-
- return st ? st
- : eat_byte_check(reader, q) ? SERD_SUCCESS : SERD_ERR_BAD_SYNTAX;
+ SerdStatus st = SERD_SUCCESS;
+
+ while (!(st && reader->strict)) {
+ const int c = peek_byte(reader);
+ uint32_t code = 0;
+ switch (c) {
+ case EOF:
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "end of file in short string\n");
+ case '\n':
+ case '\r':
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "line end in short string\n");
+ case '\\':
+ eat_byte_safe(reader, c);
+ if ((st = read_ECHAR(reader, ref, flags)) &&
+ (st = read_UCHAR(reader, ref, &code))) {
+ return r_err(reader, st, "invalid escape `\\%c'\n", peek_byte(reader));
+ }
+ break;
+ default:
+ if (c == q) {
+ eat_byte_check(reader, q);
+ return SERD_SUCCESS;
+ } else {
+ st =
+ read_character(reader, ref, flags, (uint8_t)eat_byte_safe(reader, c));
+ }
+ }
+ }
+
+ return st ? st
+ : (eat_byte_check(reader, q) ? SERD_SUCCESS : SERD_ERR_BAD_SYNTAX);
}
static SerdStatus
read_String(SerdReader* reader, Ref node, SerdNodeFlags* flags)
{
- const int q1 = peek_byte(reader);
- eat_byte_safe(reader, q1);
-
- const int q2 = peek_byte(reader);
- if (q2 == EOF) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "unexpected end of file\n");
- }
-
- if (q2 != q1) { // Short string (not triple quoted)
- return read_STRING_LITERAL(reader, node, flags, (uint8_t)q1);
- }
-
- eat_byte_safe(reader, q2);
- const int q3 = peek_byte(reader);
- if (q3 == EOF) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "unexpected end of file\n");
- }
-
- if (q3 != q1) { // Empty short string ("" or '')
- return SERD_SUCCESS;
- }
-
- if (!fancy_syntax(reader)) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "syntax does not support long literals\n");
- }
-
- eat_byte_safe(reader, q3);
- return read_STRING_LITERAL_LONG(reader, node, flags, (uint8_t)q1);
+ const int q1 = peek_byte(reader);
+ eat_byte_safe(reader, q1);
+
+ const int q2 = peek_byte(reader);
+ if (q2 == EOF) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "unexpected end of file\n");
+ }
+
+ if (q2 != q1) { // Short string (not triple quoted)
+ return read_STRING_LITERAL(reader, node, flags, (uint8_t)q1);
+ }
+
+ eat_byte_safe(reader, q2);
+ const int q3 = peek_byte(reader);
+ if (q3 == EOF) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "unexpected end of file\n");
+ }
+
+ if (q3 != q1) { // Empty short string ("" or '')
+ return SERD_SUCCESS;
+ }
+
+ if (!fancy_syntax(reader)) {
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "syntax does not support long literals\n");
+ }
+
+ eat_byte_safe(reader, q3);
+ return read_STRING_LITERAL_LONG(reader, node, flags, (uint8_t)q1);
}
static inline bool
is_PN_CHARS_BASE(const uint32_t c)
{
- return ((c >= 0x00C0 && c <= 0x00D6) || (c >= 0x00D8 && c <= 0x00F6) ||
- (c >= 0x00F8 && c <= 0x02FF) || (c >= 0x0370 && c <= 0x037D) ||
- (c >= 0x037F && c <= 0x1FFF) || (c >= 0x200C && c <= 0x200D) ||
- (c >= 0x2070 && c <= 0x218F) || (c >= 0x2C00 && c <= 0x2FEF) ||
- (c >= 0x3001 && c <= 0xD7FF) || (c >= 0xF900 && c <= 0xFDCF) ||
- (c >= 0xFDF0 && c <= 0xFFFD) || (c >= 0x10000 && c <= 0xEFFFF));
+ return ((c >= 0x00C0 && c <= 0x00D6) || (c >= 0x00D8 && c <= 0x00F6) ||
+ (c >= 0x00F8 && c <= 0x02FF) || (c >= 0x0370 && c <= 0x037D) ||
+ (c >= 0x037F && c <= 0x1FFF) || (c >= 0x200C && c <= 0x200D) ||
+ (c >= 0x2070 && c <= 0x218F) || (c >= 0x2C00 && c <= 0x2FEF) ||
+ (c >= 0x3001 && c <= 0xD7FF) || (c >= 0xF900 && c <= 0xFDCF) ||
+ (c >= 0xFDF0 && c <= 0xFFFD) || (c >= 0x10000 && c <= 0xEFFFF));
}
static SerdStatus
read_PN_CHARS_BASE(SerdReader* reader, Ref dest)
{
- uint32_t code = 0;
- const int c = peek_byte(reader);
- SerdStatus st = SERD_SUCCESS;
- if (is_alpha(c)) {
- push_byte(reader, dest, eat_byte_safe(reader, c));
- } else if (c == EOF || !(c & 0x80)) {
- return SERD_FAILURE;
- } else if ((st = read_utf8_code(reader, dest, &code,
- (uint8_t)eat_byte_safe(reader, c)))) {
- return st;
- } else if (!is_PN_CHARS_BASE(code)) {
- r_err(reader, SERD_ERR_BAD_SYNTAX,
- "invalid character U+%04X in name\n", code);
- if (reader->strict) {
- return SERD_ERR_BAD_SYNTAX;
- }
- }
- return st;
+ uint32_t code = 0;
+ const int c = peek_byte(reader);
+ SerdStatus st = SERD_SUCCESS;
+ if (is_alpha(c)) {
+ push_byte(reader, dest, eat_byte_safe(reader, c));
+ } else if (c == EOF || !(c & 0x80)) {
+ return SERD_FAILURE;
+ } else if ((st = read_utf8_code(
+ reader, dest, &code, (uint8_t)eat_byte_safe(reader, c)))) {
+ return st;
+ } else if (!is_PN_CHARS_BASE(code)) {
+ r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "invalid character U+%04X in name\n", code);
+ if (reader->strict) {
+ return SERD_ERR_BAD_SYNTAX;
+ }
+ }
+ return st;
}
static inline bool
is_PN_CHARS(const uint32_t c)
{
- return (is_PN_CHARS_BASE(c) || c == 0xB7 ||
- (c >= 0x0300 && c <= 0x036F) || (c >= 0x203F && c <= 0x2040));
+ return (is_PN_CHARS_BASE(c) || c == 0xB7 || (c >= 0x0300 && c <= 0x036F) ||
+ (c >= 0x203F && c <= 0x2040));
}
static SerdStatus
read_PN_CHARS(SerdReader* reader, Ref dest)
{
- uint32_t code = 0;
- const int c = peek_byte(reader);
- SerdStatus st = SERD_SUCCESS;
- if (is_alpha(c) || is_digit(c) || c == '_' || c == '-') {
- push_byte(reader, dest, eat_byte_safe(reader, c));
- } else if (c == EOF || !(c & 0x80)) {
- return SERD_FAILURE;
- } else if ((st = read_utf8_code(reader, dest, &code,
- (uint8_t)eat_byte_safe(reader, c)))) {
- return st;
- } else if (!is_PN_CHARS(code)) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "invalid character U+%04X in name\n", code);
- }
- return st;
+ uint32_t code = 0;
+ const int c = peek_byte(reader);
+ SerdStatus st = SERD_SUCCESS;
+ if (is_alpha(c) || is_digit(c) || c == '_' || c == '-') {
+ push_byte(reader, dest, eat_byte_safe(reader, c));
+ } else if (c == EOF || !(c & 0x80)) {
+ return SERD_FAILURE;
+ } else if ((st = read_utf8_code(
+ reader, dest, &code, (uint8_t)eat_byte_safe(reader, c)))) {
+ return st;
+ } else if (!is_PN_CHARS(code)) {
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "invalid character U+%04X in name\n", code);
+ }
+ return st;
}
static SerdStatus
read_PERCENT(SerdReader* reader, Ref dest)
{
- push_byte(reader, dest, eat_byte_safe(reader, '%'));
- const uint8_t h1 = read_HEX(reader);
- const uint8_t h2 = read_HEX(reader);
- if (h1 && h2) {
- push_byte(reader, dest, h1);
- return push_byte(reader, dest, h2);
- }
- return SERD_ERR_BAD_SYNTAX;
+ push_byte(reader, dest, eat_byte_safe(reader, '%'));
+ const uint8_t h1 = read_HEX(reader);
+ const uint8_t h2 = read_HEX(reader);
+ if (h1 && h2) {
+ push_byte(reader, dest, h1);
+ return push_byte(reader, dest, h2);
+ }
+
+ return SERD_ERR_BAD_SYNTAX;
}
static SerdStatus
read_PN_LOCAL_ESC(SerdReader* reader, Ref dest)
{
- eat_byte_safe(reader, '\\');
-
- const int c = peek_byte(reader);
- switch (c) {
- case '!':
- case '#':
- case '$':
- case '%':
- case '&':
- case '\'':
- case '(':
- case ')':
- case '*':
- case '+':
- case ',':
- case '-':
- case '.':
- case '/':
- case ';':
- case '=':
- case '?':
- case '@':
- case '_':
- case '~':
- push_byte(reader, dest, eat_byte_safe(reader, c));
- break;
- default:
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid escape\n");
- }
-
- return SERD_SUCCESS;
+ eat_byte_safe(reader, '\\');
+
+ const int c = peek_byte(reader);
+ switch (c) {
+ case '!':
+ case '#':
+ case '$':
+ case '%':
+ case '&':
+ case '\'':
+ case '(':
+ case ')':
+ case '*':
+ case '+':
+ case ',':
+ case '-':
+ case '.':
+ case '/':
+ case ';':
+ case '=':
+ case '?':
+ case '@':
+ case '_':
+ case '~':
+ push_byte(reader, dest, eat_byte_safe(reader, c));
+ break;
+ default:
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid escape\n");
+ }
+
+ return SERD_SUCCESS;
}
static SerdStatus
read_PLX(SerdReader* reader, Ref dest)
{
- const int c = peek_byte(reader);
- switch (c) {
- case '%':
- return read_PERCENT(reader, dest);
- case '\\':
- return read_PN_LOCAL_ESC(reader, dest);
- default:
- return SERD_FAILURE;
- }
+ const int c = peek_byte(reader);
+ switch (c) {
+ case '%':
+ return read_PERCENT(reader, dest);
+ case '\\':
+ return read_PN_LOCAL_ESC(reader, dest);
+ default:
+ return SERD_FAILURE;
+ }
}
static SerdStatus
read_PN_LOCAL(SerdReader* reader, Ref dest, bool* ate_dot)
{
- int c = peek_byte(reader);
- SerdStatus st = SERD_SUCCESS;
- bool trailing_unescaped_dot = false;
- switch (c) {
- case '0': case '1': case '2': case '3': case '4': case '5':
- case '6': case '7': case '8': case '9': case ':': case '_':
- push_byte(reader, dest, eat_byte_safe(reader, c));
- break;
- default:
- if ((st = read_PLX(reader, dest)) > SERD_FAILURE) {
- return r_err(reader, st, "bad escape\n");
- } else if (st != SERD_SUCCESS && read_PN_CHARS_BASE(reader, dest)) {
- return SERD_FAILURE;
- }
- }
-
- while ((c = peek_byte(reader))) { // Middle: (PN_CHARS | '.' | ':')*
- if (c == '.' || c == ':') {
- push_byte(reader, dest, eat_byte_safe(reader, c));
- } else if ((st = read_PLX(reader, dest)) > SERD_FAILURE) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "bad escape\n");
- } else if (st != SERD_SUCCESS && (st = read_PN_CHARS(reader, dest))) {
- break;
- }
- trailing_unescaped_dot = (c == '.');
- }
-
- SerdNode* const n = deref(reader, dest);
- if (trailing_unescaped_dot) {
- // Ate trailing dot, pop it from stack/node and inform caller
- --n->n_bytes;
- serd_stack_pop(&reader->stack, 1);
- *ate_dot = true;
- }
-
- return (st > SERD_FAILURE) ? st : SERD_SUCCESS;
+ int c = peek_byte(reader);
+ SerdStatus st = SERD_SUCCESS;
+ bool trailing_unescaped_dot = false;
+ switch (c) {
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ case '8':
+ case '9':
+ case ':':
+ case '_':
+ push_byte(reader, dest, eat_byte_safe(reader, c));
+ break;
+ default:
+ if ((st = read_PLX(reader, dest)) > SERD_FAILURE) {
+ return r_err(reader, st, "bad escape\n");
+ } else if (st != SERD_SUCCESS && read_PN_CHARS_BASE(reader, dest)) {
+ return SERD_FAILURE;
+ }
+ }
+
+ while ((c = peek_byte(reader))) { // Middle: (PN_CHARS | '.' | ':')*
+ if (c == '.' || c == ':') {
+ push_byte(reader, dest, eat_byte_safe(reader, c));
+ } else if ((st = read_PLX(reader, dest)) > SERD_FAILURE) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "bad escape\n");
+ } else if (st != SERD_SUCCESS && (st = read_PN_CHARS(reader, dest))) {
+ break;
+ }
+ trailing_unescaped_dot = (c == '.');
+ }
+
+ SerdNode* const n = deref(reader, dest);
+ if (trailing_unescaped_dot) {
+ // Ate trailing dot, pop it from stack/node and inform caller
+ --n->n_bytes;
+ serd_stack_pop(&reader->stack, 1);
+ *ate_dot = true;
+ }
+
+ return (st > SERD_FAILURE) ? st : SERD_SUCCESS;
}
// Read the remainder of a PN_PREFIX after some initial characters
static SerdStatus
read_PN_PREFIX_tail(SerdReader* reader, Ref dest)
{
- int c = 0;
- while ((c = peek_byte(reader))) { // Middle: (PN_CHARS | '.')*
- if (c == '.') {
- push_byte(reader, dest, eat_byte_safe(reader, c));
- } else if (read_PN_CHARS(reader, dest)) {
- break;
- }
- }
-
- const SerdNode* const n = deref(reader, dest);
- if (n->buf[n->n_bytes - 1] == '.' && read_PN_CHARS(reader, dest)) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "prefix ends with `.'\n");
- }
-
- return SERD_SUCCESS;
+ int c = 0;
+ while ((c = peek_byte(reader))) { // Middle: (PN_CHARS | '.')*
+ if (c == '.') {
+ push_byte(reader, dest, eat_byte_safe(reader, c));
+ } else if (read_PN_CHARS(reader, dest)) {
+ break;
+ }
+ }
+
+ const SerdNode* const n = deref(reader, dest);
+ if (n->buf[n->n_bytes - 1] == '.' && read_PN_CHARS(reader, dest)) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "prefix ends with `.'\n");
+ }
+
+ return SERD_SUCCESS;
}
static SerdStatus
read_PN_PREFIX(SerdReader* reader, Ref dest)
{
- if (!read_PN_CHARS_BASE(reader, dest)) {
- return read_PN_PREFIX_tail(reader, dest);
- }
- return SERD_FAILURE;
+ if (!read_PN_CHARS_BASE(reader, dest)) {
+ return read_PN_PREFIX_tail(reader, dest);
+ }
+
+ return SERD_FAILURE;
}
static SerdStatus
read_LANGTAG(SerdReader* reader, Ref* dest)
{
- int c = peek_byte(reader);
- if (!is_alpha(c)) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "unexpected `%c'\n", c);
- }
-
- *dest = push_node(reader, SERD_LITERAL, "", 0);
-
- SerdStatus st = SERD_SUCCESS;
- TRY(st, push_byte(reader, *dest, eat_byte_safe(reader, c)));
- while ((c = peek_byte(reader)) && is_alpha(c)) {
- TRY(st, push_byte(reader, *dest, eat_byte_safe(reader, c)));
- }
- while (peek_byte(reader) == '-') {
- TRY(st, push_byte(reader, *dest, eat_byte_safe(reader, '-')));
- while ((c = peek_byte(reader)) && (is_alpha(c) || is_digit(c))) {
- TRY(st, push_byte(reader, *dest, eat_byte_safe(reader, c)));
- }
- }
- return SERD_SUCCESS;
+ int c = peek_byte(reader);
+ if (!is_alpha(c)) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "unexpected `%c'\n", c);
+ }
+
+ *dest = push_node(reader, SERD_LITERAL, "", 0);
+
+ SerdStatus st = SERD_SUCCESS;
+ TRY(st, push_byte(reader, *dest, eat_byte_safe(reader, c)));
+ while ((c = peek_byte(reader)) && is_alpha(c)) {
+ TRY(st, push_byte(reader, *dest, eat_byte_safe(reader, c)));
+ }
+
+ while (peek_byte(reader) == '-') {
+ TRY(st, push_byte(reader, *dest, eat_byte_safe(reader, '-')));
+ while ((c = peek_byte(reader)) && (is_alpha(c) || is_digit(c))) {
+ TRY(st, push_byte(reader, *dest, eat_byte_safe(reader, c)));
+ }
+ }
+
+ return SERD_SUCCESS;
}
static SerdStatus
read_IRIREF_scheme(SerdReader* reader, Ref dest)
{
- int c = peek_byte(reader);
- if (!is_alpha(c)) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "bad IRI scheme start `%c'\n", c);
- }
-
- while ((c = peek_byte(reader)) != EOF) {
- if (c == '>') {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "missing IRI scheme\n");
- }
-
- if (!is_uri_scheme_char(c)) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "bad IRI scheme char U+%04X (%c)\n",
- (unsigned)c,
- (char)c);
- }
-
- push_byte(reader, dest, eat_byte_safe(reader, c));
- if (c == ':') {
- return SERD_SUCCESS; // End of scheme
- }
- }
-
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "unexpected end of file\n");
+ int c = peek_byte(reader);
+ if (!is_alpha(c)) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "bad IRI scheme start `%c'\n", c);
+ }
+
+ while ((c = peek_byte(reader)) != EOF) {
+ if (c == '>') {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "missing IRI scheme\n");
+ }
+
+ if (!is_uri_scheme_char(c)) {
+ return r_err(reader,
+ SERD_ERR_BAD_SYNTAX,
+ "bad IRI scheme char U+%04X (%c)\n",
+ (unsigned)c,
+ (char)c);
+ }
+
+ push_byte(reader, dest, eat_byte_safe(reader, c));
+ if (c == ':') {
+ return SERD_SUCCESS; // End of scheme
+ }
+ }
+
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "unexpected end of file\n");
}
static SerdStatus
read_IRIREF(SerdReader* reader, Ref* dest)
{
- if (!eat_byte_check(reader, '<')) {
- return SERD_ERR_BAD_SYNTAX;
- }
-
- *dest = push_node(reader, SERD_URI, "", 0);
-
- if (!fancy_syntax(reader) && read_IRIREF_scheme(reader, *dest)) {
- *dest = pop_node(reader, *dest);
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "expected IRI scheme\n");
- }
-
- SerdStatus st = SERD_SUCCESS;
- uint32_t code = 0;
- while (!st) {
- const int c = eat_byte_safe(reader, peek_byte(reader));
- switch (c) {
- case '"':
- case '<':
- *dest = pop_node(reader, *dest);
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "invalid IRI character `%c'\n", c);
- case '>':
- return SERD_SUCCESS;
- case '\\':
- if (read_UCHAR(reader, *dest, &code)) {
- *dest = pop_node(reader, *dest);
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "invalid IRI escape\n");
- }
- switch (code) {
- case 0: case ' ': case '<': case '>':
- *dest = pop_node(reader, *dest);
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "invalid escaped IRI character U+%04X\n", code);
- default:
- break;
- }
- break;
- case '^':
- case '`':
- case '{':
- case '|':
- case '}':
- *dest = pop_node(reader, *dest);
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "invalid IRI character `%c'\n", c);
- default:
- if (c <= 0x20) {
- r_err(reader, SERD_ERR_BAD_SYNTAX,
- "invalid IRI character (escape %%%02X)\n",
- (unsigned)c);
- if (reader->strict) {
- *dest = pop_node(reader, *dest);
- return SERD_ERR_BAD_SYNTAX;
- }
- st = SERD_FAILURE;
- push_byte(reader, *dest, c);
- } else if (!(c & 0x80)) {
- push_byte(reader, *dest, c);
- } else if (read_utf8_character(reader, *dest, (uint8_t)c)) {
- if (reader->strict) {
- *dest = pop_node(reader, *dest);
- return SERD_ERR_BAD_SYNTAX;
- }
- }
- }
- }
-
- *dest = pop_node(reader, *dest);
- return st;
+ if (!eat_byte_check(reader, '<')) {
+ return SERD_ERR_BAD_SYNTAX;
+ }
+
+ *dest = push_node(reader, SERD_URI, "", 0);
+
+ if (!fancy_syntax(reader) && read_IRIREF_scheme(reader, *dest)) {
+ *dest = pop_node(reader, *dest);
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "expected IRI scheme\n");
+ }
+
+ SerdStatus st = SERD_SUCCESS;
+ uint32_t code = 0;
+ while (!st) {
+ const int c = eat_byte_safe(reader, peek_byte(reader));
+ switch (c) {
+ case '"':
+ case '<':
+ *dest = pop_node(reader, *dest);
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "invalid IRI character `%c'\n", c);
+
+ case '>':
+ return SERD_SUCCESS;
+
+ case '\\':
+ if (read_UCHAR(reader, *dest, &code)) {
+ *dest = pop_node(reader, *dest);
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid IRI escape\n");
+ }
+
+ switch (code) {
+ case 0:
+ case ' ':
+ case '<':
+ case '>':
+ *dest = pop_node(reader, *dest);
+ return r_err(reader,
+ SERD_ERR_BAD_SYNTAX,
+ "invalid escaped IRI character U+%04X\n",
+ code);
+ default:
+ break;
+ }
+ break;
+
+ case '^':
+ case '`':
+ case '{':
+ case '|':
+ case '}':
+ *dest = pop_node(reader, *dest);
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "invalid IRI character `%c'\n", c);
+
+ default:
+ if (c <= 0x20) {
+ r_err(reader,
+ SERD_ERR_BAD_SYNTAX,
+ "invalid IRI character (escape %%%02X)\n",
+ (unsigned)c);
+ if (reader->strict) {
+ *dest = pop_node(reader, *dest);
+ return SERD_ERR_BAD_SYNTAX;
+ }
+ st = SERD_FAILURE;
+ push_byte(reader, *dest, c);
+ } else if (!(c & 0x80)) {
+ push_byte(reader, *dest, c);
+ } else if (read_utf8_character(reader, *dest, (uint8_t)c)) {
+ if (reader->strict) {
+ *dest = pop_node(reader, *dest);
+ return SERD_ERR_BAD_SYNTAX;
+ }
+ }
+ }
+ }
+
+ *dest = pop_node(reader, *dest);
+ return st;
}
static SerdStatus
read_PrefixedName(SerdReader* reader, Ref dest, bool read_prefix, bool* ate_dot)
{
- SerdStatus st = SERD_SUCCESS;
- if (read_prefix && ((st = read_PN_PREFIX(reader, dest)) > SERD_FAILURE)) {
- return st;
- }
+ SerdStatus st = SERD_SUCCESS;
+ if (read_prefix && ((st = read_PN_PREFIX(reader, dest)) > SERD_FAILURE)) {
+ return st;
+ }
- if (peek_byte(reader) != ':') {
- return SERD_FAILURE;
- }
+ if (peek_byte(reader) != ':') {
+ return SERD_FAILURE;
+ }
- push_byte(reader, dest, eat_byte_safe(reader, ':'));
+ push_byte(reader, dest, eat_byte_safe(reader, ':'));
- st = read_PN_LOCAL(reader, dest, ate_dot);
+ st = read_PN_LOCAL(reader, dest, ate_dot);
- return (st > SERD_FAILURE) ? st : SERD_SUCCESS;
+ return (st > SERD_FAILURE) ? st : SERD_SUCCESS;
}
static SerdStatus
read_0_9(SerdReader* reader, Ref str, bool at_least_one)
{
- unsigned count = 0;
- SerdStatus st = SERD_SUCCESS;
- for (int c = 0; is_digit((c = peek_byte(reader))); ++count) {
- TRY(st, push_byte(reader, str, eat_byte_safe(reader, c)));
- }
- if (at_least_one && count == 0) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "expected digit\n");
- }
- return SERD_SUCCESS;
+ unsigned count = 0;
+ SerdStatus st = SERD_SUCCESS;
+ for (int c = 0; is_digit((c = peek_byte(reader))); ++count) {
+ TRY(st, push_byte(reader, str, eat_byte_safe(reader, c)));
+ }
+
+ if (at_least_one && count == 0) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "expected digit\n");
+ }
+
+ return SERD_SUCCESS;
}
static SerdStatus
read_number(SerdReader* reader, Ref* dest, Ref* datatype, bool* ate_dot)
{
- #define XSD_DECIMAL NS_XSD "decimal"
- #define XSD_DOUBLE NS_XSD "double"
- #define XSD_INTEGER NS_XSD "integer"
-
- *dest = push_node(reader, SERD_LITERAL, "", 0);
-
- SerdStatus st = SERD_SUCCESS;
- int c = peek_byte(reader);
- bool has_decimal = false;
- if (c == '-' || c == '+') {
- push_byte(reader, *dest, eat_byte_safe(reader, c));
- }
- if ((c = peek_byte(reader)) == '.') {
- has_decimal = true;
- // decimal case 2 (e.g. '.0' or `-.0' or `+.0')
- push_byte(reader, *dest, eat_byte_safe(reader, c));
- TRY(st, read_0_9(reader, *dest, true));
- } else {
- // all other cases ::= ( '-' | '+' ) [0-9]+ ( . )? ( [0-9]+ )? ...
- TRY(st, read_0_9(reader, *dest, true));
- if ((c = peek_byte(reader)) == '.') {
- has_decimal = true;
-
- // Annoyingly, dot can be end of statement, so tentatively eat
- eat_byte_safe(reader, c);
- c = peek_byte(reader);
- if (!is_digit(c) && c != 'e' && c != 'E') {
- *ate_dot = true; // Force caller to deal with stupid grammar
- return SERD_SUCCESS; // Next byte is not a number character
- }
-
- push_byte(reader, *dest, '.');
- read_0_9(reader, *dest, false);
- }
- }
- c = peek_byte(reader);
- if (c == 'e' || c == 'E') {
- // double
- push_byte(reader, *dest, eat_byte_safe(reader, c));
- switch ((c = peek_byte(reader))) {
- case '+': case '-':
- push_byte(reader, *dest, eat_byte_safe(reader, c));
- default: break;
- }
- TRY(st, read_0_9(reader, *dest, true));
- *datatype = push_node(reader, SERD_URI,
- XSD_DOUBLE, sizeof(XSD_DOUBLE) - 1);
- } else if (has_decimal) {
- *datatype = push_node(reader, SERD_URI,
- XSD_DECIMAL, sizeof(XSD_DECIMAL) - 1);
- } else {
- *datatype = push_node(reader, SERD_URI,
- XSD_INTEGER, sizeof(XSD_INTEGER) - 1);
- }
-
- return SERD_SUCCESS;
+#define XSD_DECIMAL NS_XSD "decimal"
+#define XSD_DOUBLE NS_XSD "double"
+#define XSD_INTEGER NS_XSD "integer"
+
+ *dest = push_node(reader, SERD_LITERAL, "", 0);
+
+ SerdStatus st = SERD_SUCCESS;
+ int c = peek_byte(reader);
+ bool has_decimal = false;
+ if (c == '-' || c == '+') {
+ push_byte(reader, *dest, eat_byte_safe(reader, c));
+ }
+ if ((c = peek_byte(reader)) == '.') {
+ has_decimal = true;
+ // decimal case 2 (e.g. '.0' or `-.0' or `+.0')
+ push_byte(reader, *dest, eat_byte_safe(reader, c));
+ TRY(st, read_0_9(reader, *dest, true));
+ } else {
+ // all other cases ::= ( '-' | '+' ) [0-9]+ ( . )? ( [0-9]+ )? ...
+ TRY(st, read_0_9(reader, *dest, true));
+ if ((c = peek_byte(reader)) == '.') {
+ has_decimal = true;
+
+ // Annoyingly, dot can be end of statement, so tentatively eat
+ eat_byte_safe(reader, c);
+ c = peek_byte(reader);
+ if (!is_digit(c) && c != 'e' && c != 'E') {
+ *ate_dot = true; // Force caller to deal with stupid grammar
+ return SERD_SUCCESS; // Next byte is not a number character
+ }
+
+ push_byte(reader, *dest, '.');
+ read_0_9(reader, *dest, false);
+ }
+ }
+ c = peek_byte(reader);
+ if (c == 'e' || c == 'E') {
+ // double
+ push_byte(reader, *dest, eat_byte_safe(reader, c));
+ switch ((c = peek_byte(reader))) {
+ case '+':
+ case '-':
+ push_byte(reader, *dest, eat_byte_safe(reader, c));
+ default:
+ break;
+ }
+ TRY(st, read_0_9(reader, *dest, true));
+ *datatype = push_node(reader, SERD_URI, XSD_DOUBLE, sizeof(XSD_DOUBLE) - 1);
+ } else if (has_decimal) {
+ *datatype =
+ push_node(reader, SERD_URI, XSD_DECIMAL, sizeof(XSD_DECIMAL) - 1);
+ } else {
+ *datatype =
+ push_node(reader, SERD_URI, XSD_INTEGER, sizeof(XSD_INTEGER) - 1);
+ }
+
+ return SERD_SUCCESS;
}
static SerdStatus
read_iri(SerdReader* reader, Ref* dest, bool* ate_dot)
{
- switch (peek_byte(reader)) {
- case '<':
- return read_IRIREF(reader, dest);
- default:
- *dest = push_node(reader, SERD_CURIE, "", 0);
- return read_PrefixedName(reader, *dest, true, ate_dot);
- }
+ switch (peek_byte(reader)) {
+ case '<':
+ return read_IRIREF(reader, dest);
+ default:
+ *dest = push_node(reader, SERD_CURIE, "", 0);
+ return read_PrefixedName(reader, *dest, true, ate_dot);
+ }
}
static SerdStatus
-read_literal(SerdReader* reader, Ref* dest,
- Ref* datatype, Ref* lang, SerdNodeFlags* flags, bool* ate_dot)
+read_literal(SerdReader* reader,
+ Ref* dest,
+ Ref* datatype,
+ Ref* lang,
+ SerdNodeFlags* flags,
+ bool* ate_dot)
{
- *dest = push_node(reader, SERD_LITERAL, "", 0);
-
- SerdStatus st = read_String(reader, *dest, flags);
- if (st) {
- *dest = pop_node(reader, *dest);
- return st;
- }
-
- switch (peek_byte(reader)) {
- case '@':
- eat_byte_safe(reader, '@');
- if ((st = read_LANGTAG(reader, lang))) {
- *datatype = pop_node(reader, *datatype);
- *lang = pop_node(reader, *lang);
- *dest = pop_node(reader, *dest);
- return r_err(reader, st, "bad literal\n");
- }
- break;
- case '^':
- eat_byte_safe(reader, '^');
- eat_byte_check(reader, '^');
- if ((st = read_iri(reader, datatype, ate_dot))) {
- *datatype = pop_node(reader, *datatype);
- *lang = pop_node(reader, *lang);
- *dest = pop_node(reader, *dest);
- return r_err(reader, st, "bad literal\n");
- }
- break;
- }
- return SERD_SUCCESS;
+ *dest = push_node(reader, SERD_LITERAL, "", 0);
+
+ SerdStatus st = read_String(reader, *dest, flags);
+ if (st) {
+ *dest = pop_node(reader, *dest);
+ return st;
+ }
+
+ switch (peek_byte(reader)) {
+ case '@':
+ eat_byte_safe(reader, '@');
+ if ((st = read_LANGTAG(reader, lang))) {
+ *datatype = pop_node(reader, *datatype);
+ *lang = pop_node(reader, *lang);
+ *dest = pop_node(reader, *dest);
+ return r_err(reader, st, "bad literal\n");
+ }
+ break;
+ case '^':
+ eat_byte_safe(reader, '^');
+ eat_byte_check(reader, '^');
+ if ((st = read_iri(reader, datatype, ate_dot))) {
+ *datatype = pop_node(reader, *datatype);
+ *lang = pop_node(reader, *lang);
+ *dest = pop_node(reader, *dest);
+ return r_err(reader, st, "bad literal\n");
+ }
+ break;
+ }
+
+ return SERD_SUCCESS;
}
static SerdStatus
read_verb(SerdReader* reader, Ref* dest)
{
- if (peek_byte(reader) == '<') {
- return read_IRIREF(reader, dest);
- }
-
- /* Either a qname, or "a". Read the prefix first, and if it is in fact
- "a", produce that instead.
- */
- *dest = push_node(reader, SERD_CURIE, "", 0);
-
- SerdStatus st = read_PN_PREFIX(reader, *dest);
- bool ate_dot = false;
- SerdNode* node = deref(reader, *dest);
- const int next = peek_byte(reader);
- if (!st && node->n_bytes == 1 && node->buf[0] == 'a' &&
- next != ':' && !is_PN_CHARS_BASE((uint32_t)next)) {
- pop_node(reader, *dest);
- *dest = push_node(reader, SERD_URI, NS_RDF "type", 47);
- return SERD_SUCCESS;
- }
-
- if (st > SERD_FAILURE ||
- read_PrefixedName(reader, *dest, false, &ate_dot) ||
- ate_dot) {
- *dest = pop_node(reader, *dest);
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "bad verb\n");
- }
-
- return SERD_SUCCESS;
+ if (peek_byte(reader) == '<') {
+ return read_IRIREF(reader, dest);
+ }
+
+ /* Either a qname, or "a". Read the prefix first, and if it is in fact
+ "a", produce that instead.
+ */
+ *dest = push_node(reader, SERD_CURIE, "", 0);
+
+ SerdStatus st = read_PN_PREFIX(reader, *dest);
+ bool ate_dot = false;
+ SerdNode* node = deref(reader, *dest);
+ const int next = peek_byte(reader);
+ if (!st && node->n_bytes == 1 && node->buf[0] == 'a' && next != ':' &&
+ !is_PN_CHARS_BASE((uint32_t)next)) {
+ pop_node(reader, *dest);
+ *dest = push_node(reader, SERD_URI, NS_RDF "type", 47);
+ return SERD_SUCCESS;
+ }
+
+ if (st > SERD_FAILURE || read_PrefixedName(reader, *dest, false, &ate_dot) ||
+ ate_dot) {
+ *dest = pop_node(reader, *dest);
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "bad verb\n");
+ }
+
+ return SERD_SUCCESS;
}
static SerdStatus
read_BLANK_NODE_LABEL(SerdReader* reader, Ref* dest, bool* ate_dot)
{
- eat_byte_safe(reader, '_');
- eat_byte_check(reader, ':');
-
- const Ref ref = *dest =
- push_node(reader,
- SERD_BLANK,
- reader->bprefix ? (char*)reader->bprefix : "",
- reader->bprefix_len);
-
- int c = peek_byte(reader); // First: (PN_CHARS | '_' | [0-9])
- if (is_digit(c) || c == '_') {
- push_byte(reader, ref, eat_byte_safe(reader, c));
- } else if (read_PN_CHARS(reader, ref)) {
- *dest = pop_node(reader, *dest);
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid name start\n");
- }
-
- while ((c = peek_byte(reader))) { // Middle: (PN_CHARS | '.')*
- if (c == '.') {
- push_byte(reader, ref, eat_byte_safe(reader, c));
- } else if (read_PN_CHARS(reader, ref)) {
- break;
- }
- }
-
- SerdNode* n = deref(reader, ref);
- if (n->buf[n->n_bytes - 1] == '.' && read_PN_CHARS(reader, ref)) {
- // Ate trailing dot, pop it from stack/node and inform caller
- --n->n_bytes;
- serd_stack_pop(&reader->stack, 1);
- *ate_dot = true;
- }
-
- if (fancy_syntax(reader)) {
- if (is_digit(n->buf[reader->bprefix_len + 1])) {
- if ((n->buf[reader->bprefix_len]) == 'b') {
- ((char*)n->buf)[reader->bprefix_len] = 'B'; // Prevent clash
- reader->seen_genid = true;
- } else if (reader->seen_genid &&
- n->buf[reader->bprefix_len] == 'B') {
- *dest = pop_node(reader, *dest);
- return r_err(
- reader, SERD_ERR_ID_CLASH,
- "found both `b' and `B' blank IDs, prefix required\n");
- }
- }
- }
- return SERD_SUCCESS;
+ eat_byte_safe(reader, '_');
+ eat_byte_check(reader, ':');
+
+ const Ref ref = *dest =
+ push_node(reader,
+ SERD_BLANK,
+ reader->bprefix ? (char*)reader->bprefix : "",
+ reader->bprefix_len);
+
+ int c = peek_byte(reader); // First: (PN_CHARS | '_' | [0-9])
+ if (is_digit(c) || c == '_') {
+ push_byte(reader, ref, eat_byte_safe(reader, c));
+ } else if (read_PN_CHARS(reader, ref)) {
+ *dest = pop_node(reader, *dest);
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid name start\n");
+ }
+
+ while ((c = peek_byte(reader))) { // Middle: (PN_CHARS | '.')*
+ if (c == '.') {
+ push_byte(reader, ref, eat_byte_safe(reader, c));
+ } else if (read_PN_CHARS(reader, ref)) {
+ break;
+ }
+ }
+
+ SerdNode* n = deref(reader, ref);
+ if (n->buf[n->n_bytes - 1] == '.' && read_PN_CHARS(reader, ref)) {
+ // Ate trailing dot, pop it from stack/node and inform caller
+ --n->n_bytes;
+ serd_stack_pop(&reader->stack, 1);
+ *ate_dot = true;
+ }
+
+ if (fancy_syntax(reader)) {
+ if (is_digit(n->buf[reader->bprefix_len + 1])) {
+ if ((n->buf[reader->bprefix_len]) == 'b') {
+ ((char*)n->buf)[reader->bprefix_len] = 'B'; // Prevent clash
+ reader->seen_genid = true;
+ } else if (reader->seen_genid && n->buf[reader->bprefix_len] == 'B') {
+ *dest = pop_node(reader, *dest);
+ return r_err(reader,
+ SERD_ERR_ID_CLASH,
+ "found both `b' and `B' blank IDs, prefix required\n");
+ }
+ }
+ }
+
+ return SERD_SUCCESS;
}
static Ref
read_blankName(SerdReader* reader)
{
- eat_byte_safe(reader, '=');
- if (eat_byte_check(reader, '=') != '=') {
- r_err(reader, SERD_ERR_BAD_SYNTAX, "expected `='\n");
- return 0;
- }
-
- Ref subject = 0;
- bool ate_dot = false;
- read_ws_star(reader);
- read_iri(reader, &subject, &ate_dot);
- return subject;
+ eat_byte_safe(reader, '=');
+ if (eat_byte_check(reader, '=') != '=') {
+ r_err(reader, SERD_ERR_BAD_SYNTAX, "expected `='\n");
+ return 0;
+ }
+
+ Ref subject = 0;
+ bool ate_dot = false;
+ read_ws_star(reader);
+ read_iri(reader, &subject, &ate_dot);
+ return subject;
}
static SerdStatus
read_anon(SerdReader* reader, ReadContext ctx, bool subject, Ref* dest)
{
- const SerdStatementFlags old_flags = *ctx.flags;
- bool empty = false;
- eat_byte_safe(reader, '[');
- if ((empty = peek_delim(reader, ']'))) {
- *ctx.flags |= (subject) ? SERD_EMPTY_S : SERD_EMPTY_O;
- } else {
- *ctx.flags |= (subject) ? SERD_ANON_S_BEGIN : SERD_ANON_O_BEGIN;
- if (peek_delim(reader, '=')) {
- if (!(*dest = read_blankName(reader)) ||
- !eat_delim(reader, ';')) {
- return SERD_ERR_BAD_SYNTAX;
- }
- }
- }
-
- if (!*dest) {
- *dest = blank_id(reader);
- }
-
- SerdStatus st = SERD_SUCCESS;
- if (ctx.subject) {
- TRY(st, emit_statement(reader, ctx, *dest, 0, 0));
- }
-
- ctx.subject = *dest;
- if (!empty) {
- *ctx.flags &= ~(unsigned)SERD_LIST_CONT;
- if (!subject) {
- *ctx.flags |= SERD_ANON_CONT;
- }
- bool ate_dot_in_list = false;
- read_predicateObjectList(reader, ctx, &ate_dot_in_list);
- if (ate_dot_in_list) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "`.' inside blank\n");
- }
- read_ws_star(reader);
- if (reader->end_sink) {
- reader->end_sink(reader->handle, deref(reader, *dest));
- }
- *ctx.flags = old_flags;
- }
- return (eat_byte_check(reader, ']') == ']') ? SERD_SUCCESS
- : SERD_ERR_BAD_SYNTAX;
+ const SerdStatementFlags old_flags = *ctx.flags;
+ bool empty = false;
+ eat_byte_safe(reader, '[');
+ if ((empty = peek_delim(reader, ']'))) {
+ *ctx.flags |= (subject) ? SERD_EMPTY_S : SERD_EMPTY_O;
+ } else {
+ *ctx.flags |= (subject) ? SERD_ANON_S_BEGIN : SERD_ANON_O_BEGIN;
+ if (peek_delim(reader, '=')) {
+ if (!(*dest = read_blankName(reader)) || !eat_delim(reader, ';')) {
+ return SERD_ERR_BAD_SYNTAX;
+ }
+ }
+ }
+
+ if (!*dest) {
+ *dest = blank_id(reader);
+ }
+
+ SerdStatus st = SERD_SUCCESS;
+ if (ctx.subject) {
+ TRY(st, emit_statement(reader, ctx, *dest, 0, 0));
+ }
+
+ ctx.subject = *dest;
+ if (!empty) {
+ *ctx.flags &= ~(unsigned)SERD_LIST_CONT;
+ if (!subject) {
+ *ctx.flags |= SERD_ANON_CONT;
+ }
+
+ bool ate_dot_in_list = false;
+ read_predicateObjectList(reader, ctx, &ate_dot_in_list);
+ if (ate_dot_in_list) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "`.' inside blank\n");
+ }
+
+ read_ws_star(reader);
+ if (reader->end_sink) {
+ reader->end_sink(reader->handle, deref(reader, *dest));
+ }
+
+ *ctx.flags = old_flags;
+ }
+
+ return (eat_byte_check(reader, ']') == ']') ? SERD_SUCCESS
+ : SERD_ERR_BAD_SYNTAX;
}
/* If emit is true: recurses, calling statement_sink for every statement
@@ -1050,593 +1100,621 @@ read_anon(SerdReader* reader, ReadContext ctx, bool subject, Ref* dest)
static SerdStatus
read_object(SerdReader* reader, ReadContext* ctx, bool emit, bool* ate_dot)
{
- static const char* const XSD_BOOLEAN = NS_XSD "boolean";
- static const size_t XSD_BOOLEAN_LEN = 40;
+ static const char* const XSD_BOOLEAN = NS_XSD "boolean";
+ static const size_t XSD_BOOLEAN_LEN = 40;
#ifndef NDEBUG
- const size_t orig_stack_size = reader->stack.size;
+ const size_t orig_stack_size = reader->stack.size;
#endif
- SerdStatus ret = SERD_FAILURE;
-
- bool simple = (ctx->subject != 0);
- SerdNode* node = NULL;
- Ref o = 0;
- Ref datatype = 0;
- Ref lang = 0;
- uint32_t flags = 0;
- const int c = peek_byte(reader);
- if (!fancy_syntax(reader)) {
- switch (c) {
- case '"': case ':': case '<': case '_': break;
- default:
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "expected: ':', '<', or '_'\n");
- }
- }
- switch (c) {
- case EOF: case ')':
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "expected object\n");
- case '[':
- simple = false;
- ret = read_anon(reader, *ctx, false, &o);
- break;
- case '(':
- simple = false;
- ret = read_collection(reader, *ctx, &o);
- break;
- case '_':
- ret = read_BLANK_NODE_LABEL(reader, &o, ate_dot);
- break;
- case '<': case ':':
- ret = read_iri(reader, &o, ate_dot);
- break;
- case '+': case '-': case '.': case '0': case '1': case '2': case '3':
- case '4': case '5': case '6': case '7': case '8': case '9':
- ret = read_number(reader, &o, &datatype, ate_dot);
- break;
- case '\"':
- case '\'':
- ret = read_literal(reader, &o, &datatype, &lang, &flags, ate_dot);
- break;
- default:
- /* Either a boolean literal, or a qname. Read the prefix first, and if
- it is in fact a "true" or "false" literal, produce that instead.
- */
- o = push_node(reader, SERD_CURIE, "", 0);
- while (!read_PN_CHARS_BASE(reader, o)) {}
- node = deref(reader, o);
- if ((node->n_bytes == 4 && !memcmp(node->buf, "true", 4)) ||
- (node->n_bytes == 5 && !memcmp(node->buf, "false", 5))) {
- node->type = SERD_LITERAL;
- datatype = push_node(
- reader, SERD_URI, XSD_BOOLEAN, XSD_BOOLEAN_LEN);
- ret = SERD_SUCCESS;
- } else if (read_PN_PREFIX_tail(reader, o) > SERD_FAILURE) {
- ret = SERD_ERR_BAD_SYNTAX;
- } else {
- if ((ret = read_PrefixedName(reader, o, false, ate_dot))) {
- ret = ret > SERD_FAILURE ? ret : SERD_ERR_BAD_SYNTAX;
- pop_node(reader, o);
- return r_err(reader, ret, "expected prefixed name\n");
- }
- }
- }
-
- if (!ret && simple && o) {
- deref(reader, o)->flags = flags;
- }
-
- if (!ret && emit && simple) {
- ret = emit_statement(reader, *ctx, o, datatype, lang);
- } else if (!ret && !emit) {
- ctx->object = o;
- ctx->datatype = datatype;
- ctx->lang = lang;
- return SERD_SUCCESS;
- }
-
- pop_node(reader, lang);
- pop_node(reader, datatype);
- pop_node(reader, o);
+ SerdStatus ret = SERD_FAILURE;
+
+ bool simple = (ctx->subject != 0);
+ SerdNode* node = NULL;
+ Ref o = 0;
+ Ref datatype = 0;
+ Ref lang = 0;
+ uint32_t flags = 0;
+ const int c = peek_byte(reader);
+ if (!fancy_syntax(reader)) {
+ switch (c) {
+ case '"':
+ case ':':
+ case '<':
+ case '_':
+ break;
+ default:
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "expected: ':', '<', or '_'\n");
+ }
+ }
+ switch (c) {
+ case EOF:
+ case ')':
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "expected object\n");
+ case '[':
+ simple = false;
+ ret = read_anon(reader, *ctx, false, &o);
+ break;
+ case '(':
+ simple = false;
+ ret = read_collection(reader, *ctx, &o);
+ break;
+ case '_':
+ ret = read_BLANK_NODE_LABEL(reader, &o, ate_dot);
+ break;
+ case '<':
+ case ':':
+ ret = read_iri(reader, &o, ate_dot);
+ break;
+ case '+':
+ case '-':
+ case '.':
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ case '8':
+ case '9':
+ ret = read_number(reader, &o, &datatype, ate_dot);
+ break;
+ case '\"':
+ case '\'':
+ ret = read_literal(reader, &o, &datatype, &lang, &flags, ate_dot);
+ break;
+ default:
+ /* Either a boolean literal, or a qname. Read the prefix first, and if
+ it is in fact a "true" or "false" literal, produce that instead.
+ */
+ o = push_node(reader, SERD_CURIE, "", 0);
+ while (!read_PN_CHARS_BASE(reader, o)) {
+ }
+ node = deref(reader, o);
+ if ((node->n_bytes == 4 && !memcmp(node->buf, "true", 4)) ||
+ (node->n_bytes == 5 && !memcmp(node->buf, "false", 5))) {
+ node->type = SERD_LITERAL;
+ datatype = push_node(reader, SERD_URI, XSD_BOOLEAN, XSD_BOOLEAN_LEN);
+ ret = SERD_SUCCESS;
+ } else if (read_PN_PREFIX_tail(reader, o) > SERD_FAILURE) {
+ ret = SERD_ERR_BAD_SYNTAX;
+ } else {
+ if ((ret = read_PrefixedName(reader, o, false, ate_dot))) {
+ ret = ret > SERD_FAILURE ? ret : SERD_ERR_BAD_SYNTAX;
+ pop_node(reader, o);
+ return r_err(reader, ret, "expected prefixed name\n");
+ }
+ }
+ }
+
+ if (!ret && simple && o) {
+ deref(reader, o)->flags = flags;
+ }
+
+ if (!ret && emit && simple) {
+ ret = emit_statement(reader, *ctx, o, datatype, lang);
+ } else if (!ret && !emit) {
+ ctx->object = o;
+ ctx->datatype = datatype;
+ ctx->lang = lang;
+ return SERD_SUCCESS;
+ }
+
+ pop_node(reader, lang);
+ pop_node(reader, datatype);
+ pop_node(reader, o);
#ifndef NDEBUG
- assert(reader->stack.size == orig_stack_size);
+ assert(reader->stack.size == orig_stack_size);
#endif
- return ret;
+ return ret;
}
static SerdStatus
read_objectList(SerdReader* reader, ReadContext ctx, bool* ate_dot)
{
- SerdStatus st = SERD_SUCCESS;
- TRY(st, read_object(reader, &ctx, true, ate_dot));
- if (!fancy_syntax(reader) && peek_delim(reader, ',')) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "syntax does not support abbreviation\n");
- }
-
- while (!*ate_dot && eat_delim(reader, ',')) {
- st = read_object(reader, &ctx, true, ate_dot);
- }
- return st;
+ SerdStatus st = SERD_SUCCESS;
+ TRY(st, read_object(reader, &ctx, true, ate_dot));
+ if (!fancy_syntax(reader) && peek_delim(reader, ',')) {
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "syntax does not support abbreviation\n");
+ }
+
+ while (!*ate_dot && eat_delim(reader, ',')) {
+ st = read_object(reader, &ctx, true, ate_dot);
+ }
+
+ return st;
}
static SerdStatus
read_predicateObjectList(SerdReader* reader, ReadContext ctx, bool* ate_dot)
{
- SerdStatus st = SERD_SUCCESS;
- while (!(st = read_verb(reader, &ctx.predicate)) &&
- read_ws_star(reader) &&
- !(st = read_objectList(reader, ctx, ate_dot))) {
- ctx.predicate = pop_node(reader, ctx.predicate);
- if (*ate_dot) {
- return SERD_SUCCESS;
- }
-
- bool ate_semi = false;
- int c = 0;
- do {
- read_ws_star(reader);
- switch (c = peek_byte(reader)) {
- case EOF:
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "unexpected end of file\n");
- case '.': case ']': case '}':
- return SERD_SUCCESS;
- case ';':
- eat_byte_safe(reader, c);
- ate_semi = true;
- }
- } while (c == ';');
-
- if (!ate_semi) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "missing ';' or '.'\n");
- }
- }
-
- ctx.predicate = pop_node(reader, ctx.predicate);
- return st;
+ SerdStatus st = SERD_SUCCESS;
+ while (!(st = read_verb(reader, &ctx.predicate)) && read_ws_star(reader) &&
+ !(st = read_objectList(reader, ctx, ate_dot))) {
+ ctx.predicate = pop_node(reader, ctx.predicate);
+ if (*ate_dot) {
+ return SERD_SUCCESS;
+ }
+
+ bool ate_semi = false;
+ int c = 0;
+ do {
+ read_ws_star(reader);
+ switch (c = peek_byte(reader)) {
+ case EOF:
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "unexpected end of file\n");
+ case '.':
+ case ']':
+ case '}':
+ return SERD_SUCCESS;
+ case ';':
+ eat_byte_safe(reader, c);
+ ate_semi = true;
+ }
+ } while (c == ';');
+
+ if (!ate_semi) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "missing ';' or '.'\n");
+ }
+ }
+
+ ctx.predicate = pop_node(reader, ctx.predicate);
+ return st;
}
static SerdStatus
-end_collection(SerdReader* reader, ReadContext ctx, Ref n1, Ref n2, SerdStatus st)
+end_collection(SerdReader* reader,
+ ReadContext ctx,
+ Ref n1,
+ Ref n2,
+ SerdStatus st)
{
- pop_node(reader, n2);
- pop_node(reader, n1);
- *ctx.flags &= ~(unsigned)SERD_LIST_CONT;
- if (!st) {
- return (eat_byte_check(reader, ')') == ')') ? SERD_SUCCESS
- : SERD_ERR_BAD_SYNTAX;
- }
- return st;
+ pop_node(reader, n2);
+ pop_node(reader, n1);
+ *ctx.flags &= ~(unsigned)SERD_LIST_CONT;
+ if (!st) {
+ return (eat_byte_check(reader, ')') == ')') ? SERD_SUCCESS
+ : SERD_ERR_BAD_SYNTAX;
+ }
+
+ return st;
}
static SerdStatus
read_collection(SerdReader* reader, ReadContext ctx, Ref* dest)
{
- SerdStatus st = SERD_SUCCESS;
- eat_byte_safe(reader, '(');
- bool end = peek_delim(reader, ')');
- *dest = end ? reader->rdf_nil : blank_id(reader);
- if (ctx.subject) {
- // subject predicate _:head
- *ctx.flags |= (end ? 0 : SERD_LIST_O_BEGIN);
- TRY(st, emit_statement(reader, ctx, *dest, 0, 0));
- *ctx.flags |= SERD_LIST_CONT;
- } else {
- *ctx.flags |= (end ? 0 : SERD_LIST_S_BEGIN);
- }
-
- if (end) {
- return end_collection(reader, ctx, 0, 0, st);
- }
-
- /* The order of node allocation here is necessarily not in stack order,
- so we create two nodes and recycle them throughout. */
- Ref n1 = push_node_padded(reader, genid_size(reader), SERD_BLANK, "", 0);
- Ref n2 = 0;
- Ref node = n1;
- Ref rest = 0;
-
- ctx.subject = *dest;
- while (!peek_delim(reader, ')')) {
- // _:node rdf:first object
- ctx.predicate = reader->rdf_first;
- bool ate_dot = false;
- if ((st = read_object(reader, &ctx, true, &ate_dot)) || ate_dot) {
- return end_collection(reader, ctx, n1, n2, st);
- }
-
- if (!(end = peek_delim(reader, ')'))) {
- /* Give rest a new ID. Done as late as possible to ensure it is
- used and > IDs generated by read_object above. */
- if (!rest) {
- rest = n2 = blank_id(reader); // First pass, push
- } else {
- set_blank_id(reader, rest, genid_size(reader));
- }
- }
-
- // _:node rdf:rest _:rest
- *ctx.flags |= SERD_LIST_CONT;
- ctx.predicate = reader->rdf_rest;
- TRY(st, emit_statement(reader, ctx,
- (end ? reader->rdf_nil : rest), 0, 0));
-
- ctx.subject = rest; // _:node = _:rest
- rest = node; // _:rest = (old)_:node
- node = ctx.subject; // invariant
- }
-
- return end_collection(reader, ctx, n1, n2, st);
+ SerdStatus st = SERD_SUCCESS;
+ eat_byte_safe(reader, '(');
+
+ bool end = peek_delim(reader, ')');
+
+ *dest = end ? reader->rdf_nil : blank_id(reader);
+ if (ctx.subject) {
+ // subject predicate _:head
+ *ctx.flags |= (end ? 0 : SERD_LIST_O_BEGIN);
+ TRY(st, emit_statement(reader, ctx, *dest, 0, 0));
+ *ctx.flags |= SERD_LIST_CONT;
+ } else {
+ *ctx.flags |= (end ? 0 : SERD_LIST_S_BEGIN);
+ }
+
+ if (end) {
+ return end_collection(reader, ctx, 0, 0, st);
+ }
+
+ /* The order of node allocation here is necessarily not in stack order,
+ so we create two nodes and recycle them throughout. */
+ Ref n1 = push_node_padded(reader, genid_size(reader), SERD_BLANK, "", 0);
+ Ref n2 = 0;
+ Ref node = n1;
+ Ref rest = 0;
+
+ ctx.subject = *dest;
+ while (!peek_delim(reader, ')')) {
+ // _:node rdf:first object
+ ctx.predicate = reader->rdf_first;
+ bool ate_dot = false;
+ if ((st = read_object(reader, &ctx, true, &ate_dot)) || ate_dot) {
+ return end_collection(reader, ctx, n1, n2, st);
+ }
+
+ if (!(end = peek_delim(reader, ')'))) {
+ /* Give rest a new ID. Done as late as possible to ensure it is
+ used and > IDs generated by read_object above. */
+ if (!rest) {
+ rest = n2 = blank_id(reader); // First pass, push
+ } else {
+ set_blank_id(reader, rest, genid_size(reader));
+ }
+ }
+
+ // _:node rdf:rest _:rest
+ *ctx.flags |= SERD_LIST_CONT;
+ ctx.predicate = reader->rdf_rest;
+ TRY(st, emit_statement(reader, ctx, (end ? reader->rdf_nil : rest), 0, 0));
+
+ ctx.subject = rest; // _:node = _:rest
+ rest = node; // _:rest = (old)_:node
+ node = ctx.subject; // invariant
+ }
+
+ return end_collection(reader, ctx, n1, n2, st);
}
static SerdStatus
read_subject(SerdReader* reader, ReadContext ctx, Ref* dest, int* s_type)
{
- SerdStatus st = SERD_SUCCESS;
- bool ate_dot = false;
- switch ((*s_type = peek_byte(reader))) {
- case '[':
- read_anon(reader, ctx, true, dest);
- break;
- case '(':
- st = read_collection(reader, ctx, dest);
- break;
- case '_':
- st = read_BLANK_NODE_LABEL(reader, dest, &ate_dot);
- break;
- default:
- st = read_iri(reader, dest, &ate_dot);
- }
-
- if (ate_dot) {
- pop_node(reader, *dest);
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "subject ends with `.'\n");
- }
-
- return st;
+ SerdStatus st = SERD_SUCCESS;
+ bool ate_dot = false;
+ switch ((*s_type = peek_byte(reader))) {
+ case '[':
+ read_anon(reader, ctx, true, dest);
+ break;
+ case '(':
+ st = read_collection(reader, ctx, dest);
+ break;
+ case '_':
+ st = read_BLANK_NODE_LABEL(reader, dest, &ate_dot);
+ break;
+ default:
+ st = read_iri(reader, dest, &ate_dot);
+ }
+
+ if (ate_dot) {
+ pop_node(reader, *dest);
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "subject ends with `.'\n");
+ }
+
+ return st;
}
static SerdStatus
read_labelOrSubject(SerdReader* reader, Ref* dest)
{
- bool ate_dot = false;
- switch (peek_byte(reader)) {
- case '[':
- eat_byte_safe(reader, '[');
- read_ws_star(reader);
- if (!eat_byte_check(reader, ']')) {
- return SERD_ERR_BAD_SYNTAX;
- }
- *dest = blank_id(reader);
- return SERD_SUCCESS;
- case '_':
- return read_BLANK_NODE_LABEL(reader, dest, &ate_dot);
- default:
- if (!read_iri(reader, dest, &ate_dot)) {
- return SERD_SUCCESS;
- } else {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "expected label or subject\n");
- }
- }
+ bool ate_dot = false;
+ switch (peek_byte(reader)) {
+ case '[':
+ eat_byte_safe(reader, '[');
+ read_ws_star(reader);
+ if (!eat_byte_check(reader, ']')) {
+ return SERD_ERR_BAD_SYNTAX;
+ }
+ *dest = blank_id(reader);
+ return SERD_SUCCESS;
+ case '_':
+ return read_BLANK_NODE_LABEL(reader, dest, &ate_dot);
+ default:
+ if (!read_iri(reader, dest, &ate_dot)) {
+ return SERD_SUCCESS;
+ } else {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "expected label or subject\n");
+ }
+ }
}
static SerdStatus
read_triples(SerdReader* reader, ReadContext ctx, bool* ate_dot)
{
- SerdStatus st = SERD_FAILURE;
- if (ctx.subject) {
- read_ws_star(reader);
- switch (peek_byte(reader)) {
- case '.':
- *ate_dot = eat_byte_safe(reader, '.');
- return SERD_FAILURE;
- case '}':
- return SERD_FAILURE;
- }
- st = read_predicateObjectList(reader, ctx, ate_dot);
- }
- ctx.subject = ctx.predicate = 0;
- return st > SERD_FAILURE ? st : SERD_SUCCESS;
+ SerdStatus st = SERD_FAILURE;
+ if (ctx.subject) {
+ read_ws_star(reader);
+ switch (peek_byte(reader)) {
+ case '.':
+ *ate_dot = eat_byte_safe(reader, '.');
+ return SERD_FAILURE;
+ case '}':
+ return SERD_FAILURE;
+ }
+ st = read_predicateObjectList(reader, ctx, ate_dot);
+ }
+
+ ctx.subject = ctx.predicate = 0;
+ return st > SERD_FAILURE ? st : SERD_SUCCESS;
}
static SerdStatus
read_base(SerdReader* reader, bool sparql, bool token)
{
- SerdStatus st = SERD_SUCCESS;
- if (token) {
- TRY(st, eat_string(reader, "base", 4));
- }
-
- read_ws_star(reader);
-
- Ref uri = 0;
- TRY(st, read_IRIREF(reader, &uri));
- if (reader->base_sink) {
- TRY(st, reader->base_sink(reader->handle, deref(reader, uri)));
- }
- pop_node(reader, uri);
-
- read_ws_star(reader);
- if (!sparql) {
- return eat_byte_check(reader, '.') ? SERD_SUCCESS : SERD_ERR_BAD_SYNTAX;
- }
-
- if (peek_byte(reader) == '.') {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "full stop after SPARQL BASE\n");
- }
-
- return SERD_SUCCESS;
+ SerdStatus st = SERD_SUCCESS;
+ if (token) {
+ TRY(st, eat_string(reader, "base", 4));
+ }
+
+ read_ws_star(reader);
+
+ Ref uri = 0;
+ TRY(st, read_IRIREF(reader, &uri));
+ if (reader->base_sink) {
+ TRY(st, reader->base_sink(reader->handle, deref(reader, uri)));
+ }
+ pop_node(reader, uri);
+
+ read_ws_star(reader);
+ if (!sparql) {
+ return eat_byte_check(reader, '.') ? SERD_SUCCESS : SERD_ERR_BAD_SYNTAX;
+ }
+
+ if (peek_byte(reader) == '.') {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "full stop after SPARQL BASE\n");
+ }
+
+ return SERD_SUCCESS;
}
static SerdStatus
read_prefixID(SerdReader* reader, bool sparql, bool token)
{
- SerdStatus st = SERD_SUCCESS;
- if (token) {
- TRY(st, eat_string(reader, "prefix", 6));
- }
-
- read_ws_star(reader);
- Ref name = push_node(reader, SERD_LITERAL, "", 0);
- if ((st = read_PN_PREFIX(reader, name)) > SERD_FAILURE) {
- return st;
- }
-
- if (eat_byte_check(reader, ':') != ':') {
- pop_node(reader, name);
- return SERD_ERR_BAD_SYNTAX;
- }
-
- read_ws_star(reader);
- Ref uri = 0;
- TRY(st, read_IRIREF(reader, &uri));
-
- if (reader->prefix_sink) {
- st = reader->prefix_sink(reader->handle,
- deref(reader, name),
- deref(reader, uri));
- }
- pop_node(reader, uri);
- pop_node(reader, name);
- if (!sparql) {
- read_ws_star(reader);
- st = eat_byte_check(reader, '.') ? SERD_SUCCESS : SERD_ERR_BAD_SYNTAX;
- }
- return st;
+ SerdStatus st = SERD_SUCCESS;
+ if (token) {
+ TRY(st, eat_string(reader, "prefix", 6));
+ }
+
+ read_ws_star(reader);
+ Ref name = push_node(reader, SERD_LITERAL, "", 0);
+ if ((st = read_PN_PREFIX(reader, name)) > SERD_FAILURE) {
+ return st;
+ }
+
+ if (eat_byte_check(reader, ':') != ':') {
+ pop_node(reader, name);
+ return SERD_ERR_BAD_SYNTAX;
+ }
+
+ read_ws_star(reader);
+ Ref uri = 0;
+ TRY(st, read_IRIREF(reader, &uri));
+
+ if (reader->prefix_sink) {
+ st = reader->prefix_sink(
+ reader->handle, deref(reader, name), deref(reader, uri));
+ }
+
+ pop_node(reader, uri);
+ pop_node(reader, name);
+ if (!sparql) {
+ read_ws_star(reader);
+ st = eat_byte_check(reader, '.') ? SERD_SUCCESS : SERD_ERR_BAD_SYNTAX;
+ }
+
+ return st;
}
static SerdStatus
read_directive(SerdReader* reader)
{
- const bool sparql = peek_byte(reader) != '@';
- if (!sparql) {
- eat_byte_safe(reader, '@');
- switch (peek_byte(reader)) {
- case 'B': case 'P':
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "uppercase directive\n");
- }
- }
-
- switch (peek_byte(reader)) {
- case 'B': case 'b': return read_base(reader, sparql, true);
- case 'P': case 'p': return read_prefixID(reader, sparql, true);
- default: break;
- }
-
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid directive\n");
+ const bool sparql = peek_byte(reader) != '@';
+ if (!sparql) {
+ eat_byte_safe(reader, '@');
+ switch (peek_byte(reader)) {
+ case 'B':
+ case 'P':
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "uppercase directive\n");
+ }
+ }
+
+ switch (peek_byte(reader)) {
+ case 'B':
+ case 'b':
+ return read_base(reader, sparql, true);
+ case 'P':
+ case 'p':
+ return read_prefixID(reader, sparql, true);
+ default:
+ break;
+ }
+
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid directive\n");
}
static SerdStatus
read_wrappedGraph(SerdReader* reader, ReadContext* ctx)
{
- if (!eat_byte_check(reader, '{')) {
- return SERD_ERR_BAD_SYNTAX;
- }
-
- read_ws_star(reader);
- while (peek_byte(reader) != '}') {
- bool ate_dot = false;
- int s_type = 0;
- ctx->subject = 0;
- SerdStatus st = read_subject(reader, *ctx, &ctx->subject, &s_type);
- if (st) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX, "bad subject\n");
- }
-
- if (read_triples(reader, *ctx, &ate_dot) && s_type != '[') {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "missing predicate object list\n");
- }
-
- pop_node(reader, ctx->subject);
- read_ws_star(reader);
- if (peek_byte(reader) == '.') {
- eat_byte_safe(reader, '.');
- }
- read_ws_star(reader);
- }
-
- eat_byte_safe(reader, '}');
- read_ws_star(reader);
- if (peek_byte(reader) == '.') {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "graph followed by `.'\n");
- }
-
- return SERD_SUCCESS;
+ if (!eat_byte_check(reader, '{')) {
+ return SERD_ERR_BAD_SYNTAX;
+ }
+
+ read_ws_star(reader);
+ while (peek_byte(reader) != '}') {
+ bool ate_dot = false;
+ int s_type = 0;
+ ctx->subject = 0;
+ SerdStatus st = read_subject(reader, *ctx, &ctx->subject, &s_type);
+ if (st) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "bad subject\n");
+ }
+
+ if (read_triples(reader, *ctx, &ate_dot) && s_type != '[') {
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "missing predicate object list\n");
+ }
+
+ pop_node(reader, ctx->subject);
+ read_ws_star(reader);
+ if (peek_byte(reader) == '.') {
+ eat_byte_safe(reader, '.');
+ }
+ read_ws_star(reader);
+ }
+
+ eat_byte_safe(reader, '}');
+ read_ws_star(reader);
+ if (peek_byte(reader) == '.') {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "graph followed by `.'\n");
+ }
+
+ return SERD_SUCCESS;
}
static int
tokcmp(SerdReader* reader, Ref ref, const char* tok, size_t n)
{
- SerdNode* node = deref(reader, ref);
- if (!node || node->n_bytes != n) {
- return -1;
- }
- return serd_strncasecmp((const char*)node->buf, tok, n);
+ SerdNode* node = deref(reader, ref);
+ if (!node || node->n_bytes != n) {
+ return -1;
+ }
+
+ return serd_strncasecmp((const char*)node->buf, tok, n);
}
SerdStatus
read_n3_statement(SerdReader* reader)
{
- SerdStatementFlags flags = 0;
- ReadContext ctx = { 0, 0, 0, 0, 0, 0, &flags };
- bool ate_dot = false;
- int s_type = 0;
- SerdStatus st = SERD_SUCCESS;
- read_ws_star(reader);
- switch (peek_byte(reader)) {
- case '\0':
- eat_byte_safe(reader, '\0');
- return SERD_FAILURE;
- case EOF:
- return SERD_FAILURE;
- case '@':
- if (!fancy_syntax(reader)) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "syntax does not support directives\n");
- }
- TRY(st, read_directive(reader));
- read_ws_star(reader);
- break;
- case '{':
- if (reader->syntax == SERD_TRIG) {
- TRY(st, read_wrappedGraph(reader, &ctx));
- read_ws_star(reader);
- } else {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "syntax does not support graphs\n");
- }
- break;
- default:
- if ((st = read_subject(reader, ctx, &ctx.subject, &s_type)) >
- SERD_FAILURE) {
- return st;
- }
-
- if (!tokcmp(reader, ctx.subject, "base", 4)) {
- st = read_base(reader, true, false);
- } else if (!tokcmp(reader, ctx.subject, "prefix", 6)) {
- st = read_prefixID(reader, true, false);
- } else if (!tokcmp(reader, ctx.subject, "graph", 5)) {
- read_ws_star(reader);
- TRY(st, read_labelOrSubject(reader, &ctx.graph));
- read_ws_star(reader);
- TRY(st, read_wrappedGraph(reader, &ctx));
- pop_node(reader, ctx.graph);
- ctx.graph = 0;
- read_ws_star(reader);
- } else if (read_ws_star(reader) && peek_byte(reader) == '{') {
- if (s_type == '(' || (s_type == '[' && !*ctx.flags)) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "invalid graph name\n");
- }
- ctx.graph = ctx.subject;
- ctx.subject = 0;
- TRY(st, read_wrappedGraph(reader, &ctx));
- pop_node(reader, ctx.graph);
- read_ws_star(reader);
- } else if ((st = read_triples(reader, ctx, &ate_dot))) {
- if (st == SERD_FAILURE && s_type == '[') {
- return SERD_SUCCESS;
- }
-
- if (ate_dot) {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "unexpected end of statement\n");
- }
-
- return st > SERD_FAILURE ? st : SERD_ERR_BAD_SYNTAX;
- } else if (!ate_dot) {
- read_ws_star(reader);
- st = (eat_byte_check(reader, '.') == '.') ? SERD_SUCCESS
- : SERD_ERR_BAD_SYNTAX;
- }
- break;
- }
- return st;
+ SerdStatementFlags flags = 0;
+ ReadContext ctx = {0, 0, 0, 0, 0, 0, &flags};
+ bool ate_dot = false;
+ int s_type = 0;
+ SerdStatus st = SERD_SUCCESS;
+ read_ws_star(reader);
+ switch (peek_byte(reader)) {
+ case '\0':
+ eat_byte_safe(reader, '\0');
+ return SERD_FAILURE;
+ case EOF:
+ return SERD_FAILURE;
+ case '@':
+ if (!fancy_syntax(reader)) {
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "syntax does not support directives\n");
+ }
+ TRY(st, read_directive(reader));
+ read_ws_star(reader);
+ break;
+ case '{':
+ if (reader->syntax == SERD_TRIG) {
+ TRY(st, read_wrappedGraph(reader, &ctx));
+ read_ws_star(reader);
+ } else {
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "syntax does not support graphs\n");
+ }
+ break;
+ default:
+ if ((st = read_subject(reader, ctx, &ctx.subject, &s_type)) >
+ SERD_FAILURE) {
+ return st;
+ }
+
+ if (!tokcmp(reader, ctx.subject, "base", 4)) {
+ st = read_base(reader, true, false);
+ } else if (!tokcmp(reader, ctx.subject, "prefix", 6)) {
+ st = read_prefixID(reader, true, false);
+ } else if (!tokcmp(reader, ctx.subject, "graph", 5)) {
+ read_ws_star(reader);
+ TRY(st, read_labelOrSubject(reader, &ctx.graph));
+ read_ws_star(reader);
+ TRY(st, read_wrappedGraph(reader, &ctx));
+ pop_node(reader, ctx.graph);
+ ctx.graph = 0;
+ read_ws_star(reader);
+ } else if (read_ws_star(reader) && peek_byte(reader) == '{') {
+ if (s_type == '(' || (s_type == '[' && !*ctx.flags)) {
+ return r_err(reader, SERD_ERR_BAD_SYNTAX, "invalid graph name\n");
+ }
+ ctx.graph = ctx.subject;
+ ctx.subject = 0;
+ TRY(st, read_wrappedGraph(reader, &ctx));
+ pop_node(reader, ctx.graph);
+ read_ws_star(reader);
+ } else if ((st = read_triples(reader, ctx, &ate_dot))) {
+ if (st == SERD_FAILURE && s_type == '[') {
+ return SERD_SUCCESS;
+ }
+
+ if (ate_dot) {
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "unexpected end of statement\n");
+ }
+
+ return st > SERD_FAILURE ? st : SERD_ERR_BAD_SYNTAX;
+ } else if (!ate_dot) {
+ read_ws_star(reader);
+ st = (eat_byte_check(reader, '.') == '.') ? SERD_SUCCESS
+ : SERD_ERR_BAD_SYNTAX;
+ }
+ break;
+ }
+ return st;
}
static void
skip_until(SerdReader* reader, uint8_t byte)
{
- for (int c = 0; (c = peek_byte(reader)) && c != byte;) {
- eat_byte_safe(reader, c);
- }
+ for (int c = 0; (c = peek_byte(reader)) && c != byte;) {
+ eat_byte_safe(reader, c);
+ }
}
SerdStatus
read_turtleTrigDoc(SerdReader* reader)
{
- while (!reader->source.eof) {
- const SerdStatus st = read_n3_statement(reader);
- if (st > SERD_FAILURE) {
- if (reader->strict) {
- return st;
- }
- skip_until(reader, '\n');
- }
- }
- return SERD_SUCCESS;
+ while (!reader->source.eof) {
+ const SerdStatus st = read_n3_statement(reader);
+ if (st > SERD_FAILURE) {
+ if (reader->strict) {
+ return st;
+ }
+ skip_until(reader, '\n');
+ }
+ }
+
+ return SERD_SUCCESS;
}
SerdStatus
read_nquadsDoc(SerdReader* reader)
{
- SerdStatus st = SERD_SUCCESS;
- while (!reader->source.eof) {
- SerdStatementFlags flags = 0;
- ReadContext ctx = { 0, 0, 0, 0, 0, 0, &flags };
- bool ate_dot = false;
- int s_type = 0;
- read_ws_star(reader);
- if (peek_byte(reader) == EOF) {
- break;
- }
-
- if (peek_byte(reader) == '@') {
- return r_err(reader, SERD_ERR_BAD_SYNTAX,
- "syntax does not support directives\n");
- }
-
- // subject predicate object
- if ((st = read_subject(reader, ctx, &ctx.subject, &s_type)) ||
- !read_ws_star(reader) ||
- (st = read_IRIREF(reader, &ctx.predicate)) ||
- !read_ws_star(reader) ||
- (st = read_object(reader, &ctx, false, &ate_dot))) {
- return st;
- }
-
- if (!ate_dot) { // graphLabel?
- read_ws_star(reader);
- switch (peek_byte(reader)) {
- case '.':
- break;
- case '_':
- TRY(st, read_BLANK_NODE_LABEL(reader, &ctx.graph, &ate_dot));
- break;
- default:
- TRY(st, read_IRIREF(reader, &ctx.graph));
- }
-
- // Terminating '.'
- read_ws_star(reader);
- if (!eat_byte_check(reader, '.')) {
- return SERD_ERR_BAD_SYNTAX;
- }
- }
-
- TRY(st, emit_statement(reader, ctx, ctx.object, ctx.datatype, ctx.lang));
-
- pop_node(reader, ctx.graph);
- pop_node(reader, ctx.lang);
- pop_node(reader, ctx.datatype);
- pop_node(reader, ctx.object);
- }
- return SERD_SUCCESS;
+ SerdStatus st = SERD_SUCCESS;
+ while (!reader->source.eof) {
+ SerdStatementFlags flags = 0;
+ ReadContext ctx = {0, 0, 0, 0, 0, 0, &flags};
+ bool ate_dot = false;
+ int s_type = 0;
+ read_ws_star(reader);
+ if (peek_byte(reader) == EOF) {
+ break;
+ }
+
+ if (peek_byte(reader) == '@') {
+ return r_err(
+ reader, SERD_ERR_BAD_SYNTAX, "syntax does not support directives\n");
+ }
+
+ // subject predicate object
+ if ((st = read_subject(reader, ctx, &ctx.subject, &s_type)) ||
+ !read_ws_star(reader) || (st = read_IRIREF(reader, &ctx.predicate)) ||
+ !read_ws_star(reader) ||
+ (st = read_object(reader, &ctx, false, &ate_dot))) {
+ return st;
+ }
+
+ if (!ate_dot) { // graphLabel?
+ read_ws_star(reader);
+ switch (peek_byte(reader)) {
+ case '.':
+ break;
+ case '_':
+ TRY(st, read_BLANK_NODE_LABEL(reader, &ctx.graph, &ate_dot));
+ break;
+ default:
+ TRY(st, read_IRIREF(reader, &ctx.graph));
+ }
+
+ // Terminating '.'
+ read_ws_star(reader);
+ if (!eat_byte_check(reader, '.')) {
+ return SERD_ERR_BAD_SYNTAX;
+ }
+ }
+
+ TRY(st, emit_statement(reader, ctx, ctx.object, ctx.datatype, ctx.lang));
+
+ pop_node(reader, ctx.graph);
+ pop_node(reader, ctx.lang);
+ pop_node(reader, ctx.datatype);
+ pop_node(reader, ctx.object);
+ }
+ return SERD_SUCCESS;
}
diff --git a/src/node.c b/src/node.c
index ab153b2b..6169895d 100644
--- a/src/node.c
+++ b/src/node.c
@@ -31,93 +31,93 @@
#include <string.h>
#ifdef _WIN32
-# ifndef isnan
-# define isnan(x) _isnan(x)
-# endif
-# ifndef isinf
-# define isinf(x) (!_finite(x))
-# endif
+# ifndef isnan
+# define isnan(x) _isnan(x)
+# endif
+# ifndef isinf
+# define isinf(x) (!_finite(x))
+# endif
#endif
SerdNode
serd_node_from_string(SerdType type, const uint8_t* str)
{
- if (!str) {
- return SERD_NODE_NULL;
- }
-
- SerdNodeFlags flags = 0;
- size_t buf_n_bytes = 0;
- const size_t buf_n_chars = serd_strlen(str, &buf_n_bytes, &flags);
- SerdNode ret = {str, buf_n_bytes, buf_n_chars, flags, type};
- return ret;
+ if (!str) {
+ return SERD_NODE_NULL;
+ }
+
+ SerdNodeFlags flags = 0;
+ size_t buf_n_bytes = 0;
+ const size_t buf_n_chars = serd_strlen(str, &buf_n_bytes, &flags);
+ SerdNode ret = {str, buf_n_bytes, buf_n_chars, flags, type};
+ return ret;
}
SerdNode
serd_node_from_substring(SerdType type, const uint8_t* str, const size_t len)
{
- if (!str) {
- return SERD_NODE_NULL;
- }
-
- SerdNodeFlags flags = 0;
- size_t buf_n_bytes = 0;
- const size_t buf_n_chars = serd_substrlen(str, len, &buf_n_bytes, &flags);
- assert(buf_n_bytes <= len);
- SerdNode ret = { str, buf_n_bytes, buf_n_chars, flags, type };
- return ret;
+ if (!str) {
+ return SERD_NODE_NULL;
+ }
+
+ SerdNodeFlags flags = 0;
+ size_t buf_n_bytes = 0;
+ const size_t buf_n_chars = serd_substrlen(str, len, &buf_n_bytes, &flags);
+ assert(buf_n_bytes <= len);
+ SerdNode ret = {str, buf_n_bytes, buf_n_chars, flags, type};
+ return ret;
}
SerdNode
serd_node_copy(const SerdNode* node)
{
- if (!node || !node->buf) {
- return SERD_NODE_NULL;
- }
-
- SerdNode copy = *node;
- uint8_t* buf = (uint8_t*)malloc(copy.n_bytes + 1);
- memcpy(buf, node->buf, copy.n_bytes + 1);
- copy.buf = buf;
- return copy;
+ if (!node || !node->buf) {
+ return SERD_NODE_NULL;
+ }
+
+ SerdNode copy = *node;
+ uint8_t* buf = (uint8_t*)malloc(copy.n_bytes + 1);
+ memcpy(buf, node->buf, copy.n_bytes + 1);
+ copy.buf = buf;
+ return copy;
}
bool
serd_node_equals(const SerdNode* a, const SerdNode* b)
{
- return (a == b)
- || (a->type == b->type
- && a->n_bytes == b->n_bytes
- && a->n_chars == b->n_chars
- && ((a->buf == b->buf) || !memcmp((const char*)a->buf,
- (const char*)b->buf,
- a->n_bytes + 1)));
+ return (a == b) ||
+ (a->type == b->type && a->n_bytes == b->n_bytes &&
+ a->n_chars == b->n_chars &&
+ ((a->buf == b->buf) ||
+ !memcmp((const char*)a->buf, (const char*)b->buf, a->n_bytes + 1)));
}
static size_t
serd_uri_string_length(const SerdURI* uri)
{
- size_t len = uri->path_base.len;
+ size_t len = uri->path_base.len;
-#define ADD_LEN(field, n_delims) \
- if ((field).len) { len += (field).len + (n_delims); }
+#define ADD_LEN(field, n_delims) \
+ if ((field).len) { \
+ len += (field).len + (n_delims); \
+ }
- ADD_LEN(uri->path, 1) // + possible leading `/'
- ADD_LEN(uri->scheme, 1) // + trailing `:'
- ADD_LEN(uri->authority, 2) // + leading `//'
- ADD_LEN(uri->query, 1) // + leading `?'
- ADD_LEN(uri->fragment, 1) // + leading `#'
+ ADD_LEN(uri->path, 1) // + possible leading `/'
+ ADD_LEN(uri->scheme, 1) // + trailing `:'
+ ADD_LEN(uri->authority, 2) // + leading `//'
+ ADD_LEN(uri->query, 1) // + leading `?'
+ ADD_LEN(uri->fragment, 1) // + leading `#'
- return len + 2; // + 2 for authority `//'
+ return len + 2; // + 2 for authority `//'
}
static size_t
string_sink(const void* buf, size_t len, void* stream)
{
- uint8_t** ptr = (uint8_t**)stream;
- memcpy(*ptr, buf, len);
- *ptr += len;
- return len;
+ uint8_t** ptr = (uint8_t**)stream;
+ memcpy(*ptr, buf, len);
+ *ptr += len;
+ return len;
}
SerdNode
@@ -125,9 +125,9 @@ serd_node_new_uri_from_node(const SerdNode* uri_node,
const SerdURI* base,
SerdURI* out)
{
- return (uri_node->type == SERD_URI && uri_node->buf)
- ? serd_node_new_uri_from_string(uri_node->buf, base, out)
- : SERD_NODE_NULL;
+ return (uri_node->type == SERD_URI && uri_node->buf)
+ ? serd_node_new_uri_from_string(uri_node->buf, base, out)
+ : SERD_NODE_NULL;
}
SerdNode
@@ -135,33 +135,50 @@ serd_node_new_uri_from_string(const uint8_t* str,
const SerdURI* base,
SerdURI* out)
{
- if (!str || str[0] == '\0') {
- // Empty URI => Base URI, or nothing if no base is given
- return base ? serd_node_new_uri(base, NULL, out) : SERD_NODE_NULL;
- }
-
- SerdURI uri;
- serd_uri_parse(str, &uri);
- return serd_node_new_uri(&uri, base, out); // Resolve/Serialise
+ if (!str || str[0] == '\0') {
+ // Empty URI => Base URI, or nothing if no base is given
+ return base ? serd_node_new_uri(base, NULL, out) : SERD_NODE_NULL;
+ }
+
+ SerdURI uri;
+ serd_uri_parse(str, &uri);
+ return serd_node_new_uri(&uri, base, out); // Resolve/Serialise
}
static inline bool
is_uri_path_char(const uint8_t c)
{
- if (is_alpha(c) || is_digit(c)) {
- return true;
- }
- switch (c) {
- case '-': case '.': case '_': case '~': // unreserved
- case ':': case '@': // pchar
- case '/': // separator
- // sub-delims
- case '!': case '$': case '&': case '\'': case '(': case ')':
- case '*': case '+': case ',': case ';': case '=':
- return true;
- default:
- return false;
- }
+ if (is_alpha(c) || is_digit(c)) {
+ return true;
+ }
+
+ switch (c) {
+ // unreserved:
+ case '-':
+ case '.':
+ case '_':
+ case '~':
+ // pchar:
+ case ':':
+ case '@':
+ // separator:
+ case '/':
+ // sub-delimeters:
+ case '!':
+ case '$':
+ case '&':
+ case '\'':
+ case '(':
+ case ')':
+ case '*':
+ case '+':
+ case ',':
+ case ';':
+ case '=':
+ return true;
+ default:
+ return false;
+ }
}
SerdNode
@@ -170,69 +187,69 @@ serd_node_new_file_uri(const uint8_t* path,
SerdURI* out,
bool escape)
{
- const size_t path_len = strlen((const char*)path);
- const size_t hostname_len = hostname ? strlen((const char*)hostname) : 0;
- const bool evil = is_windows_path(path);
- size_t uri_len = 0;
- uint8_t* uri = NULL;
-
- if (path[0] == '/' || is_windows_path(path)) {
- uri_len = strlen("file://") + hostname_len + evil;
- uri = (uint8_t*)malloc(uri_len + 1);
- snprintf((char*)uri, uri_len + 1, "file://%s%s",
- hostname ? (const char*)hostname : "",
- evil ? "/" : "");
- }
-
- SerdChunk chunk = { uri, uri_len };
- for (size_t i = 0; i < path_len; ++i) {
- if (evil && path[i] == '\\') {
- serd_chunk_sink("/", 1, &chunk);
- } else if (path[i] == '%') {
- serd_chunk_sink("%%", 2, &chunk);
- } else if (!escape || is_uri_path_char(path[i])) {
- serd_chunk_sink(path + i, 1, &chunk);
- } else {
- char escape_str[4] = { '%', 0, 0, 0 };
- snprintf(escape_str + 1,
- sizeof(escape_str) - 1,
- "%X",
- (unsigned)path[i]);
- serd_chunk_sink(escape_str, 3, &chunk);
- }
- }
- serd_chunk_sink_finish(&chunk);
-
- if (out) {
- serd_uri_parse(chunk.buf, out);
- }
-
- return serd_node_from_substring(SERD_URI, chunk.buf, chunk.len);
+ const size_t path_len = strlen((const char*)path);
+ const size_t hostname_len = hostname ? strlen((const char*)hostname) : 0;
+ const bool evil = is_windows_path(path);
+ size_t uri_len = 0;
+ uint8_t* uri = NULL;
+
+ if (path[0] == '/' || is_windows_path(path)) {
+ uri_len = strlen("file://") + hostname_len + evil;
+ uri = (uint8_t*)malloc(uri_len + 1);
+ snprintf((char*)uri,
+ uri_len + 1,
+ "file://%s%s",
+ hostname ? (const char*)hostname : "",
+ evil ? "/" : "");
+ }
+
+ SerdChunk chunk = {uri, uri_len};
+ for (size_t i = 0; i < path_len; ++i) {
+ if (evil && path[i] == '\\') {
+ serd_chunk_sink("/", 1, &chunk);
+ } else if (path[i] == '%') {
+ serd_chunk_sink("%%", 2, &chunk);
+ } else if (!escape || is_uri_path_char(path[i])) {
+ serd_chunk_sink(path + i, 1, &chunk);
+ } else {
+ char escape_str[4] = {'%', 0, 0, 0};
+ snprintf(escape_str + 1, sizeof(escape_str) - 1, "%X", (unsigned)path[i]);
+ serd_chunk_sink(escape_str, 3, &chunk);
+ }
+ }
+
+ serd_chunk_sink_finish(&chunk);
+
+ if (out) {
+ serd_uri_parse(chunk.buf, out);
+ }
+
+ return serd_node_from_substring(SERD_URI, chunk.buf, chunk.len);
}
SerdNode
serd_node_new_uri(const SerdURI* uri, const SerdURI* base, SerdURI* out)
{
- SerdURI abs_uri = *uri;
- if (base) {
- serd_uri_resolve(uri, base, &abs_uri);
- }
-
- const size_t len = serd_uri_string_length(&abs_uri);
- uint8_t* buf = (uint8_t*)malloc(len + 1);
- SerdNode node = { buf, 0, 0, 0, SERD_URI };
- uint8_t* ptr = buf;
- const size_t actual_len = serd_uri_serialise(&abs_uri, string_sink, &ptr);
-
- buf[actual_len] = '\0';
- node.n_bytes = actual_len;
- node.n_chars = serd_strlen(buf, NULL, NULL);
-
- if (out) {
- serd_uri_parse(buf, out); // TODO: cleverly avoid double parse
- }
-
- return node;
+ SerdURI abs_uri = *uri;
+ if (base) {
+ serd_uri_resolve(uri, base, &abs_uri);
+ }
+
+ const size_t len = serd_uri_string_length(&abs_uri);
+ uint8_t* buf = (uint8_t*)malloc(len + 1);
+ SerdNode node = {buf, 0, 0, 0, SERD_URI};
+ uint8_t* ptr = buf;
+ const size_t actual_len = serd_uri_serialise(&abs_uri, string_sink, &ptr);
+
+ buf[actual_len] = '\0';
+ node.n_bytes = actual_len;
+ node.n_chars = serd_strlen(buf, NULL, NULL);
+
+ if (out) {
+ serd_uri_parse(buf, out); // TODO: cleverly avoid double parse
+ }
+
+ return node;
}
SerdNode
@@ -241,129 +258,131 @@ serd_node_new_relative_uri(const SerdURI* uri,
const SerdURI* root,
SerdURI* out)
{
- const size_t uri_len = serd_uri_string_length(uri);
- const size_t base_len = serd_uri_string_length(base);
- uint8_t* buf = (uint8_t*)malloc(uri_len + base_len + 1);
- SerdNode node = { buf, 0, 0, 0, SERD_URI };
- uint8_t* ptr = buf;
- const size_t actual_len = serd_uri_serialise_relative(
- uri, base, root, string_sink, &ptr);
-
- buf[actual_len] = '\0';
- node.n_bytes = actual_len;
- node.n_chars = serd_strlen(buf, NULL, NULL);
-
- if (out) {
- serd_uri_parse(buf, out); // TODO: cleverly avoid double parse
- }
-
- return node;
+ const size_t uri_len = serd_uri_string_length(uri);
+ const size_t base_len = serd_uri_string_length(base);
+ uint8_t* buf = (uint8_t*)malloc(uri_len + base_len + 1);
+ SerdNode node = {buf, 0, 0, 0, SERD_URI};
+ uint8_t* ptr = buf;
+ const size_t actual_len =
+ serd_uri_serialise_relative(uri, base, root, string_sink, &ptr);
+
+ buf[actual_len] = '\0';
+ node.n_bytes = actual_len;
+ node.n_chars = serd_strlen(buf, NULL, NULL);
+
+ if (out) {
+ serd_uri_parse(buf, out); // TODO: cleverly avoid double parse
+ }
+
+ return node;
}
static inline unsigned
serd_digits(double abs)
{
- const double lg = ceil(log10(floor(abs) + 1.0));
- return lg < 1.0 ? 1U : (unsigned)lg;
+ const double lg = ceil(log10(floor(abs) + 1.0));
+ return lg < 1.0 ? 1U : (unsigned)lg;
}
SerdNode
serd_node_new_decimal(double d, unsigned frac_digits)
{
- if (isnan(d) || isinf(d)) {
- return SERD_NODE_NULL;
- }
-
- const double abs_d = fabs(d);
- const unsigned int_digits = serd_digits(abs_d);
- char* buf = (char*)calloc(int_digits + frac_digits + 3, 1);
- SerdNode node = { (const uint8_t*)buf, 0, 0, 0, SERD_LITERAL };
- const double int_part = floor(abs_d);
-
- // Point s to decimal point location
- char* s = buf + int_digits;
- if (d < 0.0) {
- *buf = '-';
- ++s;
- }
-
- // Write integer part (right to left)
- char* t = s - 1;
- uint64_t dec = (uint64_t)int_part;
- do {
- *t-- = (char)('0' + dec % 10);
- } while ((dec /= 10) > 0);
-
- *s++ = '.';
-
- // Write fractional part (right to left)
- double frac_part = fabs(d - int_part);
- if (frac_part < DBL_EPSILON) {
- *s++ = '0';
- node.n_bytes = node.n_chars = (size_t)(s - buf);
- } else {
- uint64_t frac = (uint64_t)llround(frac_part * pow(10.0, (int)frac_digits));
- s += frac_digits - 1;
- unsigned i = 0;
-
- // Skip trailing zeros
- for (; i < frac_digits - 1 && !(frac % 10); ++i, --s, frac /= 10) {}
-
- node.n_bytes = node.n_chars = (size_t)(s - buf) + 1u;
-
- // Write digits from last trailing zero to decimal point
- for (; i < frac_digits; ++i) {
- *s-- = (char)('0' + (frac % 10));
- frac /= 10;
- }
- }
-
- return node;
+ if (isnan(d) || isinf(d)) {
+ return SERD_NODE_NULL;
+ }
+
+ const double abs_d = fabs(d);
+ const unsigned int_digits = serd_digits(abs_d);
+ char* buf = (char*)calloc(int_digits + frac_digits + 3, 1);
+ SerdNode node = {(const uint8_t*)buf, 0, 0, 0, SERD_LITERAL};
+ const double int_part = floor(abs_d);
+
+ // Point s to decimal point location
+ char* s = buf + int_digits;
+ if (d < 0.0) {
+ *buf = '-';
+ ++s;
+ }
+
+ // Write integer part (right to left)
+ char* t = s - 1;
+ uint64_t dec = (uint64_t)int_part;
+ do {
+ *t-- = (char)('0' + dec % 10);
+ } while ((dec /= 10) > 0);
+
+ *s++ = '.';
+
+ // Write fractional part (right to left)
+ double frac_part = fabs(d - int_part);
+ if (frac_part < DBL_EPSILON) {
+ *s++ = '0';
+ node.n_bytes = node.n_chars = (size_t)(s - buf);
+ } else {
+ uint64_t frac = (uint64_t)llround(frac_part * pow(10.0, (int)frac_digits));
+ s += frac_digits - 1;
+ unsigned i = 0;
+
+ // Skip trailing zeros
+ for (; i < frac_digits - 1 && !(frac % 10); ++i, --s, frac /= 10) {
+ }
+
+ node.n_bytes = node.n_chars = (size_t)(s - buf) + 1u;
+
+ // Write digits from last trailing zero to decimal point
+ for (; i < frac_digits; ++i) {
+ *s-- = (char)('0' + (frac % 10));
+ frac /= 10;
+ }
+ }
+
+ return node;
}
SerdNode
serd_node_new_integer(int64_t i)
{
- int64_t abs_i = (i < 0) ? -i : i;
- const unsigned digits = serd_digits((double)abs_i);
- char* buf = (char*)calloc(digits + 2, 1);
- SerdNode node = { (const uint8_t*)buf, 0, 0, 0, SERD_LITERAL };
-
- // Point s to the end
- char* s = buf + digits - 1;
- if (i < 0) {
- *buf = '-';
- ++s;
- }
-
- node.n_bytes = node.n_chars = (size_t)(s - buf) + 1u;
-
- // Write integer part (right to left)
- do {
- *s-- = (char)('0' + (abs_i % 10));
- } while ((abs_i /= 10) > 0);
-
- return node;
+ int64_t abs_i = (i < 0) ? -i : i;
+ const unsigned digits = serd_digits((double)abs_i);
+ char* buf = (char*)calloc(digits + 2, 1);
+ SerdNode node = {(const uint8_t*)buf, 0, 0, 0, SERD_LITERAL};
+
+ // Point s to the end
+ char* s = buf + digits - 1;
+ if (i < 0) {
+ *buf = '-';
+ ++s;
+ }
+
+ node.n_bytes = node.n_chars = (size_t)(s - buf) + 1u;
+
+ // Write integer part (right to left)
+ do {
+ *s-- = (char)('0' + (abs_i % 10));
+ } while ((abs_i /= 10) > 0);
+
+ return node;
}
SerdNode
serd_node_new_blob(const void* buf, size_t size, bool wrap_lines)
{
- const size_t len = serd_base64_get_length(size, wrap_lines);
- uint8_t* str = (uint8_t*)calloc(len + 2, 1);
- SerdNode node = { str, len, len, 0, SERD_LITERAL };
-
- if (serd_base64_encode(str, buf, size, wrap_lines)) {
- node.flags |= SERD_HAS_NEWLINE;
- }
- return node;
+ const size_t len = serd_base64_get_length(size, wrap_lines);
+ uint8_t* str = (uint8_t*)calloc(len + 2, 1);
+ SerdNode node = {str, len, len, 0, SERD_LITERAL};
+
+ if (serd_base64_encode(str, buf, size, wrap_lines)) {
+ node.flags |= SERD_HAS_NEWLINE;
+ }
+
+ return node;
}
void
serd_node_free(SerdNode* node)
{
- if (node && node->buf) {
- free((uint8_t*)node->buf);
- node->buf = NULL;
- }
+ if (node && node->buf) {
+ free((uint8_t*)node->buf);
+ node->buf = NULL;
+ }
}
diff --git a/src/node.h b/src/node.h
index f99da1a6..e380f038 100644
--- a/src/node.h
+++ b/src/node.h
@@ -22,24 +22,27 @@
#include <stddef.h>
struct SerdNodeImpl {
- size_t n_bytes; /**< Size in bytes (not including null) */
- SerdNodeFlags flags; /**< Node flags (e.g. string properties) */
- SerdType type; /**< Node type */
+ size_t n_bytes; /**< Size in bytes (not including null) */
+ SerdNodeFlags flags; /**< Node flags (e.g. string properties) */
+ SerdType type; /**< Node type */
};
static inline char*
serd_node_buffer(SerdNode* node)
{
- return (char*)(node + 1);
+ return (char*)(node + 1);
}
static inline const char*
serd_node_buffer_c(const SerdNode* node)
{
- return (const char*)(node + 1);
+ return (const char*)(node + 1);
}
-SerdNode* serd_node_malloc(size_t n_bytes, SerdNodeFlags flags, SerdType type);
-void serd_node_set(SerdNode** dst, const SerdNode* src);
+SerdNode*
+serd_node_malloc(size_t n_bytes, SerdNodeFlags flags, SerdType type);
-#endif // SERD_NODE_H
+void
+serd_node_set(SerdNode** dst, const SerdNode* src);
+
+#endif // SERD_NODE_H
diff --git a/src/reader.c b/src/reader.c
index 2b802b5f..d0e0fbe2 100644
--- a/src/reader.c
+++ b/src/reader.c
@@ -29,181 +29,183 @@
SerdStatus
r_err(SerdReader* reader, SerdStatus st, const char* fmt, ...)
{
- va_list args;
- va_start(args, fmt);
- const Cursor* const cur = &reader->source.cur;
- const SerdError e = { st, cur->filename, cur->line, cur->col, fmt, &args };
- serd_error(reader->error_sink, reader->error_handle, &e);
- va_end(args);
- return st;
+ va_list args;
+ va_start(args, fmt);
+ const Cursor* const cur = &reader->source.cur;
+ const SerdError e = {st, cur->filename, cur->line, cur->col, fmt, &args};
+ serd_error(reader->error_sink, reader->error_handle, &e);
+ va_end(args);
+ return st;
}
void
set_blank_id(SerdReader* reader, Ref ref, size_t buf_size)
{
- SerdNode* node = deref(reader, ref);
- const char* prefix = reader->bprefix ? (const char*)reader->bprefix : "";
- node->n_bytes = node->n_chars = (size_t)snprintf(
- (char*)node->buf, buf_size, "%sb%u", prefix, reader->next_id++);
+ SerdNode* node = deref(reader, ref);
+ const char* prefix = reader->bprefix ? (const char*)reader->bprefix : "";
+ node->n_bytes = node->n_chars = (size_t)snprintf(
+ (char*)node->buf, buf_size, "%sb%u", prefix, reader->next_id++);
}
size_t
genid_size(SerdReader* reader)
{
- return reader->bprefix_len + 1 + 10 + 1; // + "b" + UINT32_MAX + \0
+ return reader->bprefix_len + 1 + 10 + 1; // + "b" + UINT32_MAX + \0
}
Ref
blank_id(SerdReader* reader)
{
- Ref ref = push_node_padded(reader, genid_size(reader), SERD_BLANK, "", 0);
- set_blank_id(reader, ref, genid_size(reader));
- return ref;
+ Ref ref = push_node_padded(reader, genid_size(reader), SERD_BLANK, "", 0);
+ set_blank_id(reader, ref, genid_size(reader));
+ return ref;
}
/** fread-like wrapper for getc (which is faster). */
static size_t
serd_file_read_byte(void* buf, size_t size, size_t nmemb, void* stream)
{
- (void)size;
- (void)nmemb;
-
- const int c = getc((FILE*)stream);
- if (c == EOF) {
- *((uint8_t*)buf) = 0;
- return 0;
- }
- *((uint8_t*)buf) = (uint8_t)c;
- return 1;
+ (void)size;
+ (void)nmemb;
+
+ const int c = getc((FILE*)stream);
+ if (c == EOF) {
+ *((uint8_t*)buf) = 0;
+ return 0;
+ }
+ *((uint8_t*)buf) = (uint8_t)c;
+ return 1;
}
Ref
-push_node_padded(SerdReader* reader, size_t maxlen,
- SerdType type, const char* str, size_t n_bytes)
+push_node_padded(SerdReader* reader,
+ size_t maxlen,
+ SerdType type,
+ const char* str,
+ size_t n_bytes)
{
- void* mem = serd_stack_push_aligned(
- &reader->stack, sizeof(SerdNode) + maxlen + 1, sizeof(SerdNode));
+ void* mem = serd_stack_push_aligned(
+ &reader->stack, sizeof(SerdNode) + maxlen + 1, sizeof(SerdNode));
- SerdNode* const node = (SerdNode*)mem;
- node->n_bytes = node->n_chars = n_bytes;
- node->flags = 0;
- node->type = type;
- node->buf = NULL;
+ SerdNode* const node = (SerdNode*)mem;
+ node->n_bytes = node->n_chars = n_bytes;
+ node->flags = 0;
+ node->type = type;
+ node->buf = NULL;
- uint8_t* buf = (uint8_t*)(node + 1);
- memcpy(buf, str, n_bytes + 1);
+ uint8_t* buf = (uint8_t*)(node + 1);
+ memcpy(buf, str, n_bytes + 1);
#ifdef SERD_STACK_CHECK
- reader->allocs = (Ref*)realloc(
- reader->allocs, sizeof(reader->allocs) * (++reader->n_allocs));
- reader->allocs[reader->n_allocs - 1] = ((uint8_t*)mem - reader->stack.buf);
+ reader->allocs = (Ref*)realloc(reader->allocs,
+ sizeof(reader->allocs) * (++reader->n_allocs));
+ reader->allocs[reader->n_allocs - 1] = ((uint8_t*)mem - reader->stack.buf);
#endif
- return (Ref)((uint8_t*)node - reader->stack.buf);
+ return (Ref)((uint8_t*)node - reader->stack.buf);
}
Ref
push_node(SerdReader* reader, SerdType type, const char* str, size_t n_bytes)
{
- return push_node_padded(reader, n_bytes, type, str, n_bytes);
+ return push_node_padded(reader, n_bytes, type, str, n_bytes);
}
SerdNode*
deref(SerdReader* reader, const Ref ref)
{
- if (ref) {
- SerdNode* node = (SerdNode*)(reader->stack.buf + ref);
- node->buf = (uint8_t*)node + sizeof(SerdNode);
- return node;
- }
- return NULL;
+ if (ref) {
+ SerdNode* node = (SerdNode*)(reader->stack.buf + ref);
+ node->buf = (uint8_t*)node + sizeof(SerdNode);
+ return node;
+ }
+ return NULL;
}
Ref
pop_node(SerdReader* reader, Ref ref)
{
- if (ref && ref != reader->rdf_first && ref != reader->rdf_rest
- && ref != reader->rdf_nil) {
+ if (ref && ref != reader->rdf_first && ref != reader->rdf_rest &&
+ ref != reader->rdf_nil) {
#ifdef SERD_STACK_CHECK
- SERD_STACK_ASSERT_TOP(reader, ref);
- --reader->n_allocs;
+ SERD_STACK_ASSERT_TOP(reader, ref);
+ --reader->n_allocs;
#endif
- SerdNode* const node = deref(reader, ref);
- uint8_t* const top = reader->stack.buf + reader->stack.size;
- serd_stack_pop_aligned(&reader->stack, (size_t)(top - (uint8_t*)node));
- }
- return 0;
+ SerdNode* const node = deref(reader, ref);
+ uint8_t* const top = reader->stack.buf + reader->stack.size;
+ serd_stack_pop_aligned(&reader->stack, (size_t)(top - (uint8_t*)node));
+ }
+ return 0;
}
SerdStatus
emit_statement(SerdReader* reader, ReadContext ctx, Ref o, Ref d, Ref l)
{
- SerdNode* graph = deref(reader, ctx.graph);
- if (!graph && reader->default_graph.buf) {
- graph = &reader->default_graph;
- }
-
- const SerdStatus st =
- !reader->statement_sink
- ? SERD_SUCCESS
- : reader->statement_sink(reader->handle,
- *ctx.flags,
- graph,
- deref(reader, ctx.subject),
- deref(reader, ctx.predicate),
- deref(reader, o),
- deref(reader, d),
- deref(reader, l));
-
- *ctx.flags &= SERD_ANON_CONT|SERD_LIST_CONT; // Preserve only cont flags
- return st;
+ SerdNode* graph = deref(reader, ctx.graph);
+ if (!graph && reader->default_graph.buf) {
+ graph = &reader->default_graph;
+ }
+
+ const SerdStatus st = !reader->statement_sink
+ ? SERD_SUCCESS
+ : reader->statement_sink(reader->handle,
+ *ctx.flags,
+ graph,
+ deref(reader, ctx.subject),
+ deref(reader, ctx.predicate),
+ deref(reader, o),
+ deref(reader, d),
+ deref(reader, l));
+
+ *ctx.flags &= SERD_ANON_CONT | SERD_LIST_CONT; // Preserve only cont flags
+ return st;
}
static SerdStatus
read_statement(SerdReader* reader)
{
- return read_n3_statement(reader);
+ return read_n3_statement(reader);
}
static SerdStatus
read_doc(SerdReader* reader)
{
- return ((reader->syntax == SERD_NQUADS) ? read_nquadsDoc(reader)
- : read_turtleTrigDoc(reader));
+ return ((reader->syntax == SERD_NQUADS) ? read_nquadsDoc(reader)
+ : read_turtleTrigDoc(reader));
}
SerdReader*
-serd_reader_new(SerdSyntax syntax,
- void* handle,
- void (*free_handle)(void*),
+serd_reader_new(SerdSyntax syntax,
+ void* handle,
+ void (*free_handle)(void*),
SerdBaseSink base_sink,
SerdPrefixSink prefix_sink,
SerdStatementSink statement_sink,
SerdEndSink end_sink)
{
- SerdReader* me = (SerdReader*)calloc(1, sizeof(SerdReader));
- me->handle = handle;
- me->free_handle = free_handle;
- me->base_sink = base_sink;
- me->prefix_sink = prefix_sink;
- me->statement_sink = statement_sink;
- me->end_sink = end_sink;
- me->default_graph = SERD_NODE_NULL;
- me->stack = serd_stack_new(SERD_PAGE_SIZE);
- me->syntax = syntax;
- me->next_id = 1;
- me->strict = true;
-
- me->rdf_first = push_node(me, SERD_URI, NS_RDF "first", 48);
- me->rdf_rest = push_node(me, SERD_URI, NS_RDF "rest", 47);
- me->rdf_nil = push_node(me, SERD_URI, NS_RDF "nil", 46);
-
- return me;
+ SerdReader* me = (SerdReader*)calloc(1, sizeof(SerdReader));
+ me->handle = handle;
+ me->free_handle = free_handle;
+ me->base_sink = base_sink;
+ me->prefix_sink = prefix_sink;
+ me->statement_sink = statement_sink;
+ me->end_sink = end_sink;
+ me->default_graph = SERD_NODE_NULL;
+ me->stack = serd_stack_new(SERD_PAGE_SIZE);
+ me->syntax = syntax;
+ me->next_id = 1;
+ me->strict = true;
+
+ me->rdf_first = push_node(me, SERD_URI, NS_RDF "first", 48);
+ me->rdf_rest = push_node(me, SERD_URI, NS_RDF "rest", 47);
+ me->rdf_nil = push_node(me, SERD_URI, NS_RDF "nil", 46);
+
+ return me;
}
void
serd_reader_set_strict(SerdReader* reader, bool strict)
{
- reader->strict = strict;
+ reader->strict = strict;
}
void
@@ -211,98 +213,96 @@ serd_reader_set_error_sink(SerdReader* reader,
SerdErrorSink error_sink,
void* error_handle)
{
- reader->error_sink = error_sink;
- reader->error_handle = error_handle;
+ reader->error_sink = error_sink;
+ reader->error_handle = error_handle;
}
void
serd_reader_free(SerdReader* reader)
{
- if (!reader) {
- return;
- }
+ if (!reader) {
+ return;
+ }
- pop_node(reader, reader->rdf_nil);
- pop_node(reader, reader->rdf_rest);
- pop_node(reader, reader->rdf_first);
- serd_node_free(&reader->default_graph);
+ pop_node(reader, reader->rdf_nil);
+ pop_node(reader, reader->rdf_rest);
+ pop_node(reader, reader->rdf_first);
+ serd_node_free(&reader->default_graph);
#ifdef SERD_STACK_CHECK
- free(reader->allocs);
+ free(reader->allocs);
#endif
- free(reader->stack.buf);
- free(reader->bprefix);
- if (reader->free_handle) {
- reader->free_handle(reader->handle);
- }
- free(reader);
+ free(reader->stack.buf);
+ free(reader->bprefix);
+ if (reader->free_handle) {
+ reader->free_handle(reader->handle);
+ }
+ free(reader);
}
void*
serd_reader_get_handle(const SerdReader* reader)
{
- return reader->handle;
+ return reader->handle;
}
void
serd_reader_add_blank_prefix(SerdReader* reader, const uint8_t* prefix)
{
- free(reader->bprefix);
- reader->bprefix_len = 0;
- reader->bprefix = NULL;
-
- const size_t prefix_len = prefix ? strlen((const char*)prefix) : 0;
- if (prefix_len) {
- reader->bprefix_len = prefix_len;
- reader->bprefix = (uint8_t*)malloc(reader->bprefix_len + 1);
- memcpy(reader->bprefix, prefix, reader->bprefix_len + 1);
- }
+ free(reader->bprefix);
+ reader->bprefix_len = 0;
+ reader->bprefix = NULL;
+
+ const size_t prefix_len = prefix ? strlen((const char*)prefix) : 0;
+ if (prefix_len) {
+ reader->bprefix_len = prefix_len;
+ reader->bprefix = (uint8_t*)malloc(reader->bprefix_len + 1);
+ memcpy(reader->bprefix, prefix, reader->bprefix_len + 1);
+ }
}
void
-serd_reader_set_default_graph(SerdReader* reader,
- const SerdNode* graph)
+serd_reader_set_default_graph(SerdReader* reader, const SerdNode* graph)
{
- serd_node_free(&reader->default_graph);
- reader->default_graph = serd_node_copy(graph);
+ serd_node_free(&reader->default_graph);
+ reader->default_graph = serd_node_copy(graph);
}
SerdStatus
-serd_reader_read_file(SerdReader* reader,
- const uint8_t* uri)
+serd_reader_read_file(SerdReader* reader, const uint8_t* uri)
{
- uint8_t* const path = serd_file_uri_parse(uri, NULL);
- if (!path) {
- return SERD_ERR_BAD_ARG;
- }
-
- FILE* fd = serd_fopen((const char*)path, "rb");
- if (!fd) {
- serd_free(path);
- return SERD_ERR_UNKNOWN;
- }
-
- SerdStatus ret = serd_reader_read_file_handle(reader, fd, path);
- fclose(fd);
- free(path);
- return ret;
+ uint8_t* const path = serd_file_uri_parse(uri, NULL);
+ if (!path) {
+ return SERD_ERR_BAD_ARG;
+ }
+
+ FILE* fd = serd_fopen((const char*)path, "rb");
+ if (!fd) {
+ serd_free(path);
+ return SERD_ERR_UNKNOWN;
+ }
+
+ SerdStatus ret = serd_reader_read_file_handle(reader, fd, path);
+ fclose(fd);
+ free(path);
+ return ret;
}
static SerdStatus
skip_bom(SerdReader* me)
{
- if (serd_byte_source_peek(&me->source) == 0xEF) {
- serd_byte_source_advance(&me->source);
- if (serd_byte_source_peek(&me->source) != 0xBB ||
- serd_byte_source_advance(&me->source) ||
- serd_byte_source_peek(&me->source) != 0xBF ||
- serd_byte_source_advance(&me->source)) {
- r_err(me, SERD_ERR_BAD_SYNTAX, "corrupt byte order mark\n");
- return SERD_ERR_BAD_SYNTAX;
- }
- }
-
- return SERD_SUCCESS;
+ if (serd_byte_source_peek(&me->source) == 0xEF) {
+ serd_byte_source_advance(&me->source);
+ if (serd_byte_source_peek(&me->source) != 0xBB ||
+ serd_byte_source_advance(&me->source) ||
+ serd_byte_source_peek(&me->source) != 0xBF ||
+ serd_byte_source_advance(&me->source)) {
+ r_err(me, SERD_ERR_BAD_SYNTAX, "corrupt byte order mark\n");
+ return SERD_ERR_BAD_SYNTAX;
+ }
+ }
+
+ return SERD_SUCCESS;
}
SerdStatus
@@ -311,13 +311,13 @@ serd_reader_start_stream(SerdReader* reader,
const uint8_t* name,
bool bulk)
{
- return serd_reader_start_source_stream(
- reader,
- bulk ? (SerdSource)fread : serd_file_read_byte,
- (SerdStreamErrorFunc)ferror,
- file,
- name,
- bulk ? SERD_PAGE_SIZE : 1);
+ return serd_reader_start_source_stream(reader,
+ bulk ? (SerdSource)fread
+ : serd_file_read_byte,
+ (SerdStreamErrorFunc)ferror,
+ file,
+ name,
+ bulk ? SERD_PAGE_SIZE : 1);
}
SerdStatus
@@ -328,46 +328,46 @@ serd_reader_start_source_stream(SerdReader* reader,
const uint8_t* name,
size_t page_size)
{
- return serd_byte_source_open_source(
- &reader->source, read_func, error_func, stream, name, page_size);
+ return serd_byte_source_open_source(
+ &reader->source, read_func, error_func, stream, name, page_size);
}
static SerdStatus
serd_reader_prepare(SerdReader* reader)
{
- SerdStatus st = serd_byte_source_prepare(&reader->source);
- if (st == SERD_SUCCESS) {
- st = skip_bom(reader);
- } else if (st == SERD_FAILURE) {
- reader->source.eof = true;
- } else {
- r_err(reader, st, "read error: %s\n", strerror(errno));
- }
- return st;
+ SerdStatus st = serd_byte_source_prepare(&reader->source);
+ if (st == SERD_SUCCESS) {
+ st = skip_bom(reader);
+ } else if (st == SERD_FAILURE) {
+ reader->source.eof = true;
+ } else {
+ r_err(reader, st, "read error: %s\n", strerror(errno));
+ }
+ return st;
}
SerdStatus
serd_reader_read_chunk(SerdReader* reader)
{
- SerdStatus st = SERD_SUCCESS;
- if (!reader->source.prepared) {
- st = serd_reader_prepare(reader);
- } else if (reader->source.eof) {
- st = serd_byte_source_advance(&reader->source);
- }
-
- if (peek_byte(reader) == 0) {
- // Skip leading null byte, for reading from a null-delimited socket
- eat_byte_safe(reader, 0);
- }
-
- return st ? st : read_statement(reader);
+ SerdStatus st = SERD_SUCCESS;
+ if (!reader->source.prepared) {
+ st = serd_reader_prepare(reader);
+ } else if (reader->source.eof) {
+ st = serd_byte_source_advance(&reader->source);
+ }
+
+ if (peek_byte(reader) == 0) {
+ // Skip leading null byte, for reading from a null-delimited socket
+ eat_byte_safe(reader, 0);
+ }
+
+ return st ? st : read_statement(reader);
}
SerdStatus
serd_reader_end_stream(SerdReader* reader)
{
- return serd_byte_source_close(&reader->source);
+ return serd_byte_source_close(&reader->source);
}
SerdStatus
@@ -375,9 +375,12 @@ serd_reader_read_file_handle(SerdReader* reader,
FILE* file,
const uint8_t* name)
{
- return serd_reader_read_source(
- reader, (SerdSource)fread, (SerdStreamErrorFunc)ferror,
- file, name, SERD_PAGE_SIZE);
+ return serd_reader_read_source(reader,
+ (SerdSource)fread,
+ (SerdStreamErrorFunc)ferror,
+ file,
+ name,
+ SERD_PAGE_SIZE);
}
SerdStatus
@@ -388,33 +391,33 @@ serd_reader_read_source(SerdReader* reader,
const uint8_t* name,
size_t page_size)
{
- SerdStatus st = serd_reader_start_source_stream(
- reader, source, error, stream, name, page_size);
+ SerdStatus st = serd_reader_start_source_stream(
+ reader, source, error, stream, name, page_size);
- if (st || (st = serd_reader_prepare(reader))) {
- serd_reader_end_stream(reader);
- return st;
- }
+ if (st || (st = serd_reader_prepare(reader))) {
+ serd_reader_end_stream(reader);
+ return st;
+ }
- if ((st = read_doc(reader))) {
- serd_reader_end_stream(reader);
- return st;
- }
+ if ((st = read_doc(reader))) {
+ serd_reader_end_stream(reader);
+ return st;
+ }
- return serd_reader_end_stream(reader);
+ return serd_reader_end_stream(reader);
}
SerdStatus
serd_reader_read_string(SerdReader* reader, const uint8_t* utf8)
{
- serd_byte_source_open_string(&reader->source, utf8);
+ serd_byte_source_open_string(&reader->source, utf8);
- SerdStatus st = serd_reader_prepare(reader);
- if (!st) {
- st = read_doc(reader);
- }
+ SerdStatus st = serd_reader_prepare(reader);
+ if (!st) {
+ st = read_doc(reader);
+ }
- serd_byte_source_close(&reader->source);
+ serd_byte_source_close(&reader->source);
- return st;
+ return st;
}
diff --git a/src/reader.h b/src/reader.h
index 166e2137..85a2f109 100644
--- a/src/reader.h
+++ b/src/reader.h
@@ -28,16 +28,16 @@
#include <stdio.h>
#if defined(__GNUC__)
-# define SERD_LOG_FUNC(fmt, arg1) __attribute__((format(printf, fmt, arg1)))
+# define SERD_LOG_FUNC(fmt, arg1) __attribute__((format(printf, fmt, arg1)))
#else
-# define SERD_LOG_FUNC(fmt, arg1)
+# define SERD_LOG_FUNC(fmt, arg1)
#endif
#ifdef SERD_STACK_CHECK
-# define SERD_STACK_ASSERT_TOP(reader, ref) \
- assert(ref == reader->allocs[reader->n_allocs - 1]);
+# define SERD_STACK_ASSERT_TOP(reader, ref) \
+ assert(ref == reader->allocs[reader->n_allocs - 1]);
#else
-# define SERD_STACK_ASSERT_TOP(reader, ref)
+# define SERD_STACK_ASSERT_TOP(reader, ref)
#endif
/* Reference to a node in the stack (we can not use pointers since the
@@ -46,40 +46,40 @@
typedef size_t Ref;
typedef struct {
- Ref graph;
- Ref subject;
- Ref predicate;
- Ref object;
- Ref datatype;
- Ref lang;
- SerdStatementFlags* flags;
+ Ref graph;
+ Ref subject;
+ Ref predicate;
+ Ref object;
+ Ref datatype;
+ Ref lang;
+ SerdStatementFlags* flags;
} ReadContext;
struct SerdReaderImpl {
- void* handle;
- void (*free_handle)(void* ptr);
- SerdBaseSink base_sink;
- SerdPrefixSink prefix_sink;
- SerdStatementSink statement_sink;
- SerdEndSink end_sink;
- SerdErrorSink error_sink;
- void* error_handle;
- Ref rdf_first;
- Ref rdf_rest;
- Ref rdf_nil;
- SerdNode default_graph;
- SerdByteSource source;
- SerdStack stack;
- SerdSyntax syntax;
- unsigned next_id;
- uint8_t* buf;
- uint8_t* bprefix;
- size_t bprefix_len;
- bool strict; ///< True iff strict parsing
- bool seen_genid;
+ void* handle;
+ void (*free_handle)(void* ptr);
+ SerdBaseSink base_sink;
+ SerdPrefixSink prefix_sink;
+ SerdStatementSink statement_sink;
+ SerdEndSink end_sink;
+ SerdErrorSink error_sink;
+ void* error_handle;
+ Ref rdf_first;
+ Ref rdf_rest;
+ Ref rdf_nil;
+ SerdNode default_graph;
+ SerdByteSource source;
+ SerdStack stack;
+ SerdSyntax syntax;
+ unsigned next_id;
+ uint8_t* buf;
+ uint8_t* bprefix;
+ size_t bprefix_len;
+ bool strict; ///< True iff strict parsing
+ bool seen_genid;
#ifdef SERD_STACK_CHECK
- Ref* allocs; ///< Stack of push offsets
- size_t n_allocs; ///< Number of stack pushes
+ Ref* allocs; ///< Stack of push offsets
+ size_t n_allocs; ///< Number of stack pushes
#endif
};
@@ -87,98 +87,110 @@ SERD_LOG_FUNC(3, 4)
SerdStatus
r_err(SerdReader* reader, SerdStatus st, const char* fmt, ...);
-Ref push_node_padded(SerdReader* reader,
- size_t maxlen,
- SerdType type,
- const char* str,
- size_t n_bytes);
+Ref
+push_node_padded(SerdReader* reader,
+ size_t maxlen,
+ SerdType type,
+ const char* str,
+ size_t n_bytes);
-Ref push_node(SerdReader* reader,
- SerdType type,
- const char* str,
- size_t n_bytes);
+Ref
+push_node(SerdReader* reader, SerdType type, const char* str, size_t n_bytes);
-SERD_PURE_FUNC size_t genid_size(SerdReader* reader);
-Ref blank_id(SerdReader* reader);
-void set_blank_id(SerdReader* reader, Ref ref, size_t buf_size);
+SERD_PURE_FUNC size_t
+genid_size(SerdReader* reader);
-SerdNode* deref(SerdReader* reader, Ref ref);
+Ref
+blank_id(SerdReader* reader);
-Ref pop_node(SerdReader* reader, Ref ref);
+void
+set_blank_id(SerdReader* reader, Ref ref, size_t buf_size);
+
+SerdNode*
+deref(SerdReader* reader, Ref ref);
+
+Ref
+pop_node(SerdReader* reader, Ref ref);
SerdStatus
emit_statement(SerdReader* reader, ReadContext ctx, Ref o, Ref d, Ref l);
-SerdStatus read_n3_statement(SerdReader* reader);
-SerdStatus read_nquadsDoc(SerdReader* reader);
-SerdStatus read_turtleTrigDoc(SerdReader* reader);
+SerdStatus
+read_n3_statement(SerdReader* reader);
+
+SerdStatus
+read_nquadsDoc(SerdReader* reader);
+
+SerdStatus
+read_turtleTrigDoc(SerdReader* reader);
static inline int
peek_byte(SerdReader* reader)
{
- SerdByteSource* source = &reader->source;
+ SerdByteSource* source = &reader->source;
- return source->eof ? EOF : (int)source->read_buf[source->read_head];
+ return source->eof ? EOF : (int)source->read_buf[source->read_head];
}
static inline int
eat_byte_safe(SerdReader* reader, const int byte)
{
- (void)byte;
+ (void)byte;
- const int c = peek_byte(reader);
- assert(c == byte);
+ const int c = peek_byte(reader);
+ assert(c == byte);
- serd_byte_source_advance(&reader->source);
- return c;
+ serd_byte_source_advance(&reader->source);
+ return c;
}
static inline int
eat_byte_check(SerdReader* reader, const int byte)
{
- const int c = peek_byte(reader);
- if (c != byte) {
- r_err(reader, SERD_ERR_BAD_SYNTAX,
- "expected `%c', not `%c'\n", byte, c);
- return 0;
- }
- return eat_byte_safe(reader, byte);
+ const int c = peek_byte(reader);
+ if (c != byte) {
+ r_err(reader, SERD_ERR_BAD_SYNTAX, "expected `%c', not `%c'\n", byte, c);
+ return 0;
+ }
+ return eat_byte_safe(reader, byte);
}
static inline SerdStatus
eat_string(SerdReader* reader, const char* str, unsigned n)
{
- for (unsigned i = 0; i < n; ++i) {
- if (!eat_byte_check(reader, ((const uint8_t*)str)[i])) {
- return SERD_ERR_BAD_SYNTAX;
- }
- }
- return SERD_SUCCESS;
+ for (unsigned i = 0; i < n; ++i) {
+ if (!eat_byte_check(reader, ((const uint8_t*)str)[i])) {
+ return SERD_ERR_BAD_SYNTAX;
+ }
+ }
+ return SERD_SUCCESS;
}
static inline SerdStatus
push_byte(SerdReader* reader, Ref ref, const int c)
{
- assert(c != EOF);
- SERD_STACK_ASSERT_TOP(reader, ref);
-
- uint8_t* const s = (uint8_t*)serd_stack_push(&reader->stack, 1);
- SerdNode* const node = (SerdNode*)(reader->stack.buf + ref);
- ++node->n_bytes;
- if (!(c & 0x80)) { // Starts with 0 bit, start of new character
- ++node->n_chars;
- }
- *(s - 1) = (uint8_t)c;
- *s = '\0';
- return SERD_SUCCESS;
+ assert(c != EOF);
+ SERD_STACK_ASSERT_TOP(reader, ref);
+
+ uint8_t* const s = (uint8_t*)serd_stack_push(&reader->stack, 1);
+ SerdNode* const node = (SerdNode*)(reader->stack.buf + ref);
+
+ ++node->n_bytes;
+ if (!(c & 0x80)) { // Starts with 0 bit, start of new character
+ ++node->n_chars;
+ }
+
+ *(s - 1) = (uint8_t)c;
+ *s = '\0';
+ return SERD_SUCCESS;
}
static inline void
push_bytes(SerdReader* reader, Ref ref, const uint8_t* bytes, unsigned len)
{
- for (unsigned i = 0; i < len; ++i) {
- push_byte(reader, ref, bytes[i]);
- }
+ for (unsigned i = 0; i < len; ++i) {
+ push_byte(reader, ref, bytes[i]);
+ }
}
#endif // SERD_READER_H
diff --git a/src/serd_internal.h b/src/serd_internal.h
index 95c3b121..af9e7710 100644
--- a/src/serd_internal.h
+++ b/src/serd_internal.h
@@ -27,7 +27,7 @@
#define SERD_PAGE_SIZE 4096
#ifndef MIN
-# define MIN(a, b) (((a) < (b)) ? (a) : (b))
+# define MIN(a, b) (((a) < (b)) ? (a) : (b))
#endif
/* Error reporting */
@@ -35,12 +35,12 @@
static inline void
serd_error(SerdErrorSink error_sink, void* handle, const SerdError* e)
{
- if (error_sink) {
- error_sink(handle, e);
- } else {
- fprintf(stderr, "error: %s:%u:%u: ", e->filename, e->line, e->col);
- vfprintf(stderr, e->fmt, *e->args);
- }
+ if (error_sink) {
+ error_sink(handle, e);
+ } else {
+ fprintf(stderr, "error: %s:%u:%u: ", e->filename, e->line, e->col);
+ vfprintf(stderr, e->fmt, *e->args);
+ }
}
-#endif // SERD_INTERNAL_H
+#endif // SERD_INTERNAL_H
diff --git a/src/serdi.c b/src/serdi.c
index 69304cad..44644ff4 100644
--- a/src/serdi.c
+++ b/src/serdi.c
@@ -22,13 +22,13 @@
#include "serd/serd.h"
#ifdef _WIN32
-#define WIN32_LEAN_AND_MEAN 1
-#include <fcntl.h>
-#include <io.h>
+# define WIN32_LEAN_AND_MEAN 1
+# include <fcntl.h>
+# include <io.h>
#endif
#if defined(HAVE_POSIX_FADVISE) && defined(HAVE_FILENO)
-#include <fcntl.h>
+# include <fcntl.h>
#endif
#include <errno.h>
@@ -38,114 +38,114 @@
#include <stdlib.h>
#include <string.h>
-#define SERDI_ERROR(msg) fprintf(stderr, "serdi: " msg)
+#define SERDI_ERROR(msg) fprintf(stderr, "serdi: " msg)
#define SERDI_ERRORF(fmt, ...) fprintf(stderr, "serdi: " fmt, __VA_ARGS__)
typedef struct {
- SerdSyntax syntax;
- const char* name;
- const char* extension;
+ SerdSyntax syntax;
+ const char* name;
+ const char* extension;
} Syntax;
-static const Syntax syntaxes[] = {
- {SERD_TURTLE, "turtle", ".ttl"},
- {SERD_NTRIPLES, "ntriples", ".nt"},
- {SERD_NQUADS, "nquads", ".nq"},
- {SERD_TRIG, "trig", ".trig"},
- {(SerdSyntax)0, NULL, NULL}
-};
+static const Syntax syntaxes[] = {{SERD_TURTLE, "turtle", ".ttl"},
+ {SERD_NTRIPLES, "ntriples", ".nt"},
+ {SERD_NQUADS, "nquads", ".nq"},
+ {SERD_TRIG, "trig", ".trig"},
+ {(SerdSyntax)0, NULL, NULL}};
static SerdSyntax
get_syntax(const char* name)
{
- for (const Syntax* s = syntaxes; s->name; ++s) {
- if (!serd_strncasecmp(s->name, name, strlen(name))) {
- return s->syntax;
- }
- }
- SERDI_ERRORF("unknown syntax `%s'\n", name);
- return (SerdSyntax)0;
+ for (const Syntax* s = syntaxes; s->name; ++s) {
+ if (!serd_strncasecmp(s->name, name, strlen(name))) {
+ return s->syntax;
+ }
+ }
+
+ SERDI_ERRORF("unknown syntax `%s'\n", name);
+ return (SerdSyntax)0;
}
static SERD_PURE_FUNC SerdSyntax
guess_syntax(const char* filename)
{
- const char* ext = strrchr(filename, '.');
- if (ext) {
- for (const Syntax* s = syntaxes; s->name; ++s) {
- if (!serd_strncasecmp(s->extension, ext, strlen(ext))) {
- return s->syntax;
- }
- }
- }
- return (SerdSyntax)0;
+ const char* ext = strrchr(filename, '.');
+ if (ext) {
+ for (const Syntax* s = syntaxes; s->name; ++s) {
+ if (!serd_strncasecmp(s->extension, ext, strlen(ext))) {
+ return s->syntax;
+ }
+ }
+ }
+
+ return (SerdSyntax)0;
}
static int
print_version(void)
{
- printf("serdi " SERD_VERSION " <http://drobilla.net/software/serd>\n");
- printf("Copyright 2011-2020 David Robillard <http://drobilla.net>.\n"
- "License: <http://www.opensource.org/licenses/isc>\n"
- "This is free software; you are free to change and redistribute it."
- "\nThere is NO WARRANTY, to the extent permitted by law.\n");
- return 0;
+ printf("serdi " SERD_VERSION " <http://drobilla.net/software/serd>\n");
+ printf("Copyright 2011-2020 David Robillard <http://drobilla.net>.\n"
+ "License: <http://www.opensource.org/licenses/isc>\n"
+ "This is free software; you are free to change and redistribute it."
+ "\nThere is NO WARRANTY, to the extent permitted by law.\n");
+ return 0;
}
static int
print_usage(const char* name, bool error)
{
- FILE* const os = error ? stderr : stdout;
- fprintf(os, "%s", error ? "\n" : "");
- fprintf(os, "Usage: %s [OPTION]... INPUT [BASE_URI]\n", name);
- fprintf(os, "Read and write RDF syntax.\n");
- fprintf(os, "Use - for INPUT to read from standard input.\n\n");
- fprintf(os, " -a Write ASCII output if possible.\n");
- fprintf(os, " -b Fast bulk output for large serialisations.\n");
- fprintf(os, " -c PREFIX Chop PREFIX from matching blank node IDs.\n");
- fprintf(os, " -e Eat input one character at a time.\n");
- fprintf(os, " -f Keep full URIs in input (don't qualify).\n");
- fprintf(os, " -h Display this help and exit.\n");
- fprintf(os, " -i SYNTAX Input syntax: turtle/ntriples/trig/nquads.\n");
- fprintf(os, " -l Lax (non-strict) parsing.\n");
- fprintf(os, " -o SYNTAX Output syntax: turtle/ntriples/nquads.\n");
- fprintf(os, " -p PREFIX Add PREFIX to blank node IDs.\n");
- fprintf(os, " -q Suppress all output except data.\n");
- fprintf(os, " -r ROOT_URI Keep relative URIs within ROOT_URI.\n");
- fprintf(os, " -s INPUT Parse INPUT as string (terminates options).\n");
- fprintf(os, " -v Display version information and exit.\n");
- return error ? 1 : 0;
+ FILE* const os = error ? stderr : stdout;
+ fprintf(os, "%s", error ? "\n" : "");
+ fprintf(os, "Usage: %s [OPTION]... INPUT [BASE_URI]\n", name);
+ fprintf(os, "Read and write RDF syntax.\n");
+ fprintf(os, "Use - for INPUT to read from standard input.\n\n");
+ fprintf(os, " -a Write ASCII output if possible.\n");
+ fprintf(os, " -b Fast bulk output for large serialisations.\n");
+ fprintf(os, " -c PREFIX Chop PREFIX from matching blank node IDs.\n");
+ fprintf(os, " -e Eat input one character at a time.\n");
+ fprintf(os, " -f Keep full URIs in input (don't qualify).\n");
+ fprintf(os, " -h Display this help and exit.\n");
+ fprintf(os, " -i SYNTAX Input syntax: turtle/ntriples/trig/nquads.\n");
+ fprintf(os, " -l Lax (non-strict) parsing.\n");
+ fprintf(os, " -o SYNTAX Output syntax: turtle/ntriples/nquads.\n");
+ fprintf(os, " -p PREFIX Add PREFIX to blank node IDs.\n");
+ fprintf(os, " -q Suppress all output except data.\n");
+ fprintf(os, " -r ROOT_URI Keep relative URIs within ROOT_URI.\n");
+ fprintf(os, " -s INPUT Parse INPUT as string (terminates options).\n");
+ fprintf(os, " -v Display version information and exit.\n");
+ return error ? 1 : 0;
}
static int
missing_arg(const char* name, char opt)
{
- SERDI_ERRORF("option requires an argument -- '%c'\n", opt);
- return print_usage(name, true);
+ SERDI_ERRORF("option requires an argument -- '%c'\n", opt);
+ return print_usage(name, true);
}
static SerdStatus
quiet_error_sink(void* handle, const SerdError* e)
{
- (void)handle;
- (void)e;
- return SERD_SUCCESS;
+ (void)handle;
+ (void)e;
+ return SERD_SUCCESS;
}
static inline FILE*
serd_fopen(const char* path, const char* mode)
{
- FILE* fd = fopen(path, mode);
- if (!fd) {
- SERDI_ERRORF("failed to open file %s (%s)\n", path, strerror(errno));
- return NULL;
- }
+ FILE* fd = fopen(path, mode);
+ if (!fd) {
+ SERDI_ERRORF("failed to open file %s (%s)\n", path, strerror(errno));
+ return NULL;
+ }
#if defined(HAVE_POSIX_FADVISE) && defined(HAVE_FILENO)
- posix_fadvise(fileno(fd), 0, 0, POSIX_FADV_SEQUENTIAL|POSIX_FADV_NOREUSE);
+ posix_fadvise(fileno(fd), 0, 0, POSIX_FADV_SEQUENTIAL | POSIX_FADV_NOREUSE);
#endif
- return fd;
+ return fd;
}
static SerdStyle
@@ -155,219 +155,221 @@ choose_style(const SerdSyntax input_syntax,
const bool bulk_write,
const bool full_uris)
{
- unsigned output_style = 0u;
- if (output_syntax == SERD_NTRIPLES || ascii) {
- output_style |= SERD_STYLE_ASCII;
- } else if (output_syntax == SERD_TURTLE) {
- output_style |= SERD_STYLE_ABBREVIATED;
- if (!full_uris) {
- output_style |= SERD_STYLE_CURIED;
- }
- }
-
- if ((input_syntax == SERD_TURTLE || input_syntax == SERD_TRIG) ||
- (output_style & SERD_STYLE_CURIED)) {
- // Base URI may change and/or we're abbreviating URIs, so must resolve
- output_style |= SERD_STYLE_RESOLVED;
- }
-
- if (bulk_write) {
- output_style |= SERD_STYLE_BULK;
- }
-
- return (SerdStyle)output_style;
+ unsigned output_style = 0u;
+ if (output_syntax == SERD_NTRIPLES || ascii) {
+ output_style |= SERD_STYLE_ASCII;
+ } else if (output_syntax == SERD_TURTLE) {
+ output_style |= SERD_STYLE_ABBREVIATED;
+ if (!full_uris) {
+ output_style |= SERD_STYLE_CURIED;
+ }
+ }
+
+ if ((input_syntax == SERD_TURTLE || input_syntax == SERD_TRIG) ||
+ (output_style & SERD_STYLE_CURIED)) {
+ // Base URI may change and/or we're abbreviating URIs, so must resolve
+ output_style |= SERD_STYLE_RESOLVED;
+ }
+
+ if (bulk_write) {
+ output_style |= SERD_STYLE_BULK;
+ }
+
+ return (SerdStyle)output_style;
}
int
main(int argc, char** argv)
{
- if (argc < 2) {
- return print_usage(argv[0], true);
- }
-
- FILE* in_fd = NULL;
- SerdSyntax input_syntax = (SerdSyntax)0;
- SerdSyntax output_syntax = (SerdSyntax)0;
- bool from_file = true;
- bool ascii = false;
- bool bulk_read = true;
- bool bulk_write = false;
- bool full_uris = false;
- bool lax = false;
- bool quiet = false;
- const uint8_t* in_name = NULL;
- const uint8_t* add_prefix = NULL;
- const uint8_t* chop_prefix = NULL;
- const uint8_t* root_uri = NULL;
- int a = 1;
- for (; a < argc && argv[a][0] == '-'; ++a) {
- if (argv[a][1] == '\0') {
- in_name = (const uint8_t*)"(stdin)";
- in_fd = stdin;
- break;
- }
-
- if (argv[a][1] == 'a') {
- ascii = true;
- } else if (argv[a][1] == 'b') {
- bulk_write = true;
- } else if (argv[a][1] == 'e') {
- bulk_read = false;
- } else if (argv[a][1] == 'f') {
- full_uris = true;
- } else if (argv[a][1] == 'h') {
- return print_usage(argv[0], false);
- } else if (argv[a][1] == 'l') {
- lax = true;
- } else if (argv[a][1] == 'q') {
- quiet = true;
- } else if (argv[a][1] == 'v') {
- return print_version();
- } else if (argv[a][1] == 's') {
- in_name = (const uint8_t*)"(string)";
- from_file = false;
- ++a;
- break;
- } else if (argv[a][1] == 'i') {
- if (++a == argc) {
- return missing_arg(argv[0], 'i');
- }
-
- if (!(input_syntax = get_syntax(argv[a]))) {
- return print_usage(argv[0], true);
- }
- } else if (argv[a][1] == 'o') {
- if (++a == argc) {
- return missing_arg(argv[0], 'o');
- }
-
- if (!(output_syntax = get_syntax(argv[a]))) {
- return print_usage(argv[0], true);
- }
- } else if (argv[a][1] == 'p') {
- if (++a == argc) {
- return missing_arg(argv[0], 'p');
- }
-
- add_prefix = (const uint8_t*)argv[a];
- } else if (argv[a][1] == 'c') {
- if (++a == argc) {
- return missing_arg(argv[0], 'c');
- }
-
- chop_prefix = (const uint8_t*)argv[a];
- } else if (argv[a][1] == 'r') {
- if (++a == argc) {
- return missing_arg(argv[0], 'r');
- }
-
- root_uri = (const uint8_t*)argv[a];
- } else {
- SERDI_ERRORF("invalid option -- '%s'\n", argv[a] + 1);
- return print_usage(argv[0], true);
- }
- }
-
- if (a == argc) {
- SERDI_ERROR("missing input\n");
- return 1;
- }
+ if (argc < 2) {
+ return print_usage(argv[0], true);
+ }
+
+ FILE* in_fd = NULL;
+ SerdSyntax input_syntax = (SerdSyntax)0;
+ SerdSyntax output_syntax = (SerdSyntax)0;
+ bool from_file = true;
+ bool ascii = false;
+ bool bulk_read = true;
+ bool bulk_write = false;
+ bool full_uris = false;
+ bool lax = false;
+ bool quiet = false;
+ const uint8_t* in_name = NULL;
+ const uint8_t* add_prefix = NULL;
+ const uint8_t* chop_prefix = NULL;
+ const uint8_t* root_uri = NULL;
+ int a = 1;
+ for (; a < argc && argv[a][0] == '-'; ++a) {
+ if (argv[a][1] == '\0') {
+ in_name = (const uint8_t*)"(stdin)";
+ in_fd = stdin;
+ break;
+ }
+
+ if (argv[a][1] == 'a') {
+ ascii = true;
+ } else if (argv[a][1] == 'b') {
+ bulk_write = true;
+ } else if (argv[a][1] == 'e') {
+ bulk_read = false;
+ } else if (argv[a][1] == 'f') {
+ full_uris = true;
+ } else if (argv[a][1] == 'h') {
+ return print_usage(argv[0], false);
+ } else if (argv[a][1] == 'l') {
+ lax = true;
+ } else if (argv[a][1] == 'q') {
+ quiet = true;
+ } else if (argv[a][1] == 'v') {
+ return print_version();
+ } else if (argv[a][1] == 's') {
+ in_name = (const uint8_t*)"(string)";
+ from_file = false;
+ ++a;
+ break;
+ } else if (argv[a][1] == 'i') {
+ if (++a == argc) {
+ return missing_arg(argv[0], 'i');
+ }
+
+ if (!(input_syntax = get_syntax(argv[a]))) {
+ return print_usage(argv[0], true);
+ }
+ } else if (argv[a][1] == 'o') {
+ if (++a == argc) {
+ return missing_arg(argv[0], 'o');
+ }
+
+ if (!(output_syntax = get_syntax(argv[a]))) {
+ return print_usage(argv[0], true);
+ }
+ } else if (argv[a][1] == 'p') {
+ if (++a == argc) {
+ return missing_arg(argv[0], 'p');
+ }
+
+ add_prefix = (const uint8_t*)argv[a];
+ } else if (argv[a][1] == 'c') {
+ if (++a == argc) {
+ return missing_arg(argv[0], 'c');
+ }
+
+ chop_prefix = (const uint8_t*)argv[a];
+ } else if (argv[a][1] == 'r') {
+ if (++a == argc) {
+ return missing_arg(argv[0], 'r');
+ }
+
+ root_uri = (const uint8_t*)argv[a];
+ } else {
+ SERDI_ERRORF("invalid option -- '%s'\n", argv[a] + 1);
+ return print_usage(argv[0], true);
+ }
+ }
+
+ if (a == argc) {
+ SERDI_ERROR("missing input\n");
+ return 1;
+ }
#ifdef _WIN32
- _setmode(_fileno(stdin), _O_BINARY);
- _setmode(_fileno(stdout), _O_BINARY);
+ _setmode(_fileno(stdin), _O_BINARY);
+ _setmode(_fileno(stdout), _O_BINARY);
#endif
- uint8_t* input_path = NULL;
- const uint8_t* input = (const uint8_t*)argv[a++];
- if (from_file) {
- in_name = in_name ? in_name : input;
- if (!in_fd) {
- if (!strncmp((const char*)input, "file:", 5)) {
- input_path = serd_file_uri_parse(input, NULL);
- input = input_path;
- }
- if (!input || !(in_fd = serd_fopen((const char*)input, "rb"))) {
- return 1;
- }
- }
- }
-
- if (!input_syntax && !(input_syntax = guess_syntax((const char*)in_name))) {
- input_syntax = SERD_TRIG;
- }
-
- if (!output_syntax) {
- output_syntax = (
- (input_syntax == SERD_TURTLE || input_syntax == SERD_NTRIPLES)
- ? SERD_NTRIPLES
- : SERD_NQUADS);
- }
-
- const SerdStyle output_style =
- choose_style(input_syntax, output_syntax, ascii, bulk_write, full_uris);
-
- SerdURI base_uri = SERD_URI_NULL;
- SerdNode base = SERD_NODE_NULL;
- if (a < argc) { // Base URI given on command line
- base = serd_node_new_uri_from_string(
- (const uint8_t*)argv[a], NULL, &base_uri);
- } else if (from_file && in_fd != stdin) { // Use input file URI
- base = serd_node_new_file_uri(input, NULL, &base_uri, true);
- }
-
- FILE* const out_fd = stdout;
- SerdEnv* const env = serd_env_new(&base);
-
- SerdWriter* const writer = serd_writer_new(
- output_syntax, output_style, env, &base_uri, serd_file_sink, out_fd);
-
- SerdReader* const reader = serd_reader_new(
- input_syntax, writer, NULL,
- (SerdBaseSink)serd_writer_set_base_uri,
- (SerdPrefixSink)serd_writer_set_prefix,
- (SerdStatementSink)serd_writer_write_statement,
- (SerdEndSink)serd_writer_end_anon);
-
- serd_reader_set_strict(reader, !lax);
- if (quiet) {
- serd_reader_set_error_sink(reader, quiet_error_sink, NULL);
- serd_writer_set_error_sink(writer, quiet_error_sink, NULL);
- }
-
- SerdNode root = serd_node_from_string(SERD_URI, root_uri);
- serd_writer_set_root_uri(writer, &root);
- serd_writer_chop_blank_prefix(writer, chop_prefix);
- serd_reader_add_blank_prefix(reader, add_prefix);
-
- SerdStatus st = SERD_SUCCESS;
- if (!from_file) {
- st = serd_reader_read_string(reader, input);
- } else if (bulk_read) {
- st = serd_reader_read_file_handle(reader, in_fd, in_name);
- } else {
- st = serd_reader_start_stream(reader, in_fd, in_name, false);
- while (!st) {
- st = serd_reader_read_chunk(reader);
- }
- serd_reader_end_stream(reader);
- }
-
- serd_reader_free(reader);
- serd_writer_finish(writer);
- serd_writer_free(writer);
- serd_env_free(env);
- serd_node_free(&base);
- free(input_path);
-
- if (from_file) {
- fclose(in_fd);
- }
-
- if (fclose(out_fd)) {
- perror("serdi: write error");
- st = SERD_ERR_UNKNOWN;
- }
-
- return (st > SERD_FAILURE) ? 1 : 0;
+ uint8_t* input_path = NULL;
+ const uint8_t* input = (const uint8_t*)argv[a++];
+ if (from_file) {
+ in_name = in_name ? in_name : input;
+ if (!in_fd) {
+ if (!strncmp((const char*)input, "file:", 5)) {
+ input_path = serd_file_uri_parse(input, NULL);
+ input = input_path;
+ }
+ if (!input || !(in_fd = serd_fopen((const char*)input, "rb"))) {
+ return 1;
+ }
+ }
+ }
+
+ if (!input_syntax && !(input_syntax = guess_syntax((const char*)in_name))) {
+ input_syntax = SERD_TRIG;
+ }
+
+ if (!output_syntax) {
+ output_syntax =
+ ((input_syntax == SERD_TURTLE || input_syntax == SERD_NTRIPLES)
+ ? SERD_NTRIPLES
+ : SERD_NQUADS);
+ }
+
+ const SerdStyle output_style =
+ choose_style(input_syntax, output_syntax, ascii, bulk_write, full_uris);
+
+ SerdURI base_uri = SERD_URI_NULL;
+ SerdNode base = SERD_NODE_NULL;
+ if (a < argc) { // Base URI given on command line
+ base =
+ serd_node_new_uri_from_string((const uint8_t*)argv[a], NULL, &base_uri);
+ } else if (from_file && in_fd != stdin) { // Use input file URI
+ base = serd_node_new_file_uri(input, NULL, &base_uri, true);
+ }
+
+ FILE* const out_fd = stdout;
+ SerdEnv* const env = serd_env_new(&base);
+
+ SerdWriter* const writer = serd_writer_new(
+ output_syntax, output_style, env, &base_uri, serd_file_sink, out_fd);
+
+ SerdReader* const reader =
+ serd_reader_new(input_syntax,
+ writer,
+ NULL,
+ (SerdBaseSink)serd_writer_set_base_uri,
+ (SerdPrefixSink)serd_writer_set_prefix,
+ (SerdStatementSink)serd_writer_write_statement,
+ (SerdEndSink)serd_writer_end_anon);
+
+ serd_reader_set_strict(reader, !lax);
+ if (quiet) {
+ serd_reader_set_error_sink(reader, quiet_error_sink, NULL);
+ serd_writer_set_error_sink(writer, quiet_error_sink, NULL);
+ }
+
+ SerdNode root = serd_node_from_string(SERD_URI, root_uri);
+ serd_writer_set_root_uri(writer, &root);
+ serd_writer_chop_blank_prefix(writer, chop_prefix);
+ serd_reader_add_blank_prefix(reader, add_prefix);
+
+ SerdStatus st = SERD_SUCCESS;
+ if (!from_file) {
+ st = serd_reader_read_string(reader, input);
+ } else if (bulk_read) {
+ st = serd_reader_read_file_handle(reader, in_fd, in_name);
+ } else {
+ st = serd_reader_start_stream(reader, in_fd, in_name, false);
+ while (!st) {
+ st = serd_reader_read_chunk(reader);
+ }
+ serd_reader_end_stream(reader);
+ }
+
+ serd_reader_free(reader);
+ serd_writer_finish(writer);
+ serd_writer_free(writer);
+ serd_env_free(env);
+ serd_node_free(&base);
+ free(input_path);
+
+ if (from_file) {
+ fclose(in_fd);
+ }
+
+ if (fclose(out_fd)) {
+ perror("serdi: write error");
+ st = SERD_ERR_UNKNOWN;
+ }
+
+ return (st > SERD_FAILURE) ? 1 : 0;
}
diff --git a/src/stack.h b/src/stack.h
index 01f69de6..053255d3 100644
--- a/src/stack.h
+++ b/src/stack.h
@@ -28,9 +28,9 @@
/** A dynamic stack in memory. */
typedef struct {
- uint8_t* buf; ///< Stack memory
- size_t buf_size; ///< Allocated size of buf (>= size)
- size_t size; ///< Conceptual size of stack in buf
+ uint8_t* buf; ///< Stack memory
+ size_t buf_size; ///< Allocated size of buf (>= size)
+ size_t size; ///< Conceptual size of stack in buf
} SerdStack;
/** An offset to start the stack at. Note 0 is reserved for NULL. */
@@ -39,79 +39,81 @@ typedef struct {
static inline SerdStack
serd_stack_new(size_t size)
{
- SerdStack stack;
- stack.buf = (uint8_t*)calloc(size, 1);
- stack.buf_size = size;
- stack.size = SERD_STACK_BOTTOM;
- return stack;
+ SerdStack stack;
+ stack.buf = (uint8_t*)calloc(size, 1);
+ stack.buf_size = size;
+ stack.size = SERD_STACK_BOTTOM;
+ return stack;
}
static inline bool
serd_stack_is_empty(SerdStack* stack)
{
- return stack->size <= SERD_STACK_BOTTOM;
+ return stack->size <= SERD_STACK_BOTTOM;
}
static inline void
serd_stack_free(SerdStack* stack)
{
- free(stack->buf);
- stack->buf = NULL;
- stack->buf_size = 0;
- stack->size = 0;
+ free(stack->buf);
+ stack->buf = NULL;
+ stack->buf_size = 0;
+ stack->size = 0;
}
static inline void*
serd_stack_push(SerdStack* stack, size_t n_bytes)
{
- const size_t new_size = stack->size + n_bytes;
- if (stack->buf_size < new_size) {
- stack->buf_size += (stack->buf_size >> 1); // *= 1.5
- stack->buf = (uint8_t*)realloc(stack->buf, stack->buf_size);
- }
- uint8_t* const ret = (stack->buf + stack->size);
- stack->size = new_size;
- return ret;
+ const size_t new_size = stack->size + n_bytes;
+ if (stack->buf_size < new_size) {
+ stack->buf_size += (stack->buf_size >> 1); // *= 1.5
+ stack->buf = (uint8_t*)realloc(stack->buf, stack->buf_size);
+ }
+
+ uint8_t* const ret = (stack->buf + stack->size);
+
+ stack->size = new_size;
+ return ret;
}
static inline void
serd_stack_pop(SerdStack* stack, size_t n_bytes)
{
- assert(stack->size >= n_bytes);
- stack->size -= n_bytes;
+ assert(stack->size >= n_bytes);
+ stack->size -= n_bytes;
}
static inline void*
serd_stack_push_aligned(SerdStack* stack, size_t n_bytes, size_t align)
{
- // Push one byte to ensure space for a pad count
- serd_stack_push(stack, 1);
+ // Push one byte to ensure space for a pad count
+ serd_stack_push(stack, 1);
- // Push padding if necessary
- const size_t pad = align - stack->size % align;
- if (pad > 0) {
- serd_stack_push(stack, pad);
- }
+ // Push padding if necessary
+ const size_t pad = align - stack->size % align;
+ if (pad > 0) {
+ serd_stack_push(stack, pad);
+ }
- // Set top of stack to pad count so we can properly pop later
- assert(pad < UINT8_MAX);
- stack->buf[stack->size - 1] = (uint8_t)pad;
+ // Set top of stack to pad count so we can properly pop later
+ assert(pad < UINT8_MAX);
+ stack->buf[stack->size - 1] = (uint8_t)pad;
- // Push requested space at aligned location
- return serd_stack_push(stack, n_bytes);
+ // Push requested space at aligned location
+ return serd_stack_push(stack, n_bytes);
}
static inline void
serd_stack_pop_aligned(SerdStack* stack, size_t n_bytes)
{
- // Pop requested space down to aligned location
- serd_stack_pop(stack, n_bytes);
+ // Pop requested space down to aligned location
+ serd_stack_pop(stack, n_bytes);
- // Get amount of padding from top of stack
- const uint8_t pad = stack->buf[stack->size - 1];
+ // Get amount of padding from top of stack
+ const uint8_t pad = stack->buf[stack->size - 1];
- // Pop padding and pad count
- serd_stack_pop(stack, pad + 1u);
+ // Pop padding and pad count
+ serd_stack_pop(stack, pad + 1u);
}
-#endif // SERD_STACK_H
+#endif // SERD_STACK_H
diff --git a/src/string.c b/src/string.c
index e1e5dbda..6946ba09 100644
--- a/src/string.c
+++ b/src/string.c
@@ -25,39 +25,50 @@
void
serd_free(void* ptr)
{
- free(ptr);
+ free(ptr);
}
const uint8_t*
serd_strerror(SerdStatus status)
{
- switch (status) {
- case SERD_SUCCESS: return (const uint8_t*)"Success";
- case SERD_FAILURE: return (const uint8_t*)"Non-fatal failure";
- case SERD_ERR_UNKNOWN: return (const uint8_t*)"Unknown error";
- case SERD_ERR_BAD_SYNTAX: return (const uint8_t*)"Invalid syntax";
- case SERD_ERR_BAD_ARG: return (const uint8_t*)"Invalid argument";
- case SERD_ERR_NOT_FOUND: return (const uint8_t*)"Not found";
- case SERD_ERR_ID_CLASH: return (const uint8_t*)"Blank node ID clash";
- case SERD_ERR_BAD_CURIE: return (const uint8_t*)"Invalid CURIE";
- case SERD_ERR_INTERNAL: return (const uint8_t*)"Internal error";
- default: break;
- }
- return (const uint8_t*)"Unknown error"; // never reached
+ switch (status) {
+ case SERD_SUCCESS:
+ return (const uint8_t*)"Success";
+ case SERD_FAILURE:
+ return (const uint8_t*)"Non-fatal failure";
+ case SERD_ERR_UNKNOWN:
+ return (const uint8_t*)"Unknown error";
+ case SERD_ERR_BAD_SYNTAX:
+ return (const uint8_t*)"Invalid syntax";
+ case SERD_ERR_BAD_ARG:
+ return (const uint8_t*)"Invalid argument";
+ case SERD_ERR_NOT_FOUND:
+ return (const uint8_t*)"Not found";
+ case SERD_ERR_ID_CLASH:
+ return (const uint8_t*)"Blank node ID clash";
+ case SERD_ERR_BAD_CURIE:
+ return (const uint8_t*)"Invalid CURIE";
+ case SERD_ERR_INTERNAL:
+ return (const uint8_t*)"Internal error";
+ default:
+ break;
+ }
+ return (const uint8_t*)"Unknown error"; // never reached
}
static inline void
serd_update_flags(const uint8_t c, SerdNodeFlags* const flags)
{
- switch (c) {
- case '\r': case '\n':
- *flags |= SERD_HAS_NEWLINE;
- break;
- case '"':
- *flags |= SERD_HAS_QUOTE;
- default:
- break;
- }
+ switch (c) {
+ case '\r':
+ case '\n':
+ *flags |= SERD_HAS_NEWLINE;
+ break;
+ case '"':
+ *flags |= SERD_HAS_QUOTE;
+ default:
+ break;
+ }
}
size_t
@@ -66,101 +77,103 @@ serd_substrlen(const uint8_t* const str,
size_t* const n_bytes,
SerdNodeFlags* const flags)
{
- size_t n_chars = 0;
- size_t i = 0;
- SerdNodeFlags f = 0;
- for (; i < len && str[i]; ++i) {
- if ((str[i] & 0xC0) != 0x80) { // Start of new character
- ++n_chars;
- serd_update_flags(str[i], &f);
- }
- }
- if (n_bytes) {
- *n_bytes = i;
- }
- if (flags) {
- *flags = f;
- }
- return n_chars;
+ size_t n_chars = 0;
+ size_t i = 0;
+ SerdNodeFlags f = 0;
+ for (; i < len && str[i]; ++i) {
+ if ((str[i] & 0xC0) != 0x80) { // Start of new character
+ ++n_chars;
+ serd_update_flags(str[i], &f);
+ }
+ }
+ if (n_bytes) {
+ *n_bytes = i;
+ }
+ if (flags) {
+ *flags = f;
+ }
+ return n_chars;
}
size_t
serd_strlen(const uint8_t* str, size_t* n_bytes, SerdNodeFlags* flags)
{
- size_t n_chars = 0;
- size_t i = 0;
- SerdNodeFlags f = 0;
- for (; str[i]; ++i) {
- if ((str[i] & 0xC0) != 0x80) { // Start of new character
- ++n_chars;
- serd_update_flags(str[i], &f);
- }
- }
- if (n_bytes) {
- *n_bytes = i;
- }
- if (flags) {
- *flags = f;
- }
- return n_chars;
+ size_t n_chars = 0;
+ size_t i = 0;
+ SerdNodeFlags f = 0;
+ for (; str[i]; ++i) {
+ if ((str[i] & 0xC0) != 0x80) { // Start of new character
+ ++n_chars;
+ serd_update_flags(str[i], &f);
+ }
+ }
+ if (n_bytes) {
+ *n_bytes = i;
+ }
+ if (flags) {
+ *flags = f;
+ }
+ return n_chars;
}
static inline double
read_sign(const char** sptr)
{
- double sign = 1.0;
- switch (**sptr) {
- case '-':
- sign = -1.0;
- // fallthru
- case '+':
- ++(*sptr);
- // fallthru
- default:
- return sign;
- }
+ double sign = 1.0;
+ switch (**sptr) {
+ case '-':
+ sign = -1.0;
+ // fallthru
+ case '+':
+ ++(*sptr);
+ // fallthru
+ default:
+ return sign;
+ }
}
double
serd_strtod(const char* str, char** endptr)
{
- double result = 0.0;
-
- // Point s at the first non-whitespace character
- const char* s = str;
- while (is_space(*s)) { ++s; }
-
- // Read leading sign if necessary
- const double sign = read_sign(&s);
-
- // Parse integer part
- for (; is_digit(*s); ++s) {
- result = (result * 10.0) + (*s - '0');
- }
-
- // Parse fractional part
- if (*s == '.') {
- double denom = 10.0;
- for (++s; is_digit(*s); ++s) {
- result += (*s - '0') / denom;
- denom *= 10.0;
- }
- }
-
- // Parse exponent
- if (*s == 'e' || *s == 'E') {
- ++s;
- double expt = 0.0;
- double expt_sign = read_sign(&s);
- for (; is_digit(*s); ++s) {
- expt = (expt * 10.0) + (*s - '0');
- }
- result *= pow(10, expt * expt_sign);
- }
-
- if (endptr) {
- *endptr = (char*)s;
- }
-
- return result * sign;
+ double result = 0.0;
+
+ // Point s at the first non-whitespace character
+ const char* s = str;
+ while (is_space(*s)) {
+ ++s;
+ }
+
+ // Read leading sign if necessary
+ const double sign = read_sign(&s);
+
+ // Parse integer part
+ for (; is_digit(*s); ++s) {
+ result = (result * 10.0) + (*s - '0');
+ }
+
+ // Parse fractional part
+ if (*s == '.') {
+ double denom = 10.0;
+ for (++s; is_digit(*s); ++s) {
+ result += (*s - '0') / denom;
+ denom *= 10.0;
+ }
+ }
+
+ // Parse exponent
+ if (*s == 'e' || *s == 'E') {
+ ++s;
+ double expt = 0.0;
+ double expt_sign = read_sign(&s);
+ for (; is_digit(*s); ++s) {
+ expt = (expt * 10.0) + (*s - '0');
+ }
+ result *= pow(10, expt * expt_sign);
+ }
+
+ if (endptr) {
+ *endptr = (char*)s;
+ }
+
+ return result * sign;
}
diff --git a/src/string_utils.h b/src/string_utils.h
index a80c3a27..b6b77c95 100644
--- a/src/string_utils.h
+++ b/src/string_utils.h
@@ -24,71 +24,76 @@
#include <stdint.h>
/** Unicode replacement character in UTF-8 */
-static const uint8_t replacement_char[] = { 0xEF, 0xBF, 0xBD };
+static const uint8_t replacement_char[] = {0xEF, 0xBF, 0xBD};
/** Return true if `c` lies within [`min`...`max`] (inclusive) */
static inline bool
in_range(const int c, const int min, const int max)
{
- return (c >= min && c <= max);
+ return (c >= min && c <= max);
}
/** RFC2234: ALPHA ::= %x41-5A / %x61-7A ; A-Z / a-z */
static inline bool
is_alpha(const int c)
{
- return in_range(c, 'A', 'Z') || in_range(c, 'a', 'z');
+ return in_range(c, 'A', 'Z') || in_range(c, 'a', 'z');
}
/** RFC2234: DIGIT ::= %x30-39 ; 0-9 */
static inline bool
is_digit(const int c)
{
- return in_range(c, '0', '9');
+ return in_range(c, '0', '9');
}
/* RFC2234: HEXDIG ::= DIGIT / "A" / "B" / "C" / "D" / "E" / "F" */
static inline bool
is_hexdig(const int c)
{
- return is_digit(c) || in_range(c, 'A', 'F');
+ return is_digit(c) || in_range(c, 'A', 'F');
}
/* Turtle / JSON / C: XDIGIT ::= DIGIT / A-F / a-f */
static inline bool
is_xdigit(const int c)
{
- return is_hexdig(c) || in_range(c, 'a', 'f');
+ return is_hexdig(c) || in_range(c, 'a', 'f');
}
static inline bool
is_space(const char c)
{
- switch (c) {
- case ' ': case '\f': case '\n': case '\r': case '\t': case '\v':
- return true;
- default:
- return false;
- }
+ switch (c) {
+ case ' ':
+ case '\f':
+ case '\n':
+ case '\r':
+ case '\t':
+ case '\v':
+ return true;
+ default:
+ return false;
+ }
}
static inline bool
is_print(const int c)
{
- return c >= 0x20 && c <= 0x7E;
+ return c >= 0x20 && c <= 0x7E;
}
static inline bool
is_base64(const uint8_t c)
{
- return is_alpha(c) || is_digit(c) || c == '+' || c == '/' || c == '=';
+ return is_alpha(c) || is_digit(c) || c == '+' || c == '/' || c == '=';
}
static inline bool
is_windows_path(const uint8_t* path)
{
- return is_alpha(path[0]) && (path[1] == ':' || path[1] == '|')
- && (path[2] == '/' || path[2] == '\\');
+ return is_alpha(path[0]) && (path[1] == ':' || path[1] == '|') &&
+ (path[2] == '/' || path[2] == '\\');
}
size_t
@@ -100,65 +105,69 @@ serd_substrlen(const uint8_t* str,
static inline char
serd_to_upper(const char c)
{
- return (char)((c >= 'a' && c <= 'z') ? c - 32 : c);
+ return (char)((c >= 'a' && c <= 'z') ? c - 32 : c);
}
static inline int
serd_strncasecmp(const char* s1, const char* s2, size_t n)
{
- for (; n > 0 && *s2; s1++, s2++, --n) {
- if (serd_to_upper(*s1) != serd_to_upper(*s2)) {
- return ((*(const uint8_t*)s1 < *(const uint8_t*)s2) ? -1 : +1);
- }
- }
- return 0;
+ for (; n > 0 && *s2; s1++, s2++, --n) {
+ if (serd_to_upper(*s1) != serd_to_upper(*s2)) {
+ return ((*(const uint8_t*)s1 < *(const uint8_t*)s2) ? -1 : +1);
+ }
+ }
+
+ return 0;
}
static inline uint32_t
utf8_num_bytes(const uint8_t c)
{
- if ((c & 0x80) == 0) { // Starts with `0'
- return 1;
- }
+ if ((c & 0x80) == 0) { // Starts with `0'
+ return 1;
+ }
- if ((c & 0xE0) == 0xC0) { // Starts with `110'
- return 2;
- }
+ if ((c & 0xE0) == 0xC0) { // Starts with `110'
+ return 2;
+ }
- if ((c & 0xF0) == 0xE0) { // Starts with `1110'
- return 3;
- }
+ if ((c & 0xF0) == 0xE0) { // Starts with `1110'
+ return 3;
+ }
- if ((c & 0xF8) == 0xF0) { // Starts with `11110'
- return 4;
- }
+ if ((c & 0xF8) == 0xF0) { // Starts with `11110'
+ return 4;
+ }
- return 0;
+ return 0;
}
/// Return the code point of a UTF-8 character with known length
static inline uint32_t
parse_counted_utf8_char(const uint8_t* utf8, size_t size)
{
- uint32_t c = utf8[0] & ((1u << (8 - size)) - 1);
- for (size_t i = 1; i < size; ++i) {
- const uint8_t in = utf8[i] & 0x3F;
- c = (c << 6) | in;
- }
- return c;
+ uint32_t c = utf8[0] & ((1u << (8 - size)) - 1);
+ for (size_t i = 1; i < size; ++i) {
+ const uint8_t in = utf8[i] & 0x3F;
+ c = (c << 6) | in;
+ }
+ return c;
}
/// Parse a UTF-8 character, set *size to the length, and return the code point
static inline uint32_t
parse_utf8_char(const uint8_t* utf8, size_t* size)
{
- switch (*size = utf8_num_bytes(utf8[0])) {
- case 1: case 2: case 3: case 4:
- return parse_counted_utf8_char(utf8, *size);
- default:
- *size = 0;
- return 0;
- }
+ switch (*size = utf8_num_bytes(utf8[0])) {
+ case 1:
+ case 2:
+ case 3:
+ case 4:
+ return parse_counted_utf8_char(utf8, *size);
+ default:
+ *size = 0;
+ return 0;
+ }
}
-#endif // SERD_STRING_UTILS_H
+#endif // SERD_STRING_UTILS_H
diff --git a/src/system.c b/src/system.c
index c6796da9..3f10b86b 100644
--- a/src/system.c
+++ b/src/system.c
@@ -22,11 +22,11 @@
#include "serd_internal.h"
#if defined(HAVE_POSIX_FADVISE) && defined(HAVE_FILENO)
-# include <fcntl.h>
+# include <fcntl.h>
#endif
#ifdef _WIN32
-# include <malloc.h>
+# include <malloc.h>
#endif
#include <errno.h>
@@ -37,47 +37,48 @@
FILE*
serd_fopen(const char* path, const char* mode)
{
- FILE* fd = fopen(path, mode);
- if (!fd) {
- fprintf(stderr, "error: failed to open file %s (%s)\n",
- path, strerror(errno));
- return NULL;
- }
+ FILE* fd = fopen(path, mode);
+ if (!fd) {
+ fprintf(
+ stderr, "error: failed to open file %s (%s)\n", path, strerror(errno));
+ return NULL;
+ }
+
#if defined(HAVE_POSIX_FADVISE) && defined(HAVE_FILENO)
- posix_fadvise(fileno(fd), 0, 0, POSIX_FADV_SEQUENTIAL);
+ posix_fadvise(fileno(fd), 0, 0, POSIX_FADV_SEQUENTIAL);
#endif
- return fd;
+ return fd;
}
void*
serd_malloc_aligned(const size_t alignment, const size_t size)
{
#if defined(_WIN32)
- return _aligned_malloc(size, alignment);
+ return _aligned_malloc(size, alignment);
#elif __STDC_VERSION__ >= 201112L && defined(HAVE_ALIGNED_ALLOC)
- return aligned_alloc(alignment, size);
+ return aligned_alloc(alignment, size);
#elif defined(HAVE_POSIX_MEMALIGN)
- void* ptr = NULL;
- const int ret = posix_memalign(&ptr, alignment, size);
- return ret ? NULL : ptr;
+ void* ptr = NULL;
+ const int ret = posix_memalign(&ptr, alignment, size);
+ return ret ? NULL : ptr;
#else
- (void)alignment;
- return malloc(size);
+ (void)alignment;
+ return malloc(size);
#endif
}
void*
serd_allocate_buffer(const size_t size)
{
- return serd_malloc_aligned(SERD_PAGE_SIZE, size);
+ return serd_malloc_aligned(SERD_PAGE_SIZE, size);
}
void
serd_free_aligned(void* const ptr)
{
#ifdef _WIN32
- _aligned_free(ptr);
+ _aligned_free(ptr);
#else
- free(ptr);
+ free(ptr);
#endif
}
diff --git a/src/system.h b/src/system.h
index 57203820..fe2713b5 100644
--- a/src/system.h
+++ b/src/system.h
@@ -22,15 +22,19 @@
#include <stdio.h>
/// Open a file configured for fast sequential reading
-FILE* serd_fopen(const char* path, const char* mode);
+FILE*
+serd_fopen(const char* path, const char* mode);
/// Allocate a buffer aligned to `alignment` bytes
-SERD_MALLOC_FUNC void* serd_malloc_aligned(size_t alignment, size_t size);
+SERD_MALLOC_FUNC void*
+serd_malloc_aligned(size_t alignment, size_t size);
/// Allocate an aligned buffer for I/O
-SERD_MALLOC_FUNC void* serd_allocate_buffer(size_t size);
+SERD_MALLOC_FUNC void*
+serd_allocate_buffer(size_t size);
/// Free a buffer allocated with an aligned allocation function
-void serd_free_aligned(void* ptr);
+void
+serd_free_aligned(void* ptr);
#endif // SERD_SYSTEM_H
diff --git a/src/uri.c b/src/uri.c
index 718f24ee..6b387459 100644
--- a/src/uri.c
+++ b/src/uri.c
@@ -28,198 +28,214 @@
const uint8_t*
serd_uri_to_path(const uint8_t* uri)
{
- const uint8_t* path = uri;
- if (!is_windows_path(uri) && serd_uri_string_has_scheme(uri)) {
- if (strncmp((const char*)uri, "file:", 5)) {
- fprintf(stderr, "Non-file URI `%s'\n", uri);
- return NULL;
- }
-
- if (!strncmp((const char*)uri, "file://localhost/", 17)) {
- path = uri + 16;
- } else if (!strncmp((const char*)uri, "file://", 7)) {
- path = uri + 7;
- } else {
- fprintf(stderr, "Invalid file URI `%s'\n", uri);
- return NULL;
- }
-
- if (is_windows_path(path + 1)) {
- ++path; // Special case for terrible Windows file URIs
- }
- }
- return path;
+ const uint8_t* path = uri;
+ if (!is_windows_path(uri) && serd_uri_string_has_scheme(uri)) {
+ if (strncmp((const char*)uri, "file:", 5)) {
+ fprintf(stderr, "Non-file URI `%s'\n", uri);
+ return NULL;
+ }
+
+ if (!strncmp((const char*)uri, "file://localhost/", 17)) {
+ path = uri + 16;
+ } else if (!strncmp((const char*)uri, "file://", 7)) {
+ path = uri + 7;
+ } else {
+ fprintf(stderr, "Invalid file URI `%s'\n", uri);
+ return NULL;
+ }
+
+ if (is_windows_path(path + 1)) {
+ ++path; // Special case for terrible Windows file URIs
+ }
+ }
+ return path;
}
uint8_t*
serd_file_uri_parse(const uint8_t* uri, uint8_t** hostname)
{
- const uint8_t* path = uri;
- if (hostname) {
- *hostname = NULL;
- }
- if (!strncmp((const char*)uri, "file://", 7)) {
- const uint8_t* auth = uri + 7;
- if (*auth == '/') { // No hostname
- path = auth;
- } else { // Has hostname
- if (!(path = (const uint8_t*)strchr((const char*)auth, '/'))) {
- return NULL;
- }
- if (hostname) {
- *hostname = (uint8_t*)calloc((size_t)(path - auth + 1), 1);
- memcpy(*hostname, auth, (size_t)(path - auth));
- }
- }
- }
-
- if (is_windows_path(path + 1)) {
- ++path;
- }
-
- SerdChunk chunk = { NULL, 0 };
- for (const uint8_t* s = path; *s; ++s) {
- if (*s == '%') {
- if (*(s + 1) == '%') {
- serd_chunk_sink("%", 1, &chunk);
- ++s;
- } else if (is_hexdig(*(s + 1)) && is_hexdig(*(s + 2))) {
- const uint8_t code[3] = {*(s + 1), *(s + 2), 0};
- const uint8_t c = (uint8_t)strtoul((const char*)code, NULL, 16);
- serd_chunk_sink(&c, 1, &chunk);
- s += 2;
- } else {
- s += 2; // Junk escape, ignore
- }
- } else {
- serd_chunk_sink(s, 1, &chunk);
- }
- }
- return serd_chunk_sink_finish(&chunk);
+ const uint8_t* path = uri;
+ if (hostname) {
+ *hostname = NULL;
+ }
+ if (!strncmp((const char*)uri, "file://", 7)) {
+ const uint8_t* auth = uri + 7;
+ if (*auth == '/') { // No hostname
+ path = auth;
+ } else { // Has hostname
+ if (!(path = (const uint8_t*)strchr((const char*)auth, '/'))) {
+ return NULL;
+ }
+
+ if (hostname) {
+ *hostname = (uint8_t*)calloc((size_t)(path - auth + 1), 1);
+ memcpy(*hostname, auth, (size_t)(path - auth));
+ }
+ }
+ }
+
+ if (is_windows_path(path + 1)) {
+ ++path;
+ }
+
+ SerdChunk chunk = {NULL, 0};
+ for (const uint8_t* s = path; *s; ++s) {
+ if (*s == '%') {
+ if (*(s + 1) == '%') {
+ serd_chunk_sink("%", 1, &chunk);
+ ++s;
+ } else if (is_hexdig(*(s + 1)) && is_hexdig(*(s + 2))) {
+ const uint8_t code[3] = {*(s + 1), *(s + 2), 0};
+ const uint8_t c = (uint8_t)strtoul((const char*)code, NULL, 16);
+ serd_chunk_sink(&c, 1, &chunk);
+ s += 2;
+ } else {
+ s += 2; // Junk escape, ignore
+ }
+ } else {
+ serd_chunk_sink(s, 1, &chunk);
+ }
+ }
+
+ return serd_chunk_sink_finish(&chunk);
}
bool
serd_uri_string_has_scheme(const uint8_t* utf8)
{
- // RFC3986: scheme ::= ALPHA *( ALPHA / DIGIT / "+" / "-" / "." )
- if (!utf8 || !is_alpha(utf8[0])) {
- return false; // Invalid scheme initial character, URI is relative
- }
-
- for (uint8_t c = 0; (c = *++utf8) != '\0';) {
- if (!is_uri_scheme_char(c)) {
- return false;
- }
-
- if (c == ':') {
- return true; // End of scheme
- }
- }
-
- return false;
+ // RFC3986: scheme ::= ALPHA *( ALPHA / DIGIT / "+" / "-" / "." )
+ if (!utf8 || !is_alpha(utf8[0])) {
+ return false; // Invalid scheme initial character, URI is relative
+ }
+
+ for (uint8_t c = 0; (c = *++utf8) != '\0';) {
+ if (!is_uri_scheme_char(c)) {
+ return false;
+ }
+
+ if (c == ':') {
+ return true; // End of scheme
+ }
+ }
+
+ return false;
}
SerdStatus
serd_uri_parse(const uint8_t* utf8, SerdURI* out)
{
- *out = SERD_URI_NULL;
-
- const uint8_t* ptr = utf8;
-
- /* See http://tools.ietf.org/html/rfc3986#section-3
- URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ]
- */
-
- /* S3.1: scheme ::= ALPHA *( ALPHA / DIGIT / "+" / "-" / "." ) */
- if (is_alpha(*ptr)) {
- for (uint8_t c = *++ptr; true; c = *++ptr) {
- switch (c) {
- case '\0': case '/': case '?': case '#':
- ptr = utf8;
- goto path; // Relative URI (starts with path by definition)
- case ':':
- out->scheme.buf = utf8;
- out->scheme.len = (size_t)((ptr++) - utf8);
- goto maybe_authority; // URI with scheme
- case '+': case '-': case '.':
- continue;
- default:
- if (is_alpha(c) || is_digit(c)) {
- continue;
- }
- }
- }
- }
-
- /* S3.2: The authority component is preceded by a double slash ("//")
- and is terminated by the next slash ("/"), question mark ("?"),
- or number sign ("#") character, or by the end of the URI.
- */
+ *out = SERD_URI_NULL;
+
+ const uint8_t* ptr = utf8;
+
+ /* See http://tools.ietf.org/html/rfc3986#section-3
+ URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ]
+ */
+
+ /* S3.1: scheme ::= ALPHA *( ALPHA / DIGIT / "+" / "-" / "." ) */
+ if (is_alpha(*ptr)) {
+ for (uint8_t c = *++ptr; true; c = *++ptr) {
+ switch (c) {
+ case '\0':
+ case '/':
+ case '?':
+ case '#':
+ ptr = utf8;
+ goto path; // Relative URI (starts with path by definition)
+ case ':':
+ out->scheme.buf = utf8;
+ out->scheme.len = (size_t)((ptr++) - utf8);
+ goto maybe_authority; // URI with scheme
+ case '+':
+ case '-':
+ case '.':
+ continue;
+ default:
+ if (is_alpha(c) || is_digit(c)) {
+ continue;
+ }
+ }
+ }
+ }
+
+ /* S3.2: The authority component is preceded by a double slash ("//")
+ and is terminated by the next slash ("/"), question mark ("?"),
+ or number sign ("#") character, or by the end of the URI.
+ */
maybe_authority:
- if (*ptr == '/' && *(ptr + 1) == '/') {
- ptr += 2;
- out->authority.buf = ptr;
- for (uint8_t c = 0; (c = *ptr) != '\0'; ++ptr) {
- switch (c) {
- case '/': goto path;
- case '?': goto query;
- case '#': goto fragment;
- default:
- ++out->authority.len;
- }
- }
- }
-
- /* RFC3986 S3.3: The path is terminated by the first question mark ("?")
- or number sign ("#") character, or by the end of the URI.
- */
+ if (*ptr == '/' && *(ptr + 1) == '/') {
+ ptr += 2;
+ out->authority.buf = ptr;
+ for (uint8_t c = 0; (c = *ptr) != '\0'; ++ptr) {
+ switch (c) {
+ case '/':
+ goto path;
+ case '?':
+ goto query;
+ case '#':
+ goto fragment;
+ default:
+ ++out->authority.len;
+ }
+ }
+ }
+
+ /* RFC3986 S3.3: The path is terminated by the first question mark ("?")
+ or number sign ("#") character, or by the end of the URI.
+ */
path:
- switch (*ptr) {
- case '?': goto query;
- case '#': goto fragment;
- case '\0': goto end;
- default: break;
- }
- out->path.buf = ptr;
- out->path.len = 0;
- for (uint8_t c = 0; (c = *ptr) != '\0'; ++ptr) {
- switch (c) {
- case '?': goto query;
- case '#': goto fragment;
- default:
- ++out->path.len;
- }
- }
-
- /* RFC3986 S3.4: The query component is indicated by the first question
- mark ("?") character and terminated by a number sign ("#") character
- or by the end of the URI.
- */
+ switch (*ptr) {
+ case '?':
+ goto query;
+ case '#':
+ goto fragment;
+ case '\0':
+ goto end;
+ default:
+ break;
+ }
+ out->path.buf = ptr;
+ out->path.len = 0;
+ for (uint8_t c = 0; (c = *ptr) != '\0'; ++ptr) {
+ switch (c) {
+ case '?':
+ goto query;
+ case '#':
+ goto fragment;
+ default:
+ ++out->path.len;
+ }
+ }
+
+ /* RFC3986 S3.4: The query component is indicated by the first question
+ mark ("?") character and terminated by a number sign ("#") character
+ or by the end of the URI.
+ */
query:
- if (*ptr == '?') {
- out->query.buf = ++ptr;
- for (uint8_t c = 0; (c = *ptr) != '\0'; ++ptr) {
- if (c == '#') {
- goto fragment;
- }
- ++out->query.len;
- }
- }
-
- /* RFC3986 S3.5: A fragment identifier component is indicated by the
- presence of a number sign ("#") character and terminated by the end
- of the URI.
- */
+ if (*ptr == '?') {
+ out->query.buf = ++ptr;
+ for (uint8_t c = 0; (c = *ptr) != '\0'; ++ptr) {
+ if (c == '#') {
+ goto fragment;
+ }
+ ++out->query.len;
+ }
+ }
+
+ /* RFC3986 S3.5: A fragment identifier component is indicated by the
+ presence of a number sign ("#") character and terminated by the end
+ of the URI.
+ */
fragment:
- if (*ptr == '#') {
- out->fragment.buf = ptr;
- while (*ptr++ != '\0') {
- ++out->fragment.len;
- }
- }
+ if (*ptr == '#') {
+ out->fragment.buf = ptr;
+ while (*ptr++ != '\0') {
+ ++out->fragment.len;
+ }
+ }
end:
- return SERD_SUCCESS;
+ return SERD_SUCCESS;
}
/**
@@ -231,149 +247,153 @@ end:
static const uint8_t*
remove_dot_segments(const uint8_t* path, size_t len, size_t* up)
{
- const uint8_t* begin = path;
- const uint8_t* const end = path + len;
-
- *up = 0;
- while (begin < end) {
- switch (begin[0]) {
- case '.':
- switch (begin[1]) {
- case '/':
- begin += 2; // Chop leading "./"
- break;
- case '.':
- switch (begin[2]) {
- case '\0':
- ++*up;
- begin += 2; // Chop input ".."
- break;
- case '/':
- ++*up;
- begin += 3; // Chop leading "../"
- break;
- default:
- return begin;
- }
- break;
- case '\0':
- ++begin; // Chop input "."
- // fallthru
- default:
- return begin;
- }
- break;
- case '/':
- switch (begin[1]) {
- case '.':
- switch (begin[2]) {
- case '/':
- begin += 2; // Leading "/./" => "/"
- break;
- case '.':
- switch (begin[3]) {
- case '/':
- ++*up;
- begin += 3; // Leading "/../" => "/"
- }
- break;
- default:
- return begin;
- }
- } // else fall through
- default:
- return begin; // Finished chopping dot components
- }
- }
-
- return begin;
+ const uint8_t* begin = path;
+ const uint8_t* const end = path + len;
+
+ *up = 0;
+ while (begin < end) {
+ switch (begin[0]) {
+ case '.':
+ switch (begin[1]) {
+ case '/':
+ begin += 2; // Chop leading "./"
+ break;
+ case '.':
+ switch (begin[2]) {
+ case '\0':
+ ++*up;
+ begin += 2; // Chop input ".."
+ break;
+ case '/':
+ ++*up;
+ begin += 3; // Chop leading "../"
+ break;
+ default:
+ return begin;
+ }
+ break;
+ case '\0':
+ return ++begin; // Chop input "."
+ default:
+ return begin;
+ }
+ break;
+
+ case '/':
+ switch (begin[1]) {
+ case '.':
+ switch (begin[2]) {
+ case '/':
+ begin += 2; // Leading "/./" => "/"
+ break;
+ case '.':
+ switch (begin[3]) {
+ case '/':
+ ++*up;
+ begin += 3; // Leading "/../" => "/"
+ }
+ break;
+ default:
+ return begin;
+ }
+ }
+ return begin;
+
+ default:
+ return begin; // Finished chopping dot components
+ }
+ }
+
+ return begin;
}
/// Merge `base` and `path` in-place
static void
merge(SerdChunk* base, SerdChunk* path)
{
- size_t up = 0;
- const uint8_t* begin = remove_dot_segments(path->buf, path->len, &up);
- const uint8_t* end = path->buf + path->len;
-
- if (base->len) {
- // Find the up'th last slash
- const uint8_t* base_last = (base->buf + base->len - 1);
- ++up;
- do {
- if (*base_last == '/') {
- --up;
- }
- } while (up > 0 && (--base_last > base->buf));
-
- // Set path prefix
- base->len = (size_t)(base_last - base->buf + 1);
- }
-
- // Set path suffix
- path->buf = begin;
- path->len = (size_t)(end - begin);
+ size_t up = 0;
+ const uint8_t* begin = remove_dot_segments(path->buf, path->len, &up);
+ const uint8_t* end = path->buf + path->len;
+
+ if (base->len) {
+ // Find the up'th last slash
+ const uint8_t* base_last = (base->buf + base->len - 1);
+ ++up;
+ do {
+ if (*base_last == '/') {
+ --up;
+ }
+ } while (up > 0 && (--base_last > base->buf));
+
+ // Set path prefix
+ base->len = (size_t)(base_last - base->buf + 1);
+ }
+
+ // Set path suffix
+ path->buf = begin;
+ path->len = (size_t)(end - begin);
}
/// See http://tools.ietf.org/html/rfc3986#section-5.2.2
void
serd_uri_resolve(const SerdURI* r, const SerdURI* base, SerdURI* t)
{
- if (!base->scheme.len) {
- *t = *r; // Don't resolve against non-absolute URIs
- return;
- }
-
- t->path_base.buf = NULL;
- t->path_base.len = 0;
- if (r->scheme.len) {
- *t = *r;
- } else {
- if (r->authority.len) {
- t->authority = r->authority;
- t->path = r->path;
- t->query = r->query;
- } else {
- t->path = r->path;
- if (!r->path.len) {
- t->path_base = base->path;
- if (r->query.len) {
- t->query = r->query;
- } else {
- t->query = base->query;
- }
- } else {
- if (r->path.buf[0] != '/') {
- t->path_base = base->path;
- }
- merge(&t->path_base, &t->path);
- t->query = r->query;
- }
- t->authority = base->authority;
- }
- t->scheme = base->scheme;
- t->fragment = r->fragment;
- }
+ if (!base->scheme.len) {
+ *t = *r; // Don't resolve against non-absolute URIs
+ return;
+ }
+
+ t->path_base.buf = NULL;
+ t->path_base.len = 0;
+ if (r->scheme.len) {
+ *t = *r;
+ } else {
+ if (r->authority.len) {
+ t->authority = r->authority;
+ t->path = r->path;
+ t->query = r->query;
+ } else {
+ t->path = r->path;
+ if (!r->path.len) {
+ t->path_base = base->path;
+ if (r->query.len) {
+ t->query = r->query;
+ } else {
+ t->query = base->query;
+ }
+ } else {
+ if (r->path.buf[0] != '/') {
+ t->path_base = base->path;
+ }
+ merge(&t->path_base, &t->path);
+ t->query = r->query;
+ }
+ t->authority = base->authority;
+ }
+ t->scheme = base->scheme;
+ t->fragment = r->fragment;
+ }
}
/** Write the path of `uri` starting at index `i` */
static size_t
write_path_tail(SerdSink sink, void* stream, const SerdURI* uri, size_t i)
{
- size_t len = 0;
- if (i < uri->path_base.len) {
- len += sink(uri->path_base.buf + i, uri->path_base.len - i, stream);
- }
- if (uri->path.buf) {
- if (i < uri->path_base.len) {
- len += sink(uri->path.buf, uri->path.len, stream);
- } else {
- const size_t j = (i - uri->path_base.len);
- len += sink(uri->path.buf + j, uri->path.len - j, stream);
- }
- }
- return len;
+ size_t len = 0;
+ if (i < uri->path_base.len) {
+ len += sink(uri->path_base.buf + i, uri->path_base.len - i, stream);
+ }
+
+ if (uri->path.buf) {
+ if (i < uri->path_base.len) {
+ len += sink(uri->path.buf, uri->path.len, stream);
+ } else {
+ const size_t j = (i - uri->path_base.len);
+ len += sink(uri->path.buf + j, uri->path.len - j, stream);
+ }
+ }
+
+ return len;
}
/** Write the path of `uri` relative to the path of `base`. */
@@ -383,51 +403,51 @@ write_rel_path(SerdSink sink,
const SerdURI* uri,
const SerdURI* base)
{
- const size_t path_len = uri_path_len(uri);
- const size_t base_len = uri_path_len(base);
- const size_t min_len = (path_len < base_len) ? path_len : base_len;
-
- // Find the last separator common to both paths
- size_t last_shared_sep = 0;
- size_t i = 0;
- for (; i < min_len && uri_path_at(uri, i) == uri_path_at(base, i); ++i) {
- if (uri_path_at(uri, i) == '/') {
- last_shared_sep = i;
- }
- }
-
- if (i == path_len && i == base_len) { // Paths are identical
- return 0;
- }
-
- // Find the number of up references ("..") required
- size_t up = 0;
- for (size_t s = last_shared_sep + 1; s < base_len; ++s) {
- if (uri_path_at(base, s) == '/') {
- ++up;
- }
- }
-
- // Write up references
- size_t len = 0;
- for (size_t u = 0; u < up; ++u) {
- len += sink("../", 3, stream);
- }
-
- if (last_shared_sep == 0 && up == 0) {
- len += sink("/", 1, stream);
- }
-
- // Write suffix
- return len + write_path_tail(sink, stream, uri, last_shared_sep + 1);
+ const size_t path_len = uri_path_len(uri);
+ const size_t base_len = uri_path_len(base);
+ const size_t min_len = (path_len < base_len) ? path_len : base_len;
+
+ // Find the last separator common to both paths
+ size_t last_shared_sep = 0;
+ size_t i = 0;
+ for (; i < min_len && uri_path_at(uri, i) == uri_path_at(base, i); ++i) {
+ if (uri_path_at(uri, i) == '/') {
+ last_shared_sep = i;
+ }
+ }
+
+ if (i == path_len && i == base_len) { // Paths are identical
+ return 0;
+ }
+
+ // Find the number of up references ("..") required
+ size_t up = 0;
+ for (size_t s = last_shared_sep + 1; s < base_len; ++s) {
+ if (uri_path_at(base, s) == '/') {
+ ++up;
+ }
+ }
+
+ // Write up references
+ size_t len = 0;
+ for (size_t u = 0; u < up; ++u) {
+ len += sink("../", 3, stream);
+ }
+
+ if (last_shared_sep == 0 && up == 0) {
+ len += sink("/", 1, stream);
+ }
+
+ // Write suffix
+ return len + write_path_tail(sink, stream, uri, last_shared_sep + 1);
}
static uint8_t
serd_uri_path_starts_without_slash(const SerdURI* uri)
{
- return ((uri->path_base.len || uri->path.len) &&
- ((!uri->path_base.len || uri->path_base.buf[0] != '/') &&
- (!uri->path.len || uri->path.buf[0] != '/')));
+ return ((uri->path_base.len || uri->path.len) &&
+ ((!uri->path_base.len || uri->path_base.buf[0] != '/') &&
+ (!uri->path.len || uri->path.buf[0] != '/')));
}
/// See http://tools.ietf.org/html/rfc3986#section-5.3
@@ -438,45 +458,49 @@ serd_uri_serialise_relative(const SerdURI* uri,
SerdSink sink,
void* stream)
{
- size_t len = 0;
- const bool relative =
- root ? uri_is_under(uri, root) : uri_is_related(uri, base);
-
- if (relative) {
- len = write_rel_path(sink, stream, uri, base);
- }
- if (!relative || (!len && base->query.buf)) {
- if (uri->scheme.buf) {
- len += sink(uri->scheme.buf, uri->scheme.len, stream);
- len += sink(":", 1, stream);
- }
- if (uri->authority.buf) {
- len += sink("//", 2, stream);
- len += sink(uri->authority.buf, uri->authority.len, stream);
- if (uri->authority.len > 0 &&
- uri->authority.buf[uri->authority.len - 1] != '/' &&
- serd_uri_path_starts_without_slash(uri)) {
- // Special case: ensure path begins with a slash
- // https://tools.ietf.org/html/rfc3986#section-3.2
- len += sink("/", 1, stream);
- }
- }
- len += write_path_tail(sink, stream, uri, 0);
- }
- if (uri->query.buf) {
- len += sink("?", 1, stream);
- len += sink(uri->query.buf, uri->query.len, stream);
- }
- if (uri->fragment.buf) {
- // Note uri->fragment.buf includes the leading `#'
- len += sink(uri->fragment.buf, uri->fragment.len, stream);
- }
- return len;
+ size_t len = 0;
+ const bool relative =
+ root ? uri_is_under(uri, root) : uri_is_related(uri, base);
+
+ if (relative) {
+ len = write_rel_path(sink, stream, uri, base);
+ }
+
+ if (!relative || (!len && base->query.buf)) {
+ if (uri->scheme.buf) {
+ len += sink(uri->scheme.buf, uri->scheme.len, stream);
+ len += sink(":", 1, stream);
+ }
+ if (uri->authority.buf) {
+ len += sink("//", 2, stream);
+ len += sink(uri->authority.buf, uri->authority.len, stream);
+ if (uri->authority.len > 0 &&
+ uri->authority.buf[uri->authority.len - 1] != '/' &&
+ serd_uri_path_starts_without_slash(uri)) {
+ // Special case: ensure path begins with a slash
+ // https://tools.ietf.org/html/rfc3986#section-3.2
+ len += sink("/", 1, stream);
+ }
+ }
+ len += write_path_tail(sink, stream, uri, 0);
+ }
+
+ if (uri->query.buf) {
+ len += sink("?", 1, stream);
+ len += sink(uri->query.buf, uri->query.len, stream);
+ }
+
+ if (uri->fragment.buf) {
+ // Note uri->fragment.buf includes the leading `#'
+ len += sink(uri->fragment.buf, uri->fragment.len, stream);
+ }
+
+ return len;
}
/// See http://tools.ietf.org/html/rfc3986#section-5.3
size_t
serd_uri_serialise(const SerdURI* uri, SerdSink sink, void* stream)
{
- return serd_uri_serialise_relative(uri, NULL, NULL, sink, stream);
+ return serd_uri_serialise_relative(uri, NULL, NULL, sink, stream);
}
diff --git a/src/uri_utils.h b/src/uri_utils.h
index d7c90b12..2ba9823c 100644
--- a/src/uri_utils.h
+++ b/src/uri_utils.h
@@ -28,21 +28,21 @@
static inline bool
chunk_equals(const SerdChunk* a, const SerdChunk* b)
{
- return a->len == b->len
- && !strncmp((const char*)a->buf, (const char*)b->buf, a->len);
+ return a->len == b->len &&
+ !strncmp((const char*)a->buf, (const char*)b->buf, a->len);
}
static inline size_t
uri_path_len(const SerdURI* uri)
{
- return uri->path_base.len + uri->path.len;
+ return uri->path_base.len + uri->path.len;
}
static inline uint8_t
uri_path_at(const SerdURI* uri, size_t i)
{
- return (i < uri->path_base.len) ? uri->path_base.buf[i]
- : uri->path.buf[i - uri->path_base.len];
+ return (i < uri->path_base.len) ? uri->path_base.buf[i]
+ : uri->path.buf[i - uri->path_base.len];
}
/**
@@ -52,56 +52,59 @@ uri_path_at(const SerdURI* uri, size_t i)
static inline SERD_PURE_FUNC size_t
uri_rooted_index(const SerdURI* uri, const SerdURI* root)
{
- if (!root || !root->scheme.len ||
- !chunk_equals(&root->scheme, &uri->scheme) ||
- !chunk_equals(&root->authority, &uri->authority)) {
- return 0;
- }
-
- bool differ = false;
- const size_t path_len = uri_path_len(uri);
- const size_t root_len = uri_path_len(root);
- size_t last_root_slash = 0;
- for (size_t i = 0; i < path_len && i < root_len; ++i) {
- const uint8_t u = uri_path_at(uri, i);
- const uint8_t r = uri_path_at(root, i);
-
- differ = differ || u != r;
- if (r == '/') {
- last_root_slash = i;
- if (differ) {
- return 0;
- }
- }
- }
-
- return last_root_slash + 1;
+ if (!root || !root->scheme.len ||
+ !chunk_equals(&root->scheme, &uri->scheme) ||
+ !chunk_equals(&root->authority, &uri->authority)) {
+ return 0;
+ }
+
+ bool differ = false;
+ const size_t path_len = uri_path_len(uri);
+ const size_t root_len = uri_path_len(root);
+ size_t last_root_slash = 0;
+ for (size_t i = 0; i < path_len && i < root_len; ++i) {
+ const uint8_t u = uri_path_at(uri, i);
+ const uint8_t r = uri_path_at(root, i);
+
+ differ = differ || u != r;
+ if (r == '/') {
+ last_root_slash = i;
+ if (differ) {
+ return 0;
+ }
+ }
+ }
+
+ return last_root_slash + 1;
}
/** Return true iff `uri` shares path components with `root` */
static inline SERD_PURE_FUNC bool
uri_is_related(const SerdURI* uri, const SerdURI* root)
{
- return uri_rooted_index(uri, root) > 0;
+ return uri_rooted_index(uri, root) > 0;
}
/** Return true iff `uri` is within the base of `root` */
static inline SERD_PURE_FUNC bool
uri_is_under(const SerdURI* uri, const SerdURI* root)
{
- const size_t index = uri_rooted_index(uri, root);
- return index > 0 && uri->path.len > index;
+ const size_t index = uri_rooted_index(uri, root);
+ return index > 0 && uri->path.len > index;
}
static inline bool
is_uri_scheme_char(const int c)
{
- switch (c) {
- case ':': case '+': case '-': case '.':
- return true;
- default:
- return is_alpha(c) || is_digit(c);
- }
+ switch (c) {
+ case ':':
+ case '+':
+ case '-':
+ case '.':
+ return true;
+ default:
+ return is_alpha(c) || is_digit(c);
+ }
}
-#endif // SERD_URI_UTILS_H
+#endif // SERD_URI_UTILS_H
diff --git a/src/writer.c b/src/writer.c
index 8c0d4b3f..386276fe 100644
--- a/src/writer.c
+++ b/src/writer.c
@@ -31,95 +31,88 @@
#include <string.h>
typedef enum {
- FIELD_NONE,
- FIELD_SUBJECT,
- FIELD_PREDICATE,
- FIELD_OBJECT,
- FIELD_GRAPH
+ FIELD_NONE,
+ FIELD_SUBJECT,
+ FIELD_PREDICATE,
+ FIELD_OBJECT,
+ FIELD_GRAPH
} Field;
typedef struct {
- SerdNode graph;
- SerdNode subject;
- SerdNode predicate;
+ SerdNode graph;
+ SerdNode subject;
+ SerdNode predicate;
} WriteContext;
-static const WriteContext WRITE_CONTEXT_NULL = {
- { 0, 0, 0, 0, SERD_NOTHING },
- { 0, 0, 0, 0, SERD_NOTHING },
- { 0, 0, 0, 0, SERD_NOTHING }
-};
+static const WriteContext WRITE_CONTEXT_NULL = {{0, 0, 0, 0, SERD_NOTHING},
+ {0, 0, 0, 0, SERD_NOTHING},
+ {0, 0, 0, 0, SERD_NOTHING}};
typedef enum {
- SEP_NONE,
- SEP_END_S, ///< End of a subject ('.')
- SEP_END_P, ///< End of a predicate (';')
- SEP_END_O, ///< End of an object (',')
- SEP_S_P, ///< Between a subject and predicate (whitespace)
- SEP_P_O, ///< Between a predicate and object (whitespace)
- SEP_ANON_BEGIN, ///< Start of anonymous node ('[')
- SEP_ANON_END, ///< End of anonymous node (']')
- SEP_LIST_BEGIN, ///< Start of list ('(')
- SEP_LIST_SEP, ///< List separator (whitespace)
- SEP_LIST_END, ///< End of list (')')
- SEP_GRAPH_BEGIN, ///< Start of graph ('{')
- SEP_GRAPH_END, ///< End of graph ('}')
- SEP_URI_BEGIN, ///< URI start quote ('<')
- SEP_URI_END ///< URI end quote ('>')
+ SEP_NONE,
+ SEP_END_S, ///< End of a subject ('.')
+ SEP_END_P, ///< End of a predicate (';')
+ SEP_END_O, ///< End of an object (',')
+ SEP_S_P, ///< Between a subject and predicate (whitespace)
+ SEP_P_O, ///< Between a predicate and object (whitespace)
+ SEP_ANON_BEGIN, ///< Start of anonymous node ('[')
+ SEP_ANON_END, ///< End of anonymous node (']')
+ SEP_LIST_BEGIN, ///< Start of list ('(')
+ SEP_LIST_SEP, ///< List separator (whitespace)
+ SEP_LIST_END, ///< End of list (')')
+ SEP_GRAPH_BEGIN, ///< Start of graph ('{')
+ SEP_GRAPH_END, ///< End of graph ('}')
+ SEP_URI_BEGIN, ///< URI start quote ('<')
+ SEP_URI_END ///< URI end quote ('>')
} Sep;
typedef struct {
- const char* str; ///< Sep string
- uint8_t len; ///< Length of sep string
- uint8_t space_before; ///< Newline before sep
- uint8_t space_after_node; ///< Newline after sep if after node
- uint8_t space_after_sep; ///< Newline after sep if after sep
+ const char* str; ///< Sep string
+ uint8_t len; ///< Length of sep string
+ uint8_t space_before; ///< Newline before sep
+ uint8_t space_after_node; ///< Newline after sep if after node
+ uint8_t space_after_sep; ///< Newline after sep if after sep
} SepRule;
-static const SepRule rules[] = {
- { NULL, 0, 0, 0, 0 },
- { " .\n\n", 4, 0, 0, 0 },
- { " ;", 2, 0, 1, 1 },
- { " ,", 2, 0, 1, 0 },
- { NULL, 0, 0, 1, 0 },
- { " ", 1, 0, 0, 0 },
- { "[", 1, 0, 1, 1 },
- { "]", 1, 1, 0, 0 },
- { "(", 1, 0, 0, 0 },
- { NULL, 0, 0, 1, 0 },
- { ")", 1, 1, 0, 0 },
- { " {", 2, 0, 1, 1 },
- { " }", 2, 0, 1, 1 },
- { "<", 1, 0, 0, 0 },
- { ">", 1, 0, 0, 0 },
- { "\n", 1, 0, 1, 0 }
-};
+static const SepRule rules[] = {{NULL, 0, 0, 0, 0},
+ {" .\n\n", 4, 0, 0, 0},
+ {" ;", 2, 0, 1, 1},
+ {" ,", 2, 0, 1, 0},
+ {NULL, 0, 0, 1, 0},
+ {" ", 1, 0, 0, 0},
+ {"[", 1, 0, 1, 1},
+ {"]", 1, 1, 0, 0},
+ {"(", 1, 0, 0, 0},
+ {NULL, 0, 0, 1, 0},
+ {")", 1, 1, 0, 0},
+ {" {", 2, 0, 1, 1},
+ {" }", 2, 0, 1, 1},
+ {"<", 1, 0, 0, 0},
+ {">", 1, 0, 0, 0},
+ {"\n", 1, 0, 1, 0}};
struct SerdWriterImpl {
- SerdSyntax syntax;
- SerdStyle style;
- SerdEnv* env;
- SerdNode root_node;
- SerdURI root_uri;
- SerdURI base_uri;
- SerdStack anon_stack;
- SerdByteSink byte_sink;
- SerdErrorSink error_sink;
- void* error_handle;
- WriteContext context;
- SerdNode list_subj;
- unsigned list_depth;
- unsigned indent;
- uint8_t* bprefix;
- size_t bprefix_len;
- Sep last_sep;
- bool empty;
+ SerdSyntax syntax;
+ SerdStyle style;
+ SerdEnv* env;
+ SerdNode root_node;
+ SerdURI root_uri;
+ SerdURI base_uri;
+ SerdStack anon_stack;
+ SerdByteSink byte_sink;
+ SerdErrorSink error_sink;
+ void* error_handle;
+ WriteContext context;
+ SerdNode list_subj;
+ unsigned list_depth;
+ unsigned indent;
+ uint8_t* bprefix;
+ size_t bprefix_len;
+ Sep last_sep;
+ bool empty;
};
-typedef enum {
- WRITE_STRING,
- WRITE_LONG_STRING
-} TextContext;
+typedef enum { WRITE_STRING, WRITE_LONG_STRING } TextContext;
static bool
write_node(SerdWriter* writer,
@@ -132,58 +125,58 @@ write_node(SerdWriter* writer,
static bool
supports_abbrev(const SerdWriter* writer)
{
- return writer->syntax == SERD_TURTLE || writer->syntax == SERD_TRIG;
+ return writer->syntax == SERD_TURTLE || writer->syntax == SERD_TRIG;
}
static bool
supports_uriref(const SerdWriter* writer)
{
- return writer->syntax == SERD_TURTLE || writer->syntax == SERD_TRIG;
+ return writer->syntax == SERD_TURTLE || writer->syntax == SERD_TRIG;
}
static void
w_err(SerdWriter* writer, SerdStatus st, const char* fmt, ...)
{
- /* TODO: This results in errors with no file information, which is not
- helpful when re-serializing a file (particularly for "undefined
- namespace prefix" errors. The statement sink API needs to be changed to
- add a Cursor parameter so the source can notify the writer of the
- statement origin for better error reporting. */
-
- va_list args;
- va_start(args, fmt);
- const SerdError e = { st, (const uint8_t*)"", 0, 0, fmt, &args };
- serd_error(writer->error_sink, writer->error_handle, &e);
- va_end(args);
+ /* TODO: This results in errors with no file information, which is not
+ helpful when re-serializing a file (particularly for "undefined
+ namespace prefix" errors. The statement sink API needs to be changed to
+ add a Cursor parameter so the source can notify the writer of the
+ statement origin for better error reporting. */
+
+ va_list args;
+ va_start(args, fmt);
+ const SerdError e = {st, (const uint8_t*)"", 0, 0, fmt, &args};
+ serd_error(writer->error_sink, writer->error_handle, &e);
+ va_end(args);
}
static inline WriteContext*
anon_stack_top(SerdWriter* writer)
{
- assert(!serd_stack_is_empty(&writer->anon_stack));
- return (WriteContext*)(writer->anon_stack.buf
- + writer->anon_stack.size - sizeof(WriteContext));
+ assert(!serd_stack_is_empty(&writer->anon_stack));
+ return (WriteContext*)(writer->anon_stack.buf + writer->anon_stack.size -
+ sizeof(WriteContext));
}
static void
copy_node(SerdNode* dst, const SerdNode* src)
{
- if (src) {
- dst->buf = (uint8_t*)realloc((char*)dst->buf, src->n_bytes + 1);
- dst->n_bytes = src->n_bytes;
- dst->n_chars = src->n_chars;
- dst->flags = src->flags;
- dst->type = src->type;
- memcpy((char*)dst->buf, src->buf, src->n_bytes + 1);
- } else {
- dst->type = SERD_NOTHING;
- }
+ if (src) {
+ dst->buf = (uint8_t*)realloc((char*)dst->buf, src->n_bytes + 1);
+ dst->n_bytes = src->n_bytes;
+ dst->n_chars = src->n_chars;
+ dst->flags = src->flags;
+ dst->type = src->type;
+ memcpy((char*)dst->buf, src->buf, src->n_bytes + 1);
+ } else {
+ dst->type = SERD_NOTHING;
+ }
}
static inline size_t
sink(const void* buf, size_t len, SerdWriter* writer)
{
- return serd_byte_sink_write(buf, len, &writer->byte_sink);
+ return serd_byte_sink_write(buf, len, &writer->byte_sink);
}
// Write a single character, as an escape for single byte characters
@@ -191,253 +184,310 @@ sink(const void* buf, size_t len, SerdWriter* writer)
static size_t
write_character(SerdWriter* writer, const uint8_t* utf8, size_t* size)
{
- char escape[11] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
- const uint32_t c = parse_utf8_char(utf8, size);
- switch (*size) {
- case 0:
- w_err(writer, SERD_ERR_BAD_ARG, "invalid UTF-8: %X\n", utf8[0]);
- return sink(replacement_char, sizeof(replacement_char), writer);
- case 1:
- snprintf(escape, sizeof(escape), "\\u%04X", utf8[0]);
- return sink(escape, 6, writer);
- default:
- break;
- }
-
- if (!(writer->style & SERD_STYLE_ASCII)) {
- // Write UTF-8 character directly to UTF-8 output
- return sink(utf8, *size, writer);
- }
-
- if (c <= 0xFFFF) {
- snprintf(escape, sizeof(escape), "\\u%04X", c);
- return sink(escape, 6, writer);
- }
-
- snprintf(escape, sizeof(escape), "\\U%08X", c);
- return sink(escape, 10, writer);
+ char escape[11] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
+ const uint32_t c = parse_utf8_char(utf8, size);
+ switch (*size) {
+ case 0:
+ w_err(writer, SERD_ERR_BAD_ARG, "invalid UTF-8: %X\n", utf8[0]);
+ return sink(replacement_char, sizeof(replacement_char), writer);
+ case 1:
+ snprintf(escape, sizeof(escape), "\\u%04X", utf8[0]);
+ return sink(escape, 6, writer);
+ default:
+ break;
+ }
+
+ if (!(writer->style & SERD_STYLE_ASCII)) {
+ // Write UTF-8 character directly to UTF-8 output
+ return sink(utf8, *size, writer);
+ }
+
+ if (c <= 0xFFFF) {
+ snprintf(escape, sizeof(escape), "\\u%04X", c);
+ return sink(escape, 6, writer);
+ }
+
+ snprintf(escape, sizeof(escape), "\\U%08X", c);
+ return sink(escape, 10, writer);
}
static inline bool
uri_must_escape(const uint8_t c)
{
- switch (c) {
- case ' ': case '"': case '<': case '>': case '\\':
- case '^': case '`': case '{': case '|': case '}':
- return true;
- default:
- return !in_range(c, 0x20, 0x7E);
- }
+ switch (c) {
+ case ' ':
+ case '"':
+ case '<':
+ case '>':
+ case '\\':
+ case '^':
+ case '`':
+ case '{':
+ case '|':
+ case '}':
+ return true;
+ default:
+ return !in_range(c, 0x20, 0x7E);
+ }
}
static size_t
write_uri(SerdWriter* writer, const uint8_t* utf8, size_t n_bytes)
{
- size_t len = 0;
- for (size_t i = 0; i < n_bytes;) {
- size_t j = i; // Index of next character that must be escaped
- for (; j < n_bytes; ++j) {
- if (uri_must_escape(utf8[j])) {
- break;
- }
- }
-
- // Bulk write all characters up to this special one
- len += sink(&utf8[i], j - i, writer);
- if ((i = j) == n_bytes) {
- break; // Reached end
- }
-
- // Write UTF-8 character
- size_t size = 0;
- len += write_character(writer, utf8 + i, &size);
- i += size;
- if (size == 0) {
- // Corrupt input, scan to start of next character
- for (++i; i < n_bytes && (utf8[i] & 0x80); ++i) {}
- }
- }
- return len;
+ size_t len = 0;
+ for (size_t i = 0; i < n_bytes;) {
+ size_t j = i; // Index of next character that must be escaped
+ for (; j < n_bytes; ++j) {
+ if (uri_must_escape(utf8[j])) {
+ break;
+ }
+ }
+
+ // Bulk write all characters up to this special one
+ len += sink(&utf8[i], j - i, writer);
+ if ((i = j) == n_bytes) {
+ break; // Reached end
+ }
+
+ // Write UTF-8 character
+ size_t size = 0;
+ len += write_character(writer, utf8 + i, &size);
+ i += size;
+ if (size == 0) {
+ // Corrupt input, scan to start of next character
+ for (++i; i < n_bytes && (utf8[i] & 0x80); ++i) {
+ }
+ }
+ }
+
+ return len;
}
static bool
lname_must_escape(const uint8_t c)
{
- /* This arbitrary list of characters, most of which have nothing to do with
- Turtle, must be handled as special cases here because the RDF and SPARQL
- WGs are apparently intent on making the once elegant Turtle a baroque
- and inconsistent mess, throwing elegance and extensibility completely
- out the window for no good reason.
-
- Note '-', '.', and '_' are also in PN_LOCAL_ESC, but are valid unescaped
- in local names, so they are not escaped here. */
-
- switch (c) {
- case '\'': case '!': case '#': case '$': case '%': case '&':
- case '(': case ')': case '*': case '+': case ',': case '/':
- case ';': case '=': case '?': case '@': case '~':
- return true;
- default:
- break;
- }
- return false;
+ /* This arbitrary list of characters, most of which have nothing to do with
+ Turtle, must be handled as special cases here because the RDF and SPARQL
+ WGs are apparently intent on making the once elegant Turtle a baroque
+ and inconsistent mess, throwing elegance and extensibility completely
+ out the window for no good reason.
+
+ Note '-', '.', and '_' are also in PN_LOCAL_ESC, but are valid unescaped
+ in local names, so they are not escaped here. */
+
+ switch (c) {
+ case '\'':
+ case '!':
+ case '#':
+ case '$':
+ case '%':
+ case '&':
+ case '(':
+ case ')':
+ case '*':
+ case '+':
+ case ',':
+ case '/':
+ case ';':
+ case '=':
+ case '?':
+ case '@':
+ case '~':
+ return true;
+ default:
+ break;
+ }
+ return false;
}
static size_t
write_lname(SerdWriter* writer, const uint8_t* utf8, size_t n_bytes)
{
- size_t len = 0;
- for (size_t i = 0; i < n_bytes; ++i) {
- size_t j = i; // Index of next character that must be escaped
- for (; j < n_bytes; ++j) {
- if (lname_must_escape(utf8[j])) {
- break;
- }
- }
-
- // Bulk write all characters up to this special one
- len += sink(&utf8[i], j - i, writer);
- if ((i = j) == n_bytes) {
- break; // Reached end
- }
-
- // Write escape
- len += sink("\\", 1, writer);
- len += sink(&utf8[i], 1, writer);
- }
- return len;
+ size_t len = 0;
+ for (size_t i = 0; i < n_bytes; ++i) {
+ size_t j = i; // Index of next character that must be escaped
+ for (; j < n_bytes; ++j) {
+ if (lname_must_escape(utf8[j])) {
+ break;
+ }
+ }
+
+ // Bulk write all characters up to this special one
+ len += sink(&utf8[i], j - i, writer);
+ if ((i = j) == n_bytes) {
+ break; // Reached end
+ }
+
+ // Write escape
+ len += sink("\\", 1, writer);
+ len += sink(&utf8[i], 1, writer);
+ }
+
+ return len;
}
static size_t
-write_text(SerdWriter* writer, TextContext ctx,
- const uint8_t* utf8, size_t n_bytes)
+write_text(SerdWriter* writer,
+ TextContext ctx,
+ const uint8_t* utf8,
+ size_t n_bytes)
{
- size_t len = 0;
- for (size_t i = 0; i < n_bytes;) {
- // Fast bulk write for long strings of printable ASCII
- size_t j = i;
- for (; j < n_bytes; ++j) {
- if (utf8[j] == '\\' || utf8[j] == '"'
- || (!in_range(utf8[j], 0x20, 0x7E))) {
- break;
- }
- }
-
- len += sink(&utf8[i], j - i, writer);
- if ((i = j) == n_bytes) {
- break; // Reached end
- }
-
- const uint8_t in = utf8[i++];
- if (ctx == WRITE_LONG_STRING) {
- switch (in) {
- case '\\': len += sink("\\\\", 2, writer); continue;
- case '\b': len += sink("\\b", 2, writer); continue;
- case '\n': case '\r': case '\t': case '\f':
- len += sink(&in, 1, writer); // Write character as-is
- continue;
- case '\"':
- if (i == n_bytes) { // '"' at string end
- len += sink("\\\"", 2, writer);
- } else {
- len += sink(&in, 1, writer);
- }
- continue;
- default: break;
- }
- } else if (ctx == WRITE_STRING) {
- switch (in) {
- case '\\': len += sink("\\\\", 2, writer); continue;
- case '\n': len += sink("\\n", 2, writer); continue;
- case '\r': len += sink("\\r", 2, writer); continue;
- case '\t': len += sink("\\t", 2, writer); continue;
- case '"': len += sink("\\\"", 2, writer); continue;
- default: break;
- }
- if (writer->syntax == SERD_TURTLE) {
- switch (in) {
- case '\b': len += sink("\\b", 2, writer); continue;
- case '\f': len += sink("\\f", 2, writer); continue;
- default: break;
- }
- }
- }
-
- // Write UTF-8 character
- size_t size = 0;
- len += write_character(writer, utf8 + i - 1, &size);
- if (size == 0) {
- // Corrupt input, scan to start of next character
- for (; i < n_bytes && (utf8[i] & 0x80); ++i) {}
- } else {
- i += size - 1;
- }
- }
- return len;
+ size_t len = 0;
+ for (size_t i = 0; i < n_bytes;) {
+ // Fast bulk write for long strings of printable ASCII
+ size_t j = i;
+ for (; j < n_bytes; ++j) {
+ if (utf8[j] == '\\' || utf8[j] == '"' ||
+ (!in_range(utf8[j], 0x20, 0x7E))) {
+ break;
+ }
+ }
+
+ len += sink(&utf8[i], j - i, writer);
+ if ((i = j) == n_bytes) {
+ break; // Reached end
+ }
+
+ const uint8_t in = utf8[i++];
+ if (ctx == WRITE_LONG_STRING) {
+ switch (in) {
+ case '\\':
+ len += sink("\\\\", 2, writer);
+ continue;
+ case '\b':
+ len += sink("\\b", 2, writer);
+ continue;
+ case '\n':
+ case '\r':
+ case '\t':
+ case '\f':
+ len += sink(&in, 1, writer); // Write character as-is
+ continue;
+ case '\"':
+ if (i == n_bytes) { // '"' at string end
+ len += sink("\\\"", 2, writer);
+ } else {
+ len += sink(&in, 1, writer);
+ }
+ continue;
+ default:
+ break;
+ }
+ } else if (ctx == WRITE_STRING) {
+ switch (in) {
+ case '\\':
+ len += sink("\\\\", 2, writer);
+ continue;
+ case '\n':
+ len += sink("\\n", 2, writer);
+ continue;
+ case '\r':
+ len += sink("\\r", 2, writer);
+ continue;
+ case '\t':
+ len += sink("\\t", 2, writer);
+ continue;
+ case '"':
+ len += sink("\\\"", 2, writer);
+ continue;
+ default:
+ break;
+ }
+ if (writer->syntax == SERD_TURTLE) {
+ switch (in) {
+ case '\b':
+ len += sink("\\b", 2, writer);
+ continue;
+ case '\f':
+ len += sink("\\f", 2, writer);
+ continue;
+ default:
+ break;
+ }
+ }
+ }
+
+ // Write UTF-8 character
+ size_t size = 0;
+ len += write_character(writer, utf8 + i - 1, &size);
+ if (size == 0) {
+ // Corrupt input, scan to start of next character
+ for (; i < n_bytes && (utf8[i] & 0x80); ++i) {
+ }
+ } else {
+ i += size - 1;
+ }
+ }
+
+ return len;
}
static size_t
uri_sink(const void* buf, size_t len, void* stream)
{
- return write_uri((SerdWriter*)stream, (const uint8_t*)buf, len);
+ return write_uri((SerdWriter*)stream, (const uint8_t*)buf, len);
}
static void
write_newline(SerdWriter* writer)
{
- sink("\n", 1, writer);
- for (unsigned i = 0; i < writer->indent; ++i) {
- sink("\t", 1, writer);
- }
+ sink("\n", 1, writer);
+ for (unsigned i = 0; i < writer->indent; ++i) {
+ sink("\t", 1, writer);
+ }
}
static bool
write_sep(SerdWriter* writer, const Sep sep)
{
- const SepRule* rule = &rules[sep];
- if (rule->space_before) {
- write_newline(writer);
- }
- if (rule->str) {
- sink(rule->str, rule->len, writer);
- }
- if ((writer->last_sep && rule->space_after_sep) ||
- (!writer->last_sep && rule->space_after_node)) {
- write_newline(writer);
- } else if (writer->last_sep && rule->space_after_node) {
- sink(" ", 1, writer);
- }
- writer->last_sep = sep;
- return true;
+ const SepRule* rule = &rules[sep];
+ if (rule->space_before) {
+ write_newline(writer);
+ }
+
+ if (rule->str) {
+ sink(rule->str, rule->len, writer);
+ }
+
+ if ((writer->last_sep && rule->space_after_sep) ||
+ (!writer->last_sep && rule->space_after_node)) {
+ write_newline(writer);
+ } else if (writer->last_sep && rule->space_after_node) {
+ sink(" ", 1, writer);
+ }
+
+ writer->last_sep = sep;
+ return true;
}
static SerdStatus
reset_context(SerdWriter* writer, bool graph)
{
- if (graph) {
- writer->context.graph.type = SERD_NOTHING;
- }
- writer->context.subject.type = SERD_NOTHING;
- writer->context.predicate.type = SERD_NOTHING;
- writer->empty = false;
- return SERD_SUCCESS;
+ if (graph) {
+ writer->context.graph.type = SERD_NOTHING;
+ }
+
+ writer->context.subject.type = SERD_NOTHING;
+ writer->context.predicate.type = SERD_NOTHING;
+ writer->empty = false;
+ return SERD_SUCCESS;
}
static SerdStatus
free_context(SerdWriter* writer)
{
- serd_node_free(&writer->context.graph);
- serd_node_free(&writer->context.subject);
- serd_node_free(&writer->context.predicate);
- return reset_context(writer, true);
+ serd_node_free(&writer->context.graph);
+ serd_node_free(&writer->context.subject);
+ serd_node_free(&writer->context.predicate);
+ return reset_context(writer, true);
}
static bool
is_inline_start(const SerdWriter* writer, Field field, SerdStatementFlags flags)
{
- return (supports_abbrev(writer) &&
- ((field == FIELD_SUBJECT && (flags & SERD_ANON_S_BEGIN)) ||
- (field == FIELD_OBJECT && (flags & SERD_ANON_O_BEGIN))));
+ return (supports_abbrev(writer) &&
+ ((field == FIELD_SUBJECT && (flags & SERD_ANON_S_BEGIN)) ||
+ (field == FIELD_OBJECT && (flags & SERD_ANON_O_BEGIN))));
}
static bool
@@ -447,59 +497,60 @@ write_literal(SerdWriter* writer,
const SerdNode* lang,
SerdStatementFlags flags)
{
- if (supports_abbrev(writer) && datatype && datatype->buf) {
- const char* type_uri = (const char*)datatype->buf;
- if (!strncmp(type_uri, NS_XSD, sizeof(NS_XSD) - 1) && (
- !strcmp(type_uri + sizeof(NS_XSD) - 1, "boolean") ||
- !strcmp(type_uri + sizeof(NS_XSD) - 1, "integer"))) {
- sink(node->buf, node->n_bytes, writer);
- return true;
- }
-
- if (!strncmp(type_uri, NS_XSD, sizeof(NS_XSD) - 1) &&
- !strcmp(type_uri + sizeof(NS_XSD) - 1, "decimal") &&
- strchr((const char*)node->buf, '.') &&
- node->buf[node->n_bytes - 1] != '.') {
- /* xsd:decimal literals without trailing digits, e.g. "5.", can
- not be written bare in Turtle. We could add a 0 which is
- prettier, but changes the text and breaks round tripping.
- */
- sink(node->buf, node->n_bytes, writer);
- return true;
- }
- }
-
- if (supports_abbrev(writer)
- && (node->flags & (SERD_HAS_NEWLINE|SERD_HAS_QUOTE))) {
- sink("\"\"\"", 3, writer);
- write_text(writer, WRITE_LONG_STRING, node->buf, node->n_bytes);
- sink("\"\"\"", 3, writer);
- } else {
- sink("\"", 1, writer);
- write_text(writer, WRITE_STRING, node->buf, node->n_bytes);
- sink("\"", 1, writer);
- }
- if (lang && lang->buf) {
- sink("@", 1, writer);
- sink(lang->buf, lang->n_bytes, writer);
- } else if (datatype && datatype->buf) {
- sink("^^", 2, writer);
- return write_node(writer, datatype, NULL, NULL, FIELD_NONE, flags);
- }
- return true;
+ if (supports_abbrev(writer) && datatype && datatype->buf) {
+ const char* type_uri = (const char*)datatype->buf;
+ if (!strncmp(type_uri, NS_XSD, sizeof(NS_XSD) - 1) &&
+ (!strcmp(type_uri + sizeof(NS_XSD) - 1, "boolean") ||
+ !strcmp(type_uri + sizeof(NS_XSD) - 1, "integer"))) {
+ sink(node->buf, node->n_bytes, writer);
+ return true;
+ }
+
+ if (!strncmp(type_uri, NS_XSD, sizeof(NS_XSD) - 1) &&
+ !strcmp(type_uri + sizeof(NS_XSD) - 1, "decimal") &&
+ strchr((const char*)node->buf, '.') &&
+ node->buf[node->n_bytes - 1] != '.') {
+ /* xsd:decimal literals without trailing digits, e.g. "5.", can
+ not be written bare in Turtle. We could add a 0 which is
+ prettier, but changes the text and breaks round tripping.
+ */
+ sink(node->buf, node->n_bytes, writer);
+ return true;
+ }
+ }
+
+ if (supports_abbrev(writer) &&
+ (node->flags & (SERD_HAS_NEWLINE | SERD_HAS_QUOTE))) {
+ sink("\"\"\"", 3, writer);
+ write_text(writer, WRITE_LONG_STRING, node->buf, node->n_bytes);
+ sink("\"\"\"", 3, writer);
+ } else {
+ sink("\"", 1, writer);
+ write_text(writer, WRITE_STRING, node->buf, node->n_bytes);
+ sink("\"", 1, writer);
+ }
+ if (lang && lang->buf) {
+ sink("@", 1, writer);
+ sink(lang->buf, lang->n_bytes, writer);
+ } else if (datatype && datatype->buf) {
+ sink("^^", 2, writer);
+ return write_node(writer, datatype, NULL, NULL, FIELD_NONE, flags);
+ }
+ return true;
}
// Return true iff `buf` is a valid prefixed name suffix
static inline bool
is_name(const uint8_t* buf, const size_t len)
{
- // TODO: This is more strict than it should be.
- for (size_t i = 0; i < len; ++i) {
- if (!(is_alpha(buf[i]) || is_digit(buf[i]))) {
- return false;
- }
- }
- return true;
+ // TODO: This is more strict than it should be
+ for (size_t i = 0; i < len; ++i) {
+ if (!(is_alpha(buf[i]) || is_digit(buf[i]))) {
+ return false;
+ }
+ }
+
+ return true;
}
static bool
@@ -508,72 +559,73 @@ write_uri_node(SerdWriter* const writer,
const Field field,
const SerdStatementFlags flags)
{
- SerdNode prefix;
- SerdChunk suffix;
-
- if (is_inline_start(writer, field, flags)) {
- ++writer->indent;
- write_sep(writer, SEP_ANON_BEGIN);
- sink("== ", 3, writer);
- }
-
- const bool has_scheme = serd_uri_string_has_scheme(node->buf);
- if (supports_abbrev(writer)) {
- if (field == FIELD_PREDICATE &&
- !strcmp((const char*)node->buf, NS_RDF "type")) {
- return sink("a", 1, writer) == 1;
- }
-
- if (!strcmp((const char*)node->buf, NS_RDF "nil")) {
- return sink("()", 2, writer) == 2;
- }
-
- if (has_scheme && (writer->style & SERD_STYLE_CURIED) &&
- serd_env_qualify(writer->env, node, &prefix, &suffix) &&
- is_name(suffix.buf, suffix.len)) {
- write_uri(writer, prefix.buf, prefix.n_bytes);
- sink(":", 1, writer);
- write_uri(writer, suffix.buf, suffix.len);
- return true;
- }
- }
-
- if (!has_scheme && !supports_uriref(writer) &&
- !serd_env_get_base_uri(writer->env, NULL)->buf) {
- w_err(writer,
- SERD_ERR_BAD_ARG,
- "syntax does not support URI reference <%s>\n",
- node->buf);
- return false;
- }
-
- write_sep(writer, SEP_URI_BEGIN);
- if (writer->style & SERD_STYLE_RESOLVED) {
- SerdURI in_base_uri;
- SerdURI uri;
- SerdURI abs_uri;
- serd_env_get_base_uri(writer->env, &in_base_uri);
- serd_uri_parse(node->buf, &uri);
- serd_uri_resolve(&uri, &in_base_uri, &abs_uri);
- bool rooted = uri_is_under(&writer->base_uri, &writer->root_uri);
- SerdURI* root = rooted ? &writer->root_uri : & writer->base_uri;
- if (!uri_is_under(&abs_uri, root) ||
- writer->syntax == SERD_NTRIPLES ||
- writer->syntax == SERD_NQUADS) {
- serd_uri_serialise(&abs_uri, uri_sink, writer);
- } else {
- serd_uri_serialise_relative(
- &uri, &writer->base_uri, root, uri_sink, writer);
- }
- } else {
- write_uri(writer, node->buf, node->n_bytes);
- }
- write_sep(writer, SEP_URI_END);
- if (is_inline_start(writer, field, flags)) {
- sink(" ;", 2, writer);
- write_newline(writer);
- }
- return true;
+ SerdNode prefix;
+ SerdChunk suffix;
+
+ if (is_inline_start(writer, field, flags)) {
+ ++writer->indent;
+ write_sep(writer, SEP_ANON_BEGIN);
+ sink("== ", 3, writer);
+ }
+
+ const bool has_scheme = serd_uri_string_has_scheme(node->buf);
+ if (supports_abbrev(writer)) {
+ if (field == FIELD_PREDICATE &&
+ !strcmp((const char*)node->buf, NS_RDF "type")) {
+ return sink("a", 1, writer) == 1;
+ }
+
+ if (!strcmp((const char*)node->buf, NS_RDF "nil")) {
+ return sink("()", 2, writer) == 2;
+ }
+
+ if (has_scheme && (writer->style & SERD_STYLE_CURIED) &&
+ serd_env_qualify(writer->env, node, &prefix, &suffix) &&
+ is_name(suffix.buf, suffix.len)) {
+ write_uri(writer, prefix.buf, prefix.n_bytes);
+ sink(":", 1, writer);
+ write_uri(writer, suffix.buf, suffix.len);
+ return true;
+ }
+ }
+
+ if (!has_scheme && !supports_uriref(writer) &&
+ !serd_env_get_base_uri(writer->env, NULL)->buf) {
+ w_err(writer,
+ SERD_ERR_BAD_ARG,
+ "syntax does not support URI reference <%s>\n",
+ node->buf);
+ return false;
+ }
+
+ write_sep(writer, SEP_URI_BEGIN);
+ if (writer->style & SERD_STYLE_RESOLVED) {
+ SerdURI in_base_uri;
+ SerdURI uri;
+ SerdURI abs_uri;
+ serd_env_get_base_uri(writer->env, &in_base_uri);
+ serd_uri_parse(node->buf, &uri);
+ serd_uri_resolve(&uri, &in_base_uri, &abs_uri);
+ bool rooted = uri_is_under(&writer->base_uri, &writer->root_uri);
+ SerdURI* root = rooted ? &writer->root_uri : &writer->base_uri;
+ if (!uri_is_under(&abs_uri, root) || writer->syntax == SERD_NTRIPLES ||
+ writer->syntax == SERD_NQUADS) {
+ serd_uri_serialise(&abs_uri, uri_sink, writer);
+ } else {
+ serd_uri_serialise_relative(
+ &uri, &writer->base_uri, root, uri_sink, writer);
+ }
+ } else {
+ write_uri(writer, node->buf, node->n_bytes);
+ }
+
+ write_sep(writer, SEP_URI_END);
+ if (is_inline_start(writer, field, flags)) {
+ sink(" ;", 2, writer);
+ write_newline(writer);
+ }
+
+ return true;
}
static bool
@@ -582,36 +634,37 @@ write_curie(SerdWriter* const writer,
const Field field,
const SerdStatementFlags flags)
{
- SerdChunk prefix = {NULL, 0};
- SerdChunk suffix = {NULL, 0};
- SerdStatus st = SERD_SUCCESS;
-
- switch (writer->syntax) {
- case SERD_NTRIPLES:
- case SERD_NQUADS:
- if ((st = serd_env_expand(writer->env, node, &prefix, &suffix))) {
- w_err(writer, st, "undefined namespace prefix `%s'\n", node->buf);
- return false;
- }
- write_sep(writer, SEP_URI_BEGIN);
- write_uri(writer, prefix.buf, prefix.len);
- write_uri(writer, suffix.buf, suffix.len);
- write_sep(writer, SEP_URI_END);
- break;
- case SERD_TURTLE:
- case SERD_TRIG:
- if (is_inline_start(writer, field, flags)) {
- ++writer->indent;
- write_sep(writer, SEP_ANON_BEGIN);
- sink("== ", 3, writer);
- }
- write_lname(writer, node->buf, node->n_bytes);
- if (is_inline_start(writer, field, flags)) {
- sink(" ;", 2, writer);
- write_newline(writer);
- }
- }
- return true;
+ SerdChunk prefix = {NULL, 0};
+ SerdChunk suffix = {NULL, 0};
+ SerdStatus st = SERD_SUCCESS;
+
+ switch (writer->syntax) {
+ case SERD_NTRIPLES:
+ case SERD_NQUADS:
+ if ((st = serd_env_expand(writer->env, node, &prefix, &suffix))) {
+ w_err(writer, st, "undefined namespace prefix `%s'\n", node->buf);
+ return false;
+ }
+ write_sep(writer, SEP_URI_BEGIN);
+ write_uri(writer, prefix.buf, prefix.len);
+ write_uri(writer, suffix.buf, suffix.len);
+ write_sep(writer, SEP_URI_END);
+ break;
+ case SERD_TURTLE:
+ case SERD_TRIG:
+ if (is_inline_start(writer, field, flags)) {
+ ++writer->indent;
+ write_sep(writer, SEP_ANON_BEGIN);
+ sink("== ", 3, writer);
+ }
+ write_lname(writer, node->buf, node->n_bytes);
+ if (is_inline_start(writer, field, flags)) {
+ sink(" ;", 2, writer);
+ write_newline(writer);
+ }
+ }
+
+ return true;
}
static bool
@@ -620,44 +673,44 @@ write_blank(SerdWriter* const writer,
const Field field,
const SerdStatementFlags flags)
{
- if (supports_abbrev(writer)) {
- if (is_inline_start(writer, field, flags)) {
- ++writer->indent;
- return write_sep(writer, SEP_ANON_BEGIN);
- }
-
- if (field == FIELD_SUBJECT && (flags & SERD_LIST_S_BEGIN)) {
- assert(writer->list_depth == 0);
- copy_node(&writer->list_subj, node);
- ++writer->list_depth;
- ++writer->indent;
- return write_sep(writer, SEP_LIST_BEGIN);
- }
-
- if (field == FIELD_OBJECT && (flags & SERD_LIST_O_BEGIN)) {
- ++writer->indent;
- ++writer->list_depth;
- return write_sep(writer, SEP_LIST_BEGIN);
- }
-
- if ((field == FIELD_SUBJECT && (flags & SERD_EMPTY_S)) ||
- (field == FIELD_OBJECT && (flags & SERD_EMPTY_O))) {
- return sink("[]", 2, writer) == 2;
- }
- }
-
- sink("_:", 2, writer);
- if (writer->bprefix && !strncmp((const char*)node->buf,
- (const char*)writer->bprefix,
- writer->bprefix_len)) {
- sink(node->buf + writer->bprefix_len,
- node->n_bytes - writer->bprefix_len,
- writer);
- } else {
- sink(node->buf, node->n_bytes, writer);
- }
-
- return true;
+ if (supports_abbrev(writer)) {
+ if (is_inline_start(writer, field, flags)) {
+ ++writer->indent;
+ return write_sep(writer, SEP_ANON_BEGIN);
+ }
+
+ if (field == FIELD_SUBJECT && (flags & SERD_LIST_S_BEGIN)) {
+ assert(writer->list_depth == 0);
+ copy_node(&writer->list_subj, node);
+ ++writer->list_depth;
+ ++writer->indent;
+ return write_sep(writer, SEP_LIST_BEGIN);
+ }
+
+ if (field == FIELD_OBJECT && (flags & SERD_LIST_O_BEGIN)) {
+ ++writer->indent;
+ ++writer->list_depth;
+ return write_sep(writer, SEP_LIST_BEGIN);
+ }
+
+ if ((field == FIELD_SUBJECT && (flags & SERD_EMPTY_S)) ||
+ (field == FIELD_OBJECT && (flags & SERD_EMPTY_O))) {
+ return sink("[]", 2, writer) == 2;
+ }
+ }
+
+ sink("_:", 2, writer);
+ if (writer->bprefix && !strncmp((const char*)node->buf,
+ (const char*)writer->bprefix,
+ writer->bprefix_len)) {
+ sink(node->buf + writer->bprefix_len,
+ node->n_bytes - writer->bprefix_len,
+ writer);
+ } else {
+ sink(node->buf, node->n_bytes, writer);
+ }
+
+ return true;
}
static bool
@@ -668,39 +721,40 @@ write_node(SerdWriter* writer,
Field field,
SerdStatementFlags flags)
{
- bool ret = false;
- switch (node->type) {
- case SERD_NOTHING:
- break;
- case SERD_LITERAL:
- ret = write_literal(writer, node, datatype, lang, flags);
- break;
- case SERD_URI:
- ret = write_uri_node(writer, node, field, flags);
- break;
- case SERD_CURIE:
- ret = write_curie(writer, node, field, flags);
- break;
- case SERD_BLANK:
- ret = write_blank(writer, node, field, flags);
- break;
- }
- writer->last_sep = SEP_NONE;
- return ret;
+ bool ret = false;
+ switch (node->type) {
+ case SERD_NOTHING:
+ break;
+ case SERD_LITERAL:
+ ret = write_literal(writer, node, datatype, lang, flags);
+ break;
+ case SERD_URI:
+ ret = write_uri_node(writer, node, field, flags);
+ break;
+ case SERD_CURIE:
+ ret = write_curie(writer, node, field, flags);
+ break;
+ case SERD_BLANK:
+ ret = write_blank(writer, node, field, flags);
+ break;
+ }
+
+ writer->last_sep = SEP_NONE;
+ return ret;
}
static inline bool
is_resource(const SerdNode* node)
{
- return node && node->buf && node->type > SERD_LITERAL;
+ return node && node->buf && node->type > SERD_LITERAL;
}
static void
write_pred(SerdWriter* writer, SerdStatementFlags flags, const SerdNode* pred)
{
- write_node(writer, pred, NULL, NULL, FIELD_PREDICATE, flags);
- write_sep(writer, SEP_P_O);
- copy_node(&writer->context.predicate, pred);
+ write_node(writer, pred, NULL, NULL, FIELD_PREDICATE, flags);
+ write_sep(writer, SEP_P_O);
+ copy_node(&writer->context.predicate, pred);
}
static bool
@@ -711,18 +765,18 @@ write_list_obj(SerdWriter* writer,
const SerdNode* datatype,
const SerdNode* lang)
{
- if (!strcmp((const char*)object->buf, NS_RDF "nil")) {
- --writer->indent;
- write_sep(writer, SEP_LIST_END);
- return true;
- }
-
- if (!strcmp((const char*)predicate->buf, NS_RDF "first")) {
- write_sep(writer, SEP_LIST_SEP);
- write_node(writer, object, datatype, lang, FIELD_OBJECT, flags);
- }
-
- return false;
+ if (!strcmp((const char*)object->buf, NS_RDF "nil")) {
+ --writer->indent;
+ write_sep(writer, SEP_LIST_END);
+ return true;
+ }
+
+ if (!strcmp((const char*)predicate->buf, NS_RDF "first")) {
+ write_sep(writer, SEP_LIST_SEP);
+ write_node(writer, object, datatype, lang, FIELD_OBJECT, flags);
+ }
+
+ return false;
}
SerdStatus
@@ -735,166 +789,171 @@ serd_writer_write_statement(SerdWriter* writer,
const SerdNode* datatype,
const SerdNode* lang)
{
- if (!is_resource(subject) || !is_resource(predicate) || !object ||
- !object->buf) {
- return SERD_ERR_BAD_ARG;
- }
-
-#define TRY(write_result) \
- do { \
- if (!(write_result)) { \
- return SERD_ERR_UNKNOWN; \
- } \
- } while (0)
-
- if (writer->syntax == SERD_NTRIPLES || writer->syntax == SERD_NQUADS) {
- TRY(write_node(writer, subject, NULL, NULL, FIELD_SUBJECT, flags));
- sink(" ", 1, writer);
- TRY(write_node(writer, predicate, NULL, NULL, FIELD_PREDICATE, flags));
- sink(" ", 1, writer);
- TRY(write_node(writer, object, datatype, lang, FIELD_OBJECT, flags));
- if (writer->syntax == SERD_NQUADS && graph) {
- sink(" ", 1, writer);
- TRY(write_node(writer, graph, datatype, lang, FIELD_GRAPH, flags));
- }
- sink(" .\n", 3, writer);
- return SERD_SUCCESS;
- }
-
- if ((graph && !serd_node_equals(graph, &writer->context.graph)) ||
- (!graph && writer->context.graph.type)) {
- writer->indent = 0;
- if (writer->context.subject.type) {
- write_sep(writer, SEP_END_S);
- }
- if (writer->context.graph.type) {
- write_sep(writer, SEP_GRAPH_END);
- }
-
- reset_context(writer, true);
- if (graph) {
- TRY(write_node(writer, graph, datatype, lang, FIELD_GRAPH, flags));
- ++writer->indent;
- write_sep(writer, SEP_GRAPH_BEGIN);
- copy_node(&writer->context.graph, graph);
- }
- }
-
- if ((flags & SERD_LIST_CONT)) {
- if (write_list_obj(writer, flags, predicate, object, datatype, lang)) {
- // Reached end of list
- if (--writer->list_depth == 0 && writer->list_subj.type) {
- reset_context(writer, false);
- serd_node_free(&writer->context.subject);
- writer->context.subject = writer->list_subj;
- writer->list_subj = SERD_NODE_NULL;
- }
- return SERD_SUCCESS;
- }
- } else if (serd_node_equals(subject, &writer->context.subject)) {
- if (serd_node_equals(predicate, &writer->context.predicate)) {
- // Abbreviate S P
- if (!(flags & SERD_ANON_O_BEGIN)) {
- ++writer->indent;
- }
- write_sep(writer, SEP_END_O);
- write_node(writer, object, datatype, lang, FIELD_OBJECT, flags);
- if (!(flags & SERD_ANON_O_BEGIN)) {
- --writer->indent;
- }
- } else {
- // Abbreviate S
- Sep sep = writer->context.predicate.type ? SEP_END_P : SEP_S_P;
- write_sep(writer, sep);
- write_pred(writer, flags, predicate);
- write_node(writer, object, datatype, lang, FIELD_OBJECT, flags);
- }
- } else {
- // No abbreviation
- if (writer->context.subject.type) {
- assert(writer->indent > 0);
- --writer->indent;
- if (serd_stack_is_empty(&writer->anon_stack)) {
- write_sep(writer, SEP_END_S);
- }
- } else if (!writer->empty) {
- write_sep(writer, SEP_S_P);
- }
-
- if (!(flags & SERD_ANON_CONT)) {
- write_node(writer, subject, NULL, NULL, FIELD_SUBJECT, flags);
- ++writer->indent;
- write_sep(writer, SEP_S_P);
- } else {
- ++writer->indent;
- }
-
- reset_context(writer, false);
- copy_node(&writer->context.subject, subject);
-
- if (!(flags & SERD_LIST_S_BEGIN)) {
- write_pred(writer, flags, predicate);
- }
-
- write_node(writer, object, datatype, lang, FIELD_OBJECT, flags);
- }
-
- if (flags & (SERD_ANON_S_BEGIN|SERD_ANON_O_BEGIN)) {
- WriteContext* ctx = (WriteContext*)serd_stack_push(
- &writer->anon_stack, sizeof(WriteContext));
- *ctx = writer->context;
- WriteContext new_context = {
- serd_node_copy(graph), serd_node_copy(subject), SERD_NODE_NULL };
- if ((flags & SERD_ANON_S_BEGIN)) {
- new_context.predicate = serd_node_copy(predicate);
- }
- writer->context = new_context;
- } else {
- copy_node(&writer->context.graph, graph);
- copy_node(&writer->context.subject, subject);
- copy_node(&writer->context.predicate, predicate);
- }
-
- return SERD_SUCCESS;
+ if (!is_resource(subject) || !is_resource(predicate) || !object ||
+ !object->buf) {
+ return SERD_ERR_BAD_ARG;
+ }
+
+#define TRY(write_result) \
+ do { \
+ if (!(write_result)) { \
+ return SERD_ERR_UNKNOWN; \
+ } \
+ } while (0)
+
+ if (writer->syntax == SERD_NTRIPLES || writer->syntax == SERD_NQUADS) {
+ TRY(write_node(writer, subject, NULL, NULL, FIELD_SUBJECT, flags));
+ sink(" ", 1, writer);
+ TRY(write_node(writer, predicate, NULL, NULL, FIELD_PREDICATE, flags));
+ sink(" ", 1, writer);
+ TRY(write_node(writer, object, datatype, lang, FIELD_OBJECT, flags));
+ if (writer->syntax == SERD_NQUADS && graph) {
+ sink(" ", 1, writer);
+ TRY(write_node(writer, graph, datatype, lang, FIELD_GRAPH, flags));
+ }
+ sink(" .\n", 3, writer);
+ return SERD_SUCCESS;
+ }
+
+ if ((graph && !serd_node_equals(graph, &writer->context.graph)) ||
+ (!graph && writer->context.graph.type)) {
+ writer->indent = 0;
+
+ if (writer->context.subject.type) {
+ write_sep(writer, SEP_END_S);
+ }
+
+ if (writer->context.graph.type) {
+ write_sep(writer, SEP_GRAPH_END);
+ }
+
+ reset_context(writer, true);
+ if (graph) {
+ TRY(write_node(writer, graph, datatype, lang, FIELD_GRAPH, flags));
+ ++writer->indent;
+ write_sep(writer, SEP_GRAPH_BEGIN);
+ copy_node(&writer->context.graph, graph);
+ }
+ }
+
+ if ((flags & SERD_LIST_CONT)) {
+ if (write_list_obj(writer, flags, predicate, object, datatype, lang)) {
+ // Reached end of list
+ if (--writer->list_depth == 0 && writer->list_subj.type) {
+ reset_context(writer, false);
+ serd_node_free(&writer->context.subject);
+ writer->context.subject = writer->list_subj;
+ writer->list_subj = SERD_NODE_NULL;
+ }
+ return SERD_SUCCESS;
+ }
+ } else if (serd_node_equals(subject, &writer->context.subject)) {
+ if (serd_node_equals(predicate, &writer->context.predicate)) {
+ // Abbreviate S P
+ if (!(flags & SERD_ANON_O_BEGIN)) {
+ ++writer->indent;
+ }
+ write_sep(writer, SEP_END_O);
+ write_node(writer, object, datatype, lang, FIELD_OBJECT, flags);
+ if (!(flags & SERD_ANON_O_BEGIN)) {
+ --writer->indent;
+ }
+ } else {
+ // Abbreviate S
+ Sep sep = writer->context.predicate.type ? SEP_END_P : SEP_S_P;
+ write_sep(writer, sep);
+ write_pred(writer, flags, predicate);
+ write_node(writer, object, datatype, lang, FIELD_OBJECT, flags);
+ }
+ } else {
+ // No abbreviation
+ if (writer->context.subject.type) {
+ assert(writer->indent > 0);
+ --writer->indent;
+ if (serd_stack_is_empty(&writer->anon_stack)) {
+ write_sep(writer, SEP_END_S);
+ }
+ } else if (!writer->empty) {
+ write_sep(writer, SEP_S_P);
+ }
+
+ if (!(flags & SERD_ANON_CONT)) {
+ write_node(writer, subject, NULL, NULL, FIELD_SUBJECT, flags);
+ ++writer->indent;
+ write_sep(writer, SEP_S_P);
+ } else {
+ ++writer->indent;
+ }
+
+ reset_context(writer, false);
+ copy_node(&writer->context.subject, subject);
+
+ if (!(flags & SERD_LIST_S_BEGIN)) {
+ write_pred(writer, flags, predicate);
+ }
+
+ write_node(writer, object, datatype, lang, FIELD_OBJECT, flags);
+ }
+
+ if (flags & (SERD_ANON_S_BEGIN | SERD_ANON_O_BEGIN)) {
+ WriteContext* ctx =
+ (WriteContext*)serd_stack_push(&writer->anon_stack, sizeof(WriteContext));
+ *ctx = writer->context;
+ WriteContext new_context = {
+ serd_node_copy(graph), serd_node_copy(subject), SERD_NODE_NULL};
+ if ((flags & SERD_ANON_S_BEGIN)) {
+ new_context.predicate = serd_node_copy(predicate);
+ }
+ writer->context = new_context;
+ } else {
+ copy_node(&writer->context.graph, graph);
+ copy_node(&writer->context.subject, subject);
+ copy_node(&writer->context.predicate, predicate);
+ }
+
+ return SERD_SUCCESS;
}
SerdStatus
-serd_writer_end_anon(SerdWriter* writer,
- const SerdNode* node)
+serd_writer_end_anon(SerdWriter* writer, const SerdNode* node)
{
- if (writer->syntax == SERD_NTRIPLES || writer->syntax == SERD_NQUADS) {
- return SERD_SUCCESS;
- }
- if (serd_stack_is_empty(&writer->anon_stack) || writer->indent == 0) {
- w_err(writer, SERD_ERR_UNKNOWN,
- "unexpected end of anonymous node\n");
- return SERD_ERR_UNKNOWN;
- }
- --writer->indent;
- write_sep(writer, SEP_ANON_END);
- free_context(writer);
- writer->context = *anon_stack_top(writer);
- serd_stack_pop(&writer->anon_stack, sizeof(WriteContext));
- const bool is_subject = serd_node_equals(node, &writer->context.subject);
- if (is_subject) {
- copy_node(&writer->context.subject, node);
- writer->context.predicate.type = SERD_NOTHING;
- }
- return SERD_SUCCESS;
+ if (writer->syntax == SERD_NTRIPLES || writer->syntax == SERD_NQUADS) {
+ return SERD_SUCCESS;
+ }
+
+ if (serd_stack_is_empty(&writer->anon_stack) || writer->indent == 0) {
+ w_err(writer, SERD_ERR_UNKNOWN, "unexpected end of anonymous node\n");
+ return SERD_ERR_UNKNOWN;
+ }
+
+ --writer->indent;
+ write_sep(writer, SEP_ANON_END);
+ free_context(writer);
+ writer->context = *anon_stack_top(writer);
+ serd_stack_pop(&writer->anon_stack, sizeof(WriteContext));
+ const bool is_subject = serd_node_equals(node, &writer->context.subject);
+ if (is_subject) {
+ copy_node(&writer->context.subject, node);
+ writer->context.predicate.type = SERD_NOTHING;
+ }
+
+ return SERD_SUCCESS;
}
SerdStatus
serd_writer_finish(SerdWriter* writer)
{
- if (writer->context.subject.type) {
- write_sep(writer, SEP_END_S);
- }
- if (writer->context.graph.type) {
- write_sep(writer, SEP_GRAPH_END);
- }
- serd_byte_sink_flush(&writer->byte_sink);
- writer->indent = 0;
- return free_context(writer);
+ if (writer->context.subject.type) {
+ write_sep(writer, SEP_END_S);
+ }
+
+ if (writer->context.graph.type) {
+ write_sep(writer, SEP_GRAPH_END);
+ }
+
+ serd_byte_sink_flush(&writer->byte_sink);
+ writer->indent = 0;
+ return free_context(writer);
}
SerdWriter*
@@ -905,21 +964,23 @@ serd_writer_new(SerdSyntax syntax,
SerdSink ssink,
void* stream)
{
- const WriteContext context = WRITE_CONTEXT_NULL;
- SerdWriter* writer = (SerdWriter*)calloc(1, sizeof(SerdWriter));
- writer->syntax = syntax;
- writer->style = style;
- writer->env = env;
- writer->root_node = SERD_NODE_NULL;
- writer->root_uri = SERD_URI_NULL;
- writer->base_uri = base_uri ? *base_uri : SERD_URI_NULL;
- writer->anon_stack = serd_stack_new(4 * sizeof(WriteContext));
- writer->context = context;
- writer->list_subj = SERD_NODE_NULL;
- writer->empty = true;
- writer->byte_sink = serd_byte_sink_new(
- ssink, stream, (style & SERD_STYLE_BULK) ? SERD_PAGE_SIZE : 1);
- return writer;
+ const WriteContext context = WRITE_CONTEXT_NULL;
+ SerdWriter* writer = (SerdWriter*)calloc(1, sizeof(SerdWriter));
+
+ writer->syntax = syntax;
+ writer->style = style;
+ writer->env = env;
+ writer->root_node = SERD_NODE_NULL;
+ writer->root_uri = SERD_URI_NULL;
+ writer->base_uri = base_uri ? *base_uri : SERD_URI_NULL;
+ writer->anon_stack = serd_stack_new(4 * sizeof(WriteContext));
+ writer->context = context;
+ writer->list_subj = SERD_NODE_NULL;
+ writer->empty = true;
+ writer->byte_sink = serd_byte_sink_new(
+ ssink, stream, (style & SERD_STYLE_BULK) ? SERD_PAGE_SIZE : 1);
+
+ return writer;
}
void
@@ -927,60 +988,61 @@ serd_writer_set_error_sink(SerdWriter* writer,
SerdErrorSink error_sink,
void* error_handle)
{
- writer->error_sink = error_sink;
- writer->error_handle = error_handle;
+ writer->error_sink = error_sink;
+ writer->error_handle = error_handle;
}
void
serd_writer_chop_blank_prefix(SerdWriter* writer, const uint8_t* prefix)
{
- free(writer->bprefix);
- writer->bprefix_len = 0;
- writer->bprefix = NULL;
-
- const size_t prefix_len = prefix ? strlen((const char*)prefix) : 0;
- if (prefix_len) {
- writer->bprefix_len = prefix_len;
- writer->bprefix = (uint8_t*)malloc(writer->bprefix_len + 1);
- memcpy(writer->bprefix, prefix, writer->bprefix_len + 1);
- }
+ free(writer->bprefix);
+ writer->bprefix_len = 0;
+ writer->bprefix = NULL;
+
+ const size_t prefix_len = prefix ? strlen((const char*)prefix) : 0;
+ if (prefix_len) {
+ writer->bprefix_len = prefix_len;
+ writer->bprefix = (uint8_t*)malloc(writer->bprefix_len + 1);
+ memcpy(writer->bprefix, prefix, writer->bprefix_len + 1);
+ }
}
SerdStatus
-serd_writer_set_base_uri(SerdWriter* writer,
- const SerdNode* uri)
+serd_writer_set_base_uri(SerdWriter* writer, const SerdNode* uri)
{
- if (!serd_env_set_base_uri(writer->env, uri)) {
- serd_env_get_base_uri(writer->env, &writer->base_uri);
-
- if (writer->syntax == SERD_TURTLE || writer->syntax == SERD_TRIG) {
- if (writer->context.graph.type || writer->context.subject.type) {
- sink(" .\n\n", 4, writer);
- reset_context(writer, true);
- }
- sink("@base <", 7, writer);
- sink(uri->buf, uri->n_bytes, writer);
- sink("> .\n", 4, writer);
- }
- writer->indent = 0;
- return reset_context(writer, true);
- }
- return SERD_ERR_UNKNOWN;
+ if (!serd_env_set_base_uri(writer->env, uri)) {
+ serd_env_get_base_uri(writer->env, &writer->base_uri);
+
+ if (writer->syntax == SERD_TURTLE || writer->syntax == SERD_TRIG) {
+ if (writer->context.graph.type || writer->context.subject.type) {
+ sink(" .\n\n", 4, writer);
+ reset_context(writer, true);
+ }
+ sink("@base <", 7, writer);
+ sink(uri->buf, uri->n_bytes, writer);
+ sink("> .\n", 4, writer);
+ }
+ writer->indent = 0;
+ return reset_context(writer, true);
+ }
+
+ return SERD_ERR_UNKNOWN;
}
SerdStatus
-serd_writer_set_root_uri(SerdWriter* writer,
- const SerdNode* uri)
+serd_writer_set_root_uri(SerdWriter* writer, const SerdNode* uri)
{
- serd_node_free(&writer->root_node);
- if (uri && uri->buf) {
- writer->root_node = serd_node_copy(uri);
- serd_uri_parse(uri->buf, &writer->root_uri);
- } else {
- writer->root_node = SERD_NODE_NULL;
- writer->root_uri = SERD_URI_NULL;
- }
- return SERD_SUCCESS;
+ serd_node_free(&writer->root_node);
+
+ if (uri && uri->buf) {
+ writer->root_node = serd_node_copy(uri);
+ serd_uri_parse(uri->buf, &writer->root_uri);
+ } else {
+ writer->root_node = SERD_NODE_NULL;
+ writer->root_uri = SERD_URI_NULL;
+ }
+
+ return SERD_SUCCESS;
}
SerdStatus
@@ -988,64 +1050,65 @@ serd_writer_set_prefix(SerdWriter* writer,
const SerdNode* name,
const SerdNode* uri)
{
- if (!serd_env_set_prefix(writer->env, name, uri)) {
- if (writer->syntax == SERD_TURTLE || writer->syntax == SERD_TRIG) {
- if (writer->context.graph.type || writer->context.subject.type) {
- sink(" .\n\n", 4, writer);
- reset_context(writer, true);
- }
- sink("@prefix ", 8, writer);
- sink(name->buf, name->n_bytes, writer);
- sink(": <", 3, writer);
- write_uri(writer, uri->buf, uri->n_bytes);
- sink("> .\n", 4, writer);
- }
- writer->indent = 0;
- return reset_context(writer, true);
- }
- return SERD_ERR_UNKNOWN;
+ if (!serd_env_set_prefix(writer->env, name, uri)) {
+ if (writer->syntax == SERD_TURTLE || writer->syntax == SERD_TRIG) {
+ if (writer->context.graph.type || writer->context.subject.type) {
+ sink(" .\n\n", 4, writer);
+ reset_context(writer, true);
+ }
+ sink("@prefix ", 8, writer);
+ sink(name->buf, name->n_bytes, writer);
+ sink(": <", 3, writer);
+ write_uri(writer, uri->buf, uri->n_bytes);
+ sink("> .\n", 4, writer);
+ }
+ writer->indent = 0;
+ return reset_context(writer, true);
+ }
+
+ return SERD_ERR_UNKNOWN;
}
void
serd_writer_free(SerdWriter* writer)
{
- if (!writer) {
- return;
- }
-
- serd_writer_finish(writer);
- serd_stack_free(&writer->anon_stack);
- free(writer->bprefix);
- serd_byte_sink_free(&writer->byte_sink);
- serd_node_free(&writer->root_node);
- free(writer);
+ if (!writer) {
+ return;
+ }
+
+ serd_writer_finish(writer);
+ serd_stack_free(&writer->anon_stack);
+ free(writer->bprefix);
+ serd_byte_sink_free(&writer->byte_sink);
+ serd_node_free(&writer->root_node);
+ free(writer);
}
SerdEnv*
serd_writer_get_env(SerdWriter* writer)
{
- return writer->env;
+ return writer->env;
}
size_t
serd_file_sink(const void* buf, size_t len, void* stream)
{
- return fwrite(buf, 1, len, (FILE*)stream);
+ return fwrite(buf, 1, len, (FILE*)stream);
}
size_t
serd_chunk_sink(const void* buf, size_t len, void* stream)
{
- SerdChunk* chunk = (SerdChunk*)stream;
- chunk->buf = (uint8_t*)realloc((uint8_t*)chunk->buf, chunk->len + len);
- memcpy((uint8_t*)chunk->buf + chunk->len, buf, len);
- chunk->len += len;
- return len;
+ SerdChunk* chunk = (SerdChunk*)stream;
+ chunk->buf = (uint8_t*)realloc((uint8_t*)chunk->buf, chunk->len + len);
+ memcpy((uint8_t*)chunk->buf + chunk->len, buf, len);
+ chunk->len += len;
+ return len;
}
uint8_t*
serd_chunk_sink_finish(SerdChunk* stream)
{
- serd_chunk_sink("", 1, stream);
- return (uint8_t*)stream->buf;
+ serd_chunk_sink("", 1, stream);
+ return (uint8_t*)stream->buf;
}
diff --git a/test/test_env.c b/test/test_env.c
index 36b8f6c3..2fa9d3b5 100644
--- a/test/test_env.c
+++ b/test/test_env.c
@@ -27,74 +27,76 @@
static SerdStatus
count_prefixes(void* handle, const SerdNode* name, const SerdNode* uri)
{
- (void)name;
- (void)uri;
+ (void)name;
+ (void)uri;
- ++*(int*)handle;
- return SERD_SUCCESS;
+ ++*(int*)handle;
+ return SERD_SUCCESS;
}
static void
test_env(void)
{
- SerdNode u = serd_node_from_string(SERD_URI, USTR("http://example.org/foo"));
- SerdNode b = serd_node_from_string(SERD_CURIE, USTR("invalid"));
- SerdNode c = serd_node_from_string(SERD_CURIE, USTR("eg.2:b"));
- SerdEnv* env = serd_env_new(NULL);
- serd_env_set_prefix_from_strings(env, USTR("eg.2"), USTR("http://example.org/"));
+ SerdNode u = serd_node_from_string(SERD_URI, USTR("http://example.org/foo"));
+ SerdNode b = serd_node_from_string(SERD_CURIE, USTR("invalid"));
+ SerdNode c = serd_node_from_string(SERD_CURIE, USTR("eg.2:b"));
+ SerdEnv* env = serd_env_new(NULL);
+ serd_env_set_prefix_from_strings(
+ env, USTR("eg.2"), USTR("http://example.org/"));
- assert(!serd_env_set_base_uri(env, NULL));
- assert(serd_env_set_base_uri(env, &SERD_NODE_NULL));
- assert(serd_node_equals(serd_env_get_base_uri(env, NULL), &SERD_NODE_NULL));
+ assert(!serd_env_set_base_uri(env, NULL));
+ assert(serd_env_set_base_uri(env, &SERD_NODE_NULL));
+ assert(serd_node_equals(serd_env_get_base_uri(env, NULL), &SERD_NODE_NULL));
- SerdChunk prefix;
- SerdChunk suffix;
- assert(serd_env_expand(env, &b, &prefix, &suffix));
+ SerdChunk prefix;
+ SerdChunk suffix;
+ assert(serd_env_expand(env, &b, &prefix, &suffix));
- SerdNode xnode = serd_env_expand_node(env, &SERD_NODE_NULL);
- assert(serd_node_equals(&xnode, &SERD_NODE_NULL));
+ SerdNode xnode = serd_env_expand_node(env, &SERD_NODE_NULL);
+ assert(serd_node_equals(&xnode, &SERD_NODE_NULL));
- SerdNode xu = serd_env_expand_node(env, &u);
- assert(!strcmp((const char*)xu.buf, "http://example.org/foo"));
- serd_node_free(&xu);
+ SerdNode xu = serd_env_expand_node(env, &u);
+ assert(!strcmp((const char*)xu.buf, "http://example.org/foo"));
+ serd_node_free(&xu);
- SerdNode badpre = serd_node_from_string(SERD_CURIE, USTR("hm:what"));
- SerdNode xbadpre = serd_env_expand_node(env, &badpre);
- assert(serd_node_equals(&xbadpre, &SERD_NODE_NULL));
+ SerdNode badpre = serd_node_from_string(SERD_CURIE, USTR("hm:what"));
+ SerdNode xbadpre = serd_env_expand_node(env, &badpre);
+ assert(serd_node_equals(&xbadpre, &SERD_NODE_NULL));
- SerdNode xc = serd_env_expand_node(env, &c);
- assert(!strcmp((const char*)xc.buf, "http://example.org/b"));
- serd_node_free(&xc);
+ SerdNode xc = serd_env_expand_node(env, &c);
+ assert(!strcmp((const char*)xc.buf, "http://example.org/b"));
+ serd_node_free(&xc);
- assert(serd_env_set_prefix(env, &SERD_NODE_NULL, &SERD_NODE_NULL));
+ assert(serd_env_set_prefix(env, &SERD_NODE_NULL, &SERD_NODE_NULL));
- const SerdNode lit = serd_node_from_string(SERD_LITERAL, USTR("hello"));
- assert(serd_env_set_prefix(env, &b, &lit));
+ const SerdNode lit = serd_node_from_string(SERD_LITERAL, USTR("hello"));
+ assert(serd_env_set_prefix(env, &b, &lit));
- const SerdNode blank = serd_node_from_string(SERD_BLANK, USTR("b1"));
- const SerdNode xblank = serd_env_expand_node(env, &blank);
- assert(serd_node_equals(&xblank, &SERD_NODE_NULL));
+ const SerdNode blank = serd_node_from_string(SERD_BLANK, USTR("b1"));
+ const SerdNode xblank = serd_env_expand_node(env, &blank);
+ assert(serd_node_equals(&xblank, &SERD_NODE_NULL));
- int n_prefixes = 0;
- serd_env_set_prefix_from_strings(env, USTR("eg.2"), USTR("http://example.org/"));
- serd_env_foreach(env, count_prefixes, &n_prefixes);
- assert(n_prefixes == 1);
+ int n_prefixes = 0;
+ serd_env_set_prefix_from_strings(
+ env, USTR("eg.2"), USTR("http://example.org/"));
+ serd_env_foreach(env, count_prefixes, &n_prefixes);
+ assert(n_prefixes == 1);
- SerdNode shorter_uri = serd_node_from_string(SERD_URI, USTR("urn:foo"));
- SerdNode prefix_name;
- assert(!serd_env_qualify(env, &shorter_uri, &prefix_name, &suffix));
+ SerdNode shorter_uri = serd_node_from_string(SERD_URI, USTR("urn:foo"));
+ SerdNode prefix_name;
+ assert(!serd_env_qualify(env, &shorter_uri, &prefix_name, &suffix));
- assert(!serd_env_set_base_uri(env, &u));
- assert(serd_node_equals(serd_env_get_base_uri(env, NULL), &u));
- assert(!serd_env_set_base_uri(env, NULL));
- assert(!serd_env_get_base_uri(env, NULL)->buf);
+ assert(!serd_env_set_base_uri(env, &u));
+ assert(serd_node_equals(serd_env_get_base_uri(env, NULL), &u));
+ assert(!serd_env_set_base_uri(env, NULL));
+ assert(!serd_env_get_base_uri(env, NULL)->buf);
- serd_env_free(env);
+ serd_env_free(env);
}
int
main(void)
{
- test_env();
- return 0;
+ test_env();
+ return 0;
}
diff --git a/test/test_free_null.c b/test/test_free_null.c
index 121a66cd..a544a005 100644
--- a/test/test_free_null.c
+++ b/test/test_free_null.c
@@ -23,11 +23,11 @@
int
main(void)
{
- serd_free(NULL);
- serd_node_free(NULL);
- serd_env_free(NULL);
- serd_reader_free(NULL);
- serd_writer_free(NULL);
+ serd_free(NULL);
+ serd_node_free(NULL);
+ serd_env_free(NULL);
+ serd_reader_free(NULL);
+ serd_writer_free(NULL);
- return 0;
+ return 0;
}
diff --git a/test/test_node.c b/test/test_node.c
index 45c40aa4..f65a7c22 100644
--- a/test/test_node.c
+++ b/test/test_node.c
@@ -30,174 +30,183 @@
#define USTR(s) ((const uint8_t*)(s))
#ifndef INFINITY
-# define INFINITY (DBL_MAX + DBL_MAX)
+# define INFINITY (DBL_MAX + DBL_MAX)
#endif
#ifndef NAN
-# define NAN (INFINITY - INFINITY)
+# define NAN (INFINITY - INFINITY)
#endif
static void
test_strtod(double dbl, double max_delta)
{
- char buf[1024];
- snprintf(buf, sizeof(buf), "%f", dbl);
+ char buf[1024];
+ snprintf(buf, sizeof(buf), "%f", dbl);
- char* endptr = NULL;
- const double out = serd_strtod(buf, &endptr);
+ char* endptr = NULL;
+ const double out = serd_strtod(buf, &endptr);
+ const double diff = fabs(out - dbl);
- const double diff = fabs(out - dbl);
- assert(diff <= max_delta);
+ assert(diff <= max_delta);
}
static void
test_string_to_double(void)
{
- const double expt_test_nums[] = {
- 2.0E18, -5e19, +8e20, 2e+24, -5e-5, 8e0, 9e-0, 2e+0
- };
+ const double expt_test_nums[] = {
+ 2.0E18, -5e19, +8e20, 2e+24, -5e-5, 8e0, 9e-0, 2e+0};
- const char* expt_test_strs[] = {
- "02e18", "-5e019", "+8e20", "2E+24", "-5E-5", "8E0", "9e-0", " 2e+0"
- };
+ const char* expt_test_strs[] = {
+ "02e18", "-5e019", "+8e20", "2E+24", "-5E-5", "8E0", "9e-0", " 2e+0"};
- for (size_t i = 0; i < sizeof(expt_test_nums) / sizeof(double); ++i) {
- const double num = serd_strtod(expt_test_strs[i], NULL);
- const double delta = fabs(num - expt_test_nums[i]);
- assert(delta <= DBL_EPSILON);
+ for (size_t i = 0; i < sizeof(expt_test_nums) / sizeof(double); ++i) {
+ const double num = serd_strtod(expt_test_strs[i], NULL);
+ const double delta = fabs(num - expt_test_nums[i]);
+ assert(delta <= DBL_EPSILON);
- test_strtod(expt_test_nums[i], DBL_EPSILON);
- }
+ test_strtod(expt_test_nums[i], DBL_EPSILON);
+ }
}
static void
test_double_to_node(void)
{
- const double dbl_test_nums[] = {
- 0.0, 9.0, 10.0, .01, 2.05, -16.00001, 5.000000005, 0.0000000001, NAN, INFINITY
- };
-
- const char* dbl_test_strs[] = {
- "0.0", "9.0", "10.0", "0.01", "2.05", "-16.00001", "5.00000001", "0.0", NULL, NULL
- };
-
- for (size_t i = 0; i < sizeof(dbl_test_nums) / sizeof(double); ++i) {
- SerdNode node = serd_node_new_decimal(dbl_test_nums[i], 8);
- const bool pass = (node.buf && dbl_test_strs[i])
- ? !strcmp((const char*)node.buf, dbl_test_strs[i])
- : ((const char*)node.buf == dbl_test_strs[i]);
- assert(pass);
- const size_t len = node.buf ? strlen((const char*)node.buf) : 0;
- assert(node.n_bytes == len && node.n_chars == len);
- serd_node_free(&node);
- }
+ const double dbl_test_nums[] = {0.0,
+ 9.0,
+ 10.0,
+ .01,
+ 2.05,
+ -16.00001,
+ 5.000000005,
+ 0.0000000001,
+ NAN,
+ INFINITY};
+
+ const char* dbl_test_strs[] = {"0.0",
+ "9.0",
+ "10.0",
+ "0.01",
+ "2.05",
+ "-16.00001",
+ "5.00000001",
+ "0.0",
+ NULL,
+ NULL};
+
+ for (size_t i = 0; i < sizeof(dbl_test_nums) / sizeof(double); ++i) {
+ SerdNode node = serd_node_new_decimal(dbl_test_nums[i], 8);
+ const bool pass = (node.buf && dbl_test_strs[i])
+ ? !strcmp((const char*)node.buf, dbl_test_strs[i])
+ : ((const char*)node.buf == dbl_test_strs[i]);
+ assert(pass);
+ const size_t len = node.buf ? strlen((const char*)node.buf) : 0;
+ assert(node.n_bytes == len && node.n_chars == len);
+ serd_node_free(&node);
+ }
}
static void
test_integer_to_node(void)
{
- const long int_test_nums[] = {
- 0, -0, -23, 23, -12340, 1000, -1000
- };
-
- const char* int_test_strs[] = {
- "0", "0", "-23", "23", "-12340", "1000", "-1000"
- };
-
- for (size_t i = 0; i < sizeof(int_test_nums) / sizeof(double); ++i) {
- SerdNode node = serd_node_new_integer(int_test_nums[i]);
- assert(!strcmp((const char*)node.buf, (const char*)int_test_strs[i]));
- const size_t len = strlen((const char*)node.buf);
- assert(node.n_bytes == len && node.n_chars == len);
- serd_node_free(&node);
- }
+ const long int_test_nums[] = {0, -0, -23, 23, -12340, 1000, -1000};
+
+ const char* int_test_strs[] = {
+ "0", "0", "-23", "23", "-12340", "1000", "-1000"};
+
+ for (size_t i = 0; i < sizeof(int_test_nums) / sizeof(double); ++i) {
+ SerdNode node = serd_node_new_integer(int_test_nums[i]);
+ assert(!strcmp((const char*)node.buf, (const char*)int_test_strs[i]));
+ const size_t len = strlen((const char*)node.buf);
+ assert(node.n_bytes == len && node.n_chars == len);
+ serd_node_free(&node);
+ }
}
static void
test_blob_to_node(void)
{
- for (size_t size = 1; size < 256; ++size) {
- uint8_t* const data = (uint8_t*)malloc(size);
- for (size_t i = 0; i < size; ++i) {
- data[i] = (uint8_t)((size + i) % 256);
- }
-
- SerdNode blob = serd_node_new_blob(data, size, size % 5);
-
- assert(blob.n_bytes == blob.n_chars);
- assert(blob.n_bytes == strlen((const char*)blob.buf));
-
- size_t out_size = 0;
- uint8_t* out = (uint8_t*)serd_base64_decode(
- blob.buf, blob.n_bytes, &out_size);
- assert(out_size == size);
-
- for (size_t i = 0; i < size; ++i) {
- assert(out[i] == data[i]);
- }
-
- serd_node_free(&blob);
- serd_free(out);
- free(data);
- }
+ for (size_t size = 1; size < 256; ++size) {
+ uint8_t* const data = (uint8_t*)malloc(size);
+ for (size_t i = 0; i < size; ++i) {
+ data[i] = (uint8_t)((size + i) % 256);
+ }
+
+ SerdNode blob = serd_node_new_blob(data, size, size % 5);
+
+ assert(blob.n_bytes == blob.n_chars);
+ assert(blob.n_bytes == strlen((const char*)blob.buf));
+
+ size_t out_size = 0;
+ uint8_t* out =
+ (uint8_t*)serd_base64_decode(blob.buf, blob.n_bytes, &out_size);
+ assert(out_size == size);
+
+ for (size_t i = 0; i < size; ++i) {
+ assert(out[i] == data[i]);
+ }
+
+ serd_node_free(&blob);
+ serd_free(out);
+ free(data);
+ }
}
static void
test_node_equals(void)
{
- const uint8_t replacement_char_str[] = { 0xEF, 0xBF, 0xBD, 0 };
- SerdNode lhs = serd_node_from_string(SERD_LITERAL, replacement_char_str);
- SerdNode rhs = serd_node_from_string(SERD_LITERAL, USTR("123"));
- assert(!serd_node_equals(&lhs, &rhs));
+ const uint8_t replacement_char_str[] = {0xEF, 0xBF, 0xBD, 0};
+ SerdNode lhs = serd_node_from_string(SERD_LITERAL, replacement_char_str);
+ SerdNode rhs = serd_node_from_string(SERD_LITERAL, USTR("123"));
+ assert(!serd_node_equals(&lhs, &rhs));
- SerdNode qnode = serd_node_from_string(SERD_CURIE, USTR("foo:bar"));
- assert(!serd_node_equals(&lhs, &qnode));
- assert(serd_node_equals(&lhs, &lhs));
+ SerdNode qnode = serd_node_from_string(SERD_CURIE, USTR("foo:bar"));
+ assert(!serd_node_equals(&lhs, &qnode));
+ assert(serd_node_equals(&lhs, &lhs));
- SerdNode null_copy = serd_node_copy(&SERD_NODE_NULL);
- assert(serd_node_equals(&SERD_NODE_NULL, &null_copy));
+ SerdNode null_copy = serd_node_copy(&SERD_NODE_NULL);
+ assert(serd_node_equals(&SERD_NODE_NULL, &null_copy));
}
static void
test_node_from_string(void)
{
- SerdNode node = serd_node_from_string(SERD_LITERAL, (const uint8_t*)"hello\"");
- assert(node.n_bytes == 6 && node.n_chars == 6 &&
- node.flags == SERD_HAS_QUOTE &&
- !strcmp((const char*)node.buf, "hello\""));
+ SerdNode node =
+ serd_node_from_string(SERD_LITERAL, (const uint8_t*)"hello\"");
- node = serd_node_from_string(SERD_URI, NULL);
- assert(serd_node_equals(&node, &SERD_NODE_NULL));
+ assert(node.n_bytes == 6 && node.n_chars == 6 &&
+ node.flags == SERD_HAS_QUOTE &&
+ !strcmp((const char*)node.buf, "hello\""));
+
+ node = serd_node_from_string(SERD_URI, NULL);
+ assert(serd_node_equals(&node, &SERD_NODE_NULL));
}
static void
test_node_from_substring(void)
{
- SerdNode empty = serd_node_from_substring(SERD_LITERAL, NULL, 32);
- assert(!empty.buf && !empty.n_bytes && !empty.n_chars && !empty.flags &&
- !empty.type);
-
- SerdNode a_b = serd_node_from_substring(SERD_LITERAL, USTR("a\"bc"), 3);
- assert(a_b.n_bytes == 3 && a_b.n_chars == 3 &&
- a_b.flags == SERD_HAS_QUOTE &&
- !strncmp((const char*)a_b.buf, "a\"b", 3));
-
- a_b = serd_node_from_substring(SERD_LITERAL, USTR("a\"bc"), 10);
- assert(a_b.n_bytes == 4 && a_b.n_chars == 4 &&
- a_b.flags == SERD_HAS_QUOTE &&
- !strncmp((const char*)a_b.buf, "a\"bc", 4));
+ SerdNode empty = serd_node_from_substring(SERD_LITERAL, NULL, 32);
+ assert(!empty.buf && !empty.n_bytes && !empty.n_chars && !empty.flags &&
+ !empty.type);
+
+ SerdNode a_b = serd_node_from_substring(SERD_LITERAL, USTR("a\"bc"), 3);
+ assert(a_b.n_bytes == 3 && a_b.n_chars == 3 && a_b.flags == SERD_HAS_QUOTE &&
+ !strncmp((const char*)a_b.buf, "a\"b", 3));
+
+ a_b = serd_node_from_substring(SERD_LITERAL, USTR("a\"bc"), 10);
+ assert(a_b.n_bytes == 4 && a_b.n_chars == 4 && a_b.flags == SERD_HAS_QUOTE &&
+ !strncmp((const char*)a_b.buf, "a\"bc", 4));
}
int
main(void)
{
- test_string_to_double();
- test_double_to_node();
- test_integer_to_node();
- test_blob_to_node();
- test_node_equals();
- test_node_from_string();
- test_node_from_substring();
-
- printf("Success\n");
- return 0;
+ test_string_to_double();
+ test_double_to_node();
+ test_integer_to_node();
+ test_blob_to_node();
+ test_node_equals();
+ test_node_from_string();
+ test_node_from_substring();
+
+ printf("Success\n");
+ return 0;
}
diff --git a/test/test_read_chunk.c b/test/test_read_chunk.c
index cc9c2e19..2e7f2b0d 100644
--- a/test/test_read_chunk.c
+++ b/test/test_read_chunk.c
@@ -30,22 +30,22 @@ static size_t n_end = 0;
static SerdStatus
on_base(void* handle, const SerdNode* uri)
{
- (void)handle;
- (void)uri;
+ (void)handle;
+ (void)uri;
- ++n_base;
- return SERD_SUCCESS;
+ ++n_base;
+ return SERD_SUCCESS;
}
static SerdStatus
on_prefix(void* handle, const SerdNode* name, const SerdNode* uri)
{
- (void)handle;
- (void)name;
- (void)uri;
+ (void)handle;
+ (void)name;
+ (void)uri;
- ++n_prefix;
- return SERD_SUCCESS;
+ ++n_prefix;
+ return SERD_SUCCESS;
}
static SerdStatus
@@ -58,64 +58,65 @@ on_statement(void* handle,
const SerdNode* object_datatype,
const SerdNode* object_lang)
{
- (void)handle;
- (void)flags;
- (void)graph;
- (void)subject;
- (void)predicate;
- (void)object;
- (void)object_datatype;
- (void)object_lang;
-
- ++n_statement;
- return SERD_SUCCESS;
+ (void)handle;
+ (void)flags;
+ (void)graph;
+ (void)subject;
+ (void)predicate;
+ (void)object;
+ (void)object_datatype;
+ (void)object_lang;
+
+ ++n_statement;
+ return SERD_SUCCESS;
}
static SerdStatus
on_end(void* handle, const SerdNode* node)
{
- (void)handle;
- (void)node;
+ (void)handle;
+ (void)node;
- ++n_end;
- return SERD_SUCCESS;
+ ++n_end;
+ return SERD_SUCCESS;
}
int
main(void)
{
- FILE* file = tmpfile();
-
- fprintf(file,
- "@prefix eg: <http://example.org/> .\n"
- "@base <http://example.org/base> .\n"
- "eg:s1 eg:p1 eg:o1 ;\n"
- " eg:p2 eg:o2 ,\n"
- " eg:o3 .\n"
- "eg:s2 eg:p1 eg:o1 ;\n"
- " eg:p2 eg:o2 .\n"
- "eg:s3 eg:p1 eg:o1 .\n"
- "eg:s4 eg:p1 [ eg:p3 eg:o1 ] .\n");
-
- fseek(file, 0, SEEK_SET);
-
- SerdReader* reader = serd_reader_new(
- SERD_TURTLE, NULL, NULL, on_base, on_prefix, on_statement, on_end);
- assert(reader);
-
- assert(!serd_reader_start_stream(reader, file, NULL, true));
-
- assert(!serd_reader_read_chunk(reader) && n_prefix == 1);
- assert(!serd_reader_read_chunk(reader) && n_base == 1);
- assert(!serd_reader_read_chunk(reader) && n_statement == 3);
- assert(!serd_reader_read_chunk(reader) && n_statement == 5);
- assert(!serd_reader_read_chunk(reader) && n_statement == 6);
- assert(!serd_reader_read_chunk(reader) && n_statement == 8 && n_end == 1);
- assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
- assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
- assert(!serd_reader_end_stream(reader));
-
- serd_reader_free(reader);
- fclose(file);
- return 0;
+ FILE* file = tmpfile();
+
+ fprintf(file,
+ "@prefix eg: <http://example.org/> .\n"
+ "@base <http://example.org/base> .\n"
+ "eg:s1 eg:p1 eg:o1 ;\n"
+ " eg:p2 eg:o2 ,\n"
+ " eg:o3 .\n"
+ "eg:s2 eg:p1 eg:o1 ;\n"
+ " eg:p2 eg:o2 .\n"
+ "eg:s3 eg:p1 eg:o1 .\n"
+ "eg:s4 eg:p1 [ eg:p3 eg:o1 ] .\n");
+
+ fseek(file, 0, SEEK_SET);
+
+ SerdReader* reader = serd_reader_new(
+ SERD_TURTLE, NULL, NULL, on_base, on_prefix, on_statement, on_end);
+
+ assert(reader);
+ assert(!serd_reader_start_stream(reader, file, NULL, true));
+
+ assert(!serd_reader_read_chunk(reader) && n_prefix == 1);
+ assert(!serd_reader_read_chunk(reader) && n_base == 1);
+ assert(!serd_reader_read_chunk(reader) && n_statement == 3);
+ assert(!serd_reader_read_chunk(reader) && n_statement == 5);
+ assert(!serd_reader_read_chunk(reader) && n_statement == 6);
+ assert(!serd_reader_read_chunk(reader) && n_statement == 8 && n_end == 1);
+ assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
+ assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
+
+ assert(!serd_reader_end_stream(reader));
+ serd_reader_free(reader);
+ fclose(file);
+
+ return 0;
}
diff --git a/test/test_reader_writer.c b/test/test_reader_writer.c
index 22576f72..d47bcccc 100644
--- a/test/test_reader_writer.c
+++ b/test/test_reader_writer.c
@@ -28,8 +28,8 @@
#define USTR(s) ((const uint8_t*)(s))
typedef struct {
- int n_statements;
- const SerdNode* graph;
+ int n_statements;
+ const SerdNode* graph;
} ReaderTest;
static SerdStatus
@@ -42,315 +42,322 @@ test_sink(void* handle,
const SerdNode* object_datatype,
const SerdNode* object_lang)
{
- (void)flags;
- (void)subject;
- (void)predicate;
- (void)object;
- (void)object_datatype;
- (void)object_lang;
-
- ReaderTest* rt = (ReaderTest*)handle;
- ++rt->n_statements;
- rt->graph = graph;
- return SERD_SUCCESS;
+ (void)flags;
+ (void)subject;
+ (void)predicate;
+ (void)object;
+ (void)object_datatype;
+ (void)object_lang;
+
+ ReaderTest* rt = (ReaderTest*)handle;
+ ++rt->n_statements;
+ rt->graph = graph;
+ return SERD_SUCCESS;
}
/// Returns EOF after a statement, then succeeds again (like a socket)
static size_t
eof_test_read(void* buf, size_t size, size_t nmemb, void* stream)
{
- assert(nmemb == 1);
+ assert(nmemb == 1);
- static const char* const string = "_:s1 <http://example.org/p> _:o1 .\n"
- "_:s2 <http://example.org/p> _:o2 .\n";
+ static const char* const string = "_:s1 <http://example.org/p> _:o1 .\n"
+ "_:s2 <http://example.org/p> _:o2 .\n";
- size_t* count = (size_t*)stream;
- if (*count == 34 || *count == 35 || *count + nmemb >= strlen(string)) {
- ++*count;
- return 0;
- }
+ size_t* count = (size_t*)stream;
+ if (*count == 34 || *count == 35 || *count + nmemb >= strlen(string)) {
+ ++*count;
+ return 0;
+ }
- memcpy((char*)buf, string + *count, size * nmemb);
- *count += nmemb;
- return nmemb;
+ memcpy((char*)buf, string + *count, size * nmemb);
+ *count += nmemb;
+ return nmemb;
}
static int
eof_test_error(void* stream)
{
- (void)stream;
- return 0;
+ (void)stream;
+ return 0;
}
static void
test_read_chunks(void)
{
- ReaderTest* const rt = (ReaderTest*)calloc(1, sizeof(ReaderTest));
- FILE* const f = tmpfile();
- static const char null = 0;
- SerdReader* const reader =
- serd_reader_new(SERD_TURTLE, rt, free, NULL, NULL, test_sink, NULL);
-
- assert(reader);
- assert(serd_reader_get_handle(reader) == rt);
- assert(f);
-
- SerdStatus st = serd_reader_start_stream(reader, f, NULL, false);
- assert(st == SERD_SUCCESS);
-
- // Write two statement separated by null characters
- fprintf(f, "@prefix eg: <http://example.org/> .\n");
- fprintf(f, "eg:s eg:p eg:o1 .\n");
- fwrite(&null, sizeof(null), 1, f);
- fprintf(f, "eg:s eg:p eg:o2 .\n");
- fwrite(&null, sizeof(null), 1, f);
- fseek(f, 0, SEEK_SET);
-
- // Read prefix
- st = serd_reader_read_chunk(reader);
- assert(st == SERD_SUCCESS);
- assert(rt->n_statements == 0);
-
- // Read first statement
- st = serd_reader_read_chunk(reader);
- assert(st == SERD_SUCCESS);
- assert(rt->n_statements == 1);
-
- // Read terminator
- st = serd_reader_read_chunk(reader);
- assert(st == SERD_FAILURE);
- assert(rt->n_statements == 1);
-
- // Read second statement (after null terminator)
- st = serd_reader_read_chunk(reader);
- assert(st == SERD_SUCCESS);
- assert(rt->n_statements == 2);
-
- // Read terminator
- st = serd_reader_read_chunk(reader);
- assert(st == SERD_FAILURE);
- assert(rt->n_statements == 2);
-
- // EOF
- st = serd_reader_read_chunk(reader);
- assert(st == SERD_FAILURE);
- assert(rt->n_statements == 2);
-
- serd_reader_free(reader);
- fclose(f);
+ ReaderTest* const rt = (ReaderTest*)calloc(1, sizeof(ReaderTest));
+ FILE* const f = tmpfile();
+ static const char null = 0;
+ SerdReader* const reader =
+ serd_reader_new(SERD_TURTLE, rt, free, NULL, NULL, test_sink, NULL);
+
+ assert(reader);
+ assert(serd_reader_get_handle(reader) == rt);
+ assert(f);
+
+ SerdStatus st = serd_reader_start_stream(reader, f, NULL, false);
+ assert(st == SERD_SUCCESS);
+
+ // Write two statement separated by null characters
+ fprintf(f, "@prefix eg: <http://example.org/> .\n");
+ fprintf(f, "eg:s eg:p eg:o1 .\n");
+ fwrite(&null, sizeof(null), 1, f);
+ fprintf(f, "eg:s eg:p eg:o2 .\n");
+ fwrite(&null, sizeof(null), 1, f);
+ fseek(f, 0, SEEK_SET);
+
+ // Read prefix
+ st = serd_reader_read_chunk(reader);
+ assert(st == SERD_SUCCESS);
+ assert(rt->n_statements == 0);
+
+ // Read first statement
+ st = serd_reader_read_chunk(reader);
+ assert(st == SERD_SUCCESS);
+ assert(rt->n_statements == 1);
+
+ // Read terminator
+ st = serd_reader_read_chunk(reader);
+ assert(st == SERD_FAILURE);
+ assert(rt->n_statements == 1);
+
+ // Read second statement (after null terminator)
+ st = serd_reader_read_chunk(reader);
+ assert(st == SERD_SUCCESS);
+ assert(rt->n_statements == 2);
+
+ // Read terminator
+ st = serd_reader_read_chunk(reader);
+ assert(st == SERD_FAILURE);
+ assert(rt->n_statements == 2);
+
+ // EOF
+ st = serd_reader_read_chunk(reader);
+ assert(st == SERD_FAILURE);
+ assert(rt->n_statements == 2);
+
+ serd_reader_free(reader);
+ fclose(f);
}
static void
test_read_string(void)
{
- ReaderTest* rt = (ReaderTest*)calloc(1, sizeof(ReaderTest));
- SerdReader* reader =
- serd_reader_new(SERD_TURTLE, rt, free, NULL, NULL, test_sink, NULL);
+ ReaderTest* rt = (ReaderTest*)calloc(1, sizeof(ReaderTest));
+ SerdReader* reader =
+ serd_reader_new(SERD_TURTLE, rt, free, NULL, NULL, test_sink, NULL);
- assert(reader);
- assert(serd_reader_get_handle(reader) == rt);
+ assert(reader);
+ assert(serd_reader_get_handle(reader) == rt);
- // Test reading a string that ends exactly at the end of input (no newline)
- const SerdStatus st =
- serd_reader_read_string(reader,
- USTR("<http://example.org/s> <http://example.org/p> "
- "<http://example.org/o> ."));
+ // Test reading a string that ends exactly at the end of input (no newline)
+ const SerdStatus st = serd_reader_read_string(
+ reader,
+ USTR("<http://example.org/s> <http://example.org/p> "
+ "<http://example.org/o> ."));
- assert(!st);
- assert(rt->n_statements == 1);
+ assert(!st);
+ assert(rt->n_statements == 1);
- serd_reader_free(reader);
+ serd_reader_free(reader);
}
static void
test_writer(const char* const path)
{
- FILE* fd = fopen(path, "wb");
- SerdEnv* env = serd_env_new(NULL);
- assert(fd);
-
- SerdWriter* writer = serd_writer_new(
- SERD_TURTLE, (SerdStyle)0, env, NULL, serd_file_sink, fd);
- assert(writer);
-
- serd_writer_chop_blank_prefix(writer, USTR("tmp"));
- serd_writer_chop_blank_prefix(writer, NULL);
-
- const SerdNode lit = serd_node_from_string(SERD_LITERAL, USTR("hello"));
-
- assert(serd_writer_set_base_uri(writer, &lit));
- assert(serd_writer_set_prefix(writer, &lit, &lit));
- assert(serd_writer_end_anon(writer, NULL));
- assert(serd_writer_get_env(writer) == env);
-
- uint8_t buf[] = { 0x80, 0, 0, 0, 0 };
- SerdNode s = serd_node_from_string(SERD_URI, USTR(""));
- SerdNode p = serd_node_from_string(SERD_URI, USTR("http://example.org/pred"));
- SerdNode o = serd_node_from_string(SERD_LITERAL, buf);
-
- // Write 3 invalid statements (should write nothing)
- const SerdNode* junk[][5] = { { &s, &p, &SERD_NODE_NULL, NULL, NULL },
- { &s, &SERD_NODE_NULL, &o, NULL, NULL },
- { &SERD_NODE_NULL, &p, &o, NULL, NULL },
- { &s, &o, &o, NULL, NULL },
- { &o, &p, &o, NULL, NULL },
- { &s, &p, &SERD_NODE_NULL, NULL, NULL } };
- for (size_t i = 0; i < sizeof(junk) / (sizeof(SerdNode*) * 5); ++i) {
- assert(serd_writer_write_statement(
- writer, 0, NULL,
- junk[i][0], junk[i][1], junk[i][2], junk[i][3], junk[i][4]));
- }
-
- const SerdNode t = serd_node_from_string(SERD_URI, USTR("urn:Type"));
- const SerdNode l = serd_node_from_string(SERD_LITERAL, USTR("en"));
- const SerdNode* good[][5] = { { &s, &p, &o, NULL, NULL },
- { &s, &p, &o, &SERD_NODE_NULL, &SERD_NODE_NULL },
- { &s, &p, &o, &t, NULL },
- { &s, &p, &o, NULL, &l },
- { &s, &p, &o, &t, &l },
- { &s, &p, &o, &t, &SERD_NODE_NULL },
- { &s, &p, &o, &SERD_NODE_NULL, &l },
- { &s, &p, &o, NULL, &SERD_NODE_NULL },
- { &s, &p, &o, &SERD_NODE_NULL, NULL },
- { &s, &p, &o, &SERD_NODE_NULL, NULL } };
- for (size_t i = 0; i < sizeof(good) / (sizeof(SerdNode*) * 5); ++i) {
- assert(!serd_writer_write_statement(
- writer, 0, NULL,
- good[i][0], good[i][1], good[i][2], good[i][3], good[i][4]));
- }
-
- // Write statements with bad UTF-8 (should be replaced)
- const uint8_t bad_str[] = { 0xFF, 0x90, 'h', 'i', 0 };
- SerdNode bad_lit = serd_node_from_string(SERD_LITERAL, bad_str);
- SerdNode bad_uri = serd_node_from_string(SERD_URI, bad_str);
- assert(!serd_writer_write_statement(writer, 0, NULL,
- &s, &p, &bad_lit, NULL, NULL));
- assert(!serd_writer_write_statement(writer, 0, NULL,
- &s, &p, &bad_uri, NULL, NULL));
-
- // Write 1 valid statement
- o = serd_node_from_string(SERD_LITERAL, USTR("hello"));
- assert(!serd_writer_write_statement(writer, 0, NULL,
- &s, &p, &o, NULL, NULL));
-
- serd_writer_free(writer);
-
- // Test chunk sink
- SerdChunk chunk = { NULL, 0 };
- writer = serd_writer_new(
- SERD_TURTLE, (SerdStyle)0, env, NULL, serd_chunk_sink, &chunk);
-
- o = serd_node_from_string(SERD_URI, USTR("http://example.org/base"));
- assert(!serd_writer_set_base_uri(writer, &o));
-
- serd_writer_free(writer);
- uint8_t* out = serd_chunk_sink_finish(&chunk);
-
- assert(!strcmp((const char*)out, "@base <http://example.org/base> .\n"));
- serd_free(out);
-
- // Test writing empty node
- SerdNode nothing = serd_node_from_string(SERD_NOTHING, USTR(""));
- FILE* const empty = tmpfile();
-
- writer = serd_writer_new(
- SERD_TURTLE, (SerdStyle)0, env, NULL, serd_file_sink, empty);
-
- // FIXME: error handling
- serd_writer_write_statement(writer, 0, NULL,
- &s, &p, &nothing, NULL, NULL);
-
- assert((size_t)ftell(empty) == strlen("<>\n\t<http://example.org/pred> "));
-
- serd_writer_free(writer);
- fclose(empty);
-
- serd_env_free(env);
- fclose(fd);
+ FILE* fd = fopen(path, "wb");
+ SerdEnv* env = serd_env_new(NULL);
+ assert(fd);
+
+ SerdWriter* writer =
+ serd_writer_new(SERD_TURTLE, (SerdStyle)0, env, NULL, serd_file_sink, fd);
+ assert(writer);
+
+ serd_writer_chop_blank_prefix(writer, USTR("tmp"));
+ serd_writer_chop_blank_prefix(writer, NULL);
+
+ const SerdNode lit = serd_node_from_string(SERD_LITERAL, USTR("hello"));
+
+ assert(serd_writer_set_base_uri(writer, &lit));
+ assert(serd_writer_set_prefix(writer, &lit, &lit));
+ assert(serd_writer_end_anon(writer, NULL));
+ assert(serd_writer_get_env(writer) == env);
+
+ uint8_t buf[] = {0x80, 0, 0, 0, 0};
+ SerdNode s = serd_node_from_string(SERD_URI, USTR(""));
+ SerdNode p = serd_node_from_string(SERD_URI, USTR("http://example.org/pred"));
+ SerdNode o = serd_node_from_string(SERD_LITERAL, buf);
+
+ // Write 3 invalid statements (should write nothing)
+ const SerdNode* junk[][5] = {{&s, &p, &SERD_NODE_NULL, NULL, NULL},
+ {&s, &SERD_NODE_NULL, &o, NULL, NULL},
+ {&SERD_NODE_NULL, &p, &o, NULL, NULL},
+ {&s, &o, &o, NULL, NULL},
+ {&o, &p, &o, NULL, NULL},
+ {&s, &p, &SERD_NODE_NULL, NULL, NULL}};
+ for (size_t i = 0; i < sizeof(junk) / (sizeof(SerdNode*) * 5); ++i) {
+ assert(serd_writer_write_statement(writer,
+ 0,
+ NULL,
+ junk[i][0],
+ junk[i][1],
+ junk[i][2],
+ junk[i][3],
+ junk[i][4]));
+ }
+
+ const SerdNode t = serd_node_from_string(SERD_URI, USTR("urn:Type"));
+ const SerdNode l = serd_node_from_string(SERD_LITERAL, USTR("en"));
+ const SerdNode* good[][5] = {{&s, &p, &o, NULL, NULL},
+ {&s, &p, &o, &SERD_NODE_NULL, &SERD_NODE_NULL},
+ {&s, &p, &o, &t, NULL},
+ {&s, &p, &o, NULL, &l},
+ {&s, &p, &o, &t, &l},
+ {&s, &p, &o, &t, &SERD_NODE_NULL},
+ {&s, &p, &o, &SERD_NODE_NULL, &l},
+ {&s, &p, &o, NULL, &SERD_NODE_NULL},
+ {&s, &p, &o, &SERD_NODE_NULL, NULL},
+ {&s, &p, &o, &SERD_NODE_NULL, NULL}};
+ for (size_t i = 0; i < sizeof(good) / (sizeof(SerdNode*) * 5); ++i) {
+ assert(!serd_writer_write_statement(writer,
+ 0,
+ NULL,
+ good[i][0],
+ good[i][1],
+ good[i][2],
+ good[i][3],
+ good[i][4]));
+ }
+
+ // Write statements with bad UTF-8 (should be replaced)
+ const uint8_t bad_str[] = {0xFF, 0x90, 'h', 'i', 0};
+ SerdNode bad_lit = serd_node_from_string(SERD_LITERAL, bad_str);
+ SerdNode bad_uri = serd_node_from_string(SERD_URI, bad_str);
+ assert(!serd_writer_write_statement(
+ writer, 0, NULL, &s, &p, &bad_lit, NULL, NULL));
+ assert(!serd_writer_write_statement(
+ writer, 0, NULL, &s, &p, &bad_uri, NULL, NULL));
+
+ // Write 1 valid statement
+ o = serd_node_from_string(SERD_LITERAL, USTR("hello"));
+ assert(!serd_writer_write_statement(writer, 0, NULL, &s, &p, &o, NULL, NULL));
+
+ serd_writer_free(writer);
+
+ // Test chunk sink
+ SerdChunk chunk = {NULL, 0};
+ writer = serd_writer_new(
+ SERD_TURTLE, (SerdStyle)0, env, NULL, serd_chunk_sink, &chunk);
+
+ o = serd_node_from_string(SERD_URI, USTR("http://example.org/base"));
+ assert(!serd_writer_set_base_uri(writer, &o));
+
+ serd_writer_free(writer);
+ uint8_t* out = serd_chunk_sink_finish(&chunk);
+
+ assert(!strcmp((const char*)out, "@base <http://example.org/base> .\n"));
+ serd_free(out);
+
+ // Test writing empty node
+ SerdNode nothing = serd_node_from_string(SERD_NOTHING, USTR(""));
+ FILE* const empty = tmpfile();
+
+ writer = serd_writer_new(
+ SERD_TURTLE, (SerdStyle)0, env, NULL, serd_file_sink, empty);
+
+ // FIXME: error handling
+ serd_writer_write_statement(writer, 0, NULL, &s, &p, &nothing, NULL, NULL);
+
+ assert((size_t)ftell(empty) == strlen("<>\n\t<http://example.org/pred> "));
+
+ serd_writer_free(writer);
+ fclose(empty);
+
+ serd_env_free(env);
+ fclose(fd);
}
static void
test_reader(const char* path)
{
- ReaderTest* rt = (ReaderTest*)calloc(1, sizeof(ReaderTest));
- SerdReader* reader = serd_reader_new(
- SERD_TURTLE, rt, free,
- NULL, NULL, test_sink, NULL);
- assert(reader);
- assert(serd_reader_get_handle(reader) == rt);
+ ReaderTest* rt = (ReaderTest*)calloc(1, sizeof(ReaderTest));
+ SerdReader* reader =
+ serd_reader_new(SERD_TURTLE, rt, free, NULL, NULL, test_sink, NULL);
+ assert(reader);
+ assert(serd_reader_get_handle(reader) == rt);
- SerdNode g = serd_node_from_string(SERD_URI, USTR("http://example.org/"));
- serd_reader_set_default_graph(reader, &g);
- serd_reader_add_blank_prefix(reader, USTR("tmp"));
+ SerdNode g = serd_node_from_string(SERD_URI, USTR("http://example.org/"));
+ serd_reader_set_default_graph(reader, &g);
+ serd_reader_add_blank_prefix(reader, USTR("tmp"));
#if defined(__GNUC__)
-# pragma GCC diagnostic push
-# pragma GCC diagnostic ignored "-Wnonnull"
+# pragma GCC diagnostic push
+# pragma GCC diagnostic ignored "-Wnonnull"
#endif
- serd_reader_add_blank_prefix(reader, NULL);
+ serd_reader_add_blank_prefix(reader, NULL);
#if defined(__GNUC__)
-# pragma GCC diagnostic pop
+# pragma GCC diagnostic pop
#endif
- assert(serd_reader_read_file(reader, USTR("http://notafile")));
- assert(serd_reader_read_file(reader, USTR("file:///better/not/exist")));
- assert(serd_reader_read_file(reader, USTR("file://")));
-
- const SerdStatus st = serd_reader_read_file(reader, USTR(path));
- assert(!st);
- assert(rt->n_statements == 13);
- assert(rt->graph && rt->graph->buf &&
- !strcmp((const char*)rt->graph->buf, "http://example.org/"));
-
- assert(serd_reader_read_string(reader, USTR("This isn't Turtle at all.")));
-
- // A read of a big page hits EOF then fails to read chunks immediately
- {
- FILE* temp = tmpfile();
- assert(temp);
- fprintf(temp, "_:s <http://example.org/p> _:o .\n");
- fflush(temp);
- fseek(temp, 0L, SEEK_SET);
-
- serd_reader_start_stream(reader, temp, NULL, true);
-
- assert(serd_reader_read_chunk(reader) == SERD_SUCCESS);
- assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
- assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
-
- serd_reader_end_stream(reader);
- fclose(temp);
- }
-
- // A byte-wise reader that hits EOF once then continues (like a socket)
- {
- size_t n_reads = 0;
- serd_reader_start_source_stream(reader,
- (SerdSource)eof_test_read,
- (SerdStreamErrorFunc)eof_test_error,
- &n_reads,
- NULL,
- 1);
-
- assert(serd_reader_read_chunk(reader) == SERD_SUCCESS);
- assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
- assert(serd_reader_read_chunk(reader) == SERD_SUCCESS);
- assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
- }
-
- serd_reader_free(reader);
+ assert(serd_reader_read_file(reader, USTR("http://notafile")));
+ assert(serd_reader_read_file(reader, USTR("file:///better/not/exist")));
+ assert(serd_reader_read_file(reader, USTR("file://")));
+
+ const SerdStatus st = serd_reader_read_file(reader, USTR(path));
+ assert(!st);
+ assert(rt->n_statements == 13);
+ assert(rt->graph && rt->graph->buf &&
+ !strcmp((const char*)rt->graph->buf, "http://example.org/"));
+
+ assert(serd_reader_read_string(reader, USTR("This isn't Turtle at all.")));
+
+ // A read of a big page hits EOF then fails to read chunks immediately
+ {
+ FILE* temp = tmpfile();
+ assert(temp);
+ fprintf(temp, "_:s <http://example.org/p> _:o .\n");
+ fflush(temp);
+ fseek(temp, 0L, SEEK_SET);
+
+ serd_reader_start_stream(reader, temp, NULL, true);
+
+ assert(serd_reader_read_chunk(reader) == SERD_SUCCESS);
+ assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
+ assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
+
+ serd_reader_end_stream(reader);
+ fclose(temp);
+ }
+
+ // A byte-wise reader that hits EOF once then continues (like a socket)
+ {
+ size_t n_reads = 0;
+ serd_reader_start_source_stream(reader,
+ (SerdSource)eof_test_read,
+ (SerdStreamErrorFunc)eof_test_error,
+ &n_reads,
+ NULL,
+ 1);
+
+ assert(serd_reader_read_chunk(reader) == SERD_SUCCESS);
+ assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
+ assert(serd_reader_read_chunk(reader) == SERD_SUCCESS);
+ assert(serd_reader_read_chunk(reader) == SERD_FAILURE);
+ }
+
+ serd_reader_free(reader);
}
int
main(void)
{
- test_read_chunks();
- test_read_string();
+ test_read_chunks();
+ test_read_string();
- const char* const path = "serd_test.ttl";
- test_writer(path);
- test_reader(path);
+ const char* const path = "serd_test.ttl";
+ test_writer(path);
+ test_reader(path);
- printf("Success\n");
- return 0;
+ printf("Success\n");
+ return 0;
}
diff --git a/test/test_string.c b/test/test_string.c
index 1e11f5b4..225eca4f 100644
--- a/test/test_string.c
+++ b/test/test_string.c
@@ -26,39 +26,39 @@
static void
test_strlen(void)
{
- const uint8_t str[] = { '"', '5', 0xE2, 0x82, 0xAC, '"', '\n', 0 };
+ const uint8_t str[] = {'"', '5', 0xE2, 0x82, 0xAC, '"', '\n', 0};
- size_t n_bytes = 0;
- SerdNodeFlags flags = 0;
- size_t len = serd_strlen(str, &n_bytes, &flags);
- assert(len == 5 && n_bytes == 7 &&
- flags == (SERD_HAS_QUOTE | SERD_HAS_NEWLINE));
- len = serd_strlen(str, NULL, &flags);
- assert(len == 5);
+ size_t n_bytes = 0;
+ SerdNodeFlags flags = 0;
+ size_t len = serd_strlen(str, &n_bytes, &flags);
+ assert(len == 5 && n_bytes == 7 &&
+ flags == (SERD_HAS_QUOTE | SERD_HAS_NEWLINE));
+ len = serd_strlen(str, NULL, &flags);
+ assert(len == 5);
- assert(serd_strlen(str, &n_bytes, NULL) == 5);
+ assert(serd_strlen(str, &n_bytes, NULL) == 5);
}
static void
test_strerror(void)
{
- const uint8_t* msg = serd_strerror(SERD_SUCCESS);
- assert(!strcmp((const char*)msg, "Success"));
- for (int i = SERD_FAILURE; i <= SERD_ERR_INTERNAL; ++i) {
- msg = serd_strerror((SerdStatus)i);
- assert(strcmp((const char*)msg, "Success"));
- }
+ const uint8_t* msg = serd_strerror(SERD_SUCCESS);
+ assert(!strcmp((const char*)msg, "Success"));
+ for (int i = SERD_FAILURE; i <= SERD_ERR_INTERNAL; ++i) {
+ msg = serd_strerror((SerdStatus)i);
+ assert(strcmp((const char*)msg, "Success"));
+ }
- msg = serd_strerror((SerdStatus)-1);
- assert(!strcmp((const char*)msg, "Unknown error"));
+ msg = serd_strerror((SerdStatus)-1);
+ assert(!strcmp((const char*)msg, "Unknown error"));
}
int
main(void)
{
- test_strlen();
- test_strerror();
+ test_strlen();
+ test_strerror();
- printf("Success\n");
- return 0;
+ printf("Success\n");
+ return 0;
}
diff --git a/test/test_uri.c b/test/test_uri.c
index 9538ffb0..f910e1f0 100644
--- a/test/test_uri.c
+++ b/test/test_uri.c
@@ -33,151 +33,153 @@ test_file_uri(const char* hostname,
const char* expected_uri,
const char* expected_path)
{
- if (!expected_path) {
- expected_path = path;
- }
-
- SerdNode node = serd_node_new_file_uri(
- USTR(path), USTR(hostname), 0, escape);
-
- uint8_t* out_hostname = NULL;
- uint8_t* out_path = serd_file_uri_parse(node.buf, &out_hostname);
- assert(!strcmp((const char*)node.buf, expected_uri));
- assert((hostname && out_hostname) || (!hostname && !out_hostname));
- assert(!strcmp((const char*)out_path, (const char*)expected_path));
-
- serd_free(out_path);
- serd_free(out_hostname);
- serd_node_free(&node);
+ if (!expected_path) {
+ expected_path = path;
+ }
+
+ SerdNode node = serd_node_new_file_uri(USTR(path), USTR(hostname), 0, escape);
+
+ uint8_t* out_hostname = NULL;
+ uint8_t* out_path = serd_file_uri_parse(node.buf, &out_hostname);
+ assert(!strcmp((const char*)node.buf, expected_uri));
+ assert((hostname && out_hostname) || (!hostname && !out_hostname));
+ assert(!strcmp((const char*)out_path, (const char*)expected_path));
+
+ serd_free(out_path);
+ serd_free(out_hostname);
+ serd_node_free(&node);
}
#if defined(__GNUC__)
-# pragma GCC diagnostic push
-# pragma GCC diagnostic ignored "-Wdeprecated-declarations"
+# pragma GCC diagnostic push
+# pragma GCC diagnostic ignored "-Wdeprecated-declarations"
#endif
static void
test_uri_to_path(void)
{
- const uint8_t* uri = (const uint8_t*)"file:///home/user/foo.ttl";
- assert(!strcmp((const char*)serd_uri_to_path(uri), "/home/user/foo.ttl"));
+ const uint8_t* uri = (const uint8_t*)"file:///home/user/foo.ttl";
+ assert(!strcmp((const char*)serd_uri_to_path(uri), "/home/user/foo.ttl"));
- uri = (const uint8_t*)"file://localhost/home/user/foo.ttl";
- assert(!strcmp((const char*)serd_uri_to_path(uri), "/home/user/foo.ttl"));
+ uri = (const uint8_t*)"file://localhost/home/user/foo.ttl";
+ assert(!strcmp((const char*)serd_uri_to_path(uri), "/home/user/foo.ttl"));
- uri = (const uint8_t*)"file:illegal/file/uri";
- assert(!serd_uri_to_path(uri));
+ uri = (const uint8_t*)"file:illegal/file/uri";
+ assert(!serd_uri_to_path(uri));
- uri = (const uint8_t*)"file:///c:/awful/system";
- assert(!strcmp((const char*)serd_uri_to_path(uri), "c:/awful/system"));
+ uri = (const uint8_t*)"file:///c:/awful/system";
+ assert(!strcmp((const char*)serd_uri_to_path(uri), "c:/awful/system"));
- uri = (const uint8_t*)"file:///c:awful/system";
- assert(!strcmp((const char*)serd_uri_to_path(uri), "/c:awful/system"));
+ uri = (const uint8_t*)"file:///c:awful/system";
+ assert(!strcmp((const char*)serd_uri_to_path(uri), "/c:awful/system"));
- uri = (const uint8_t*)"file:///0/1";
- assert(!strcmp((const char*)serd_uri_to_path(uri), "/0/1"));
+ uri = (const uint8_t*)"file:///0/1";
+ assert(!strcmp((const char*)serd_uri_to_path(uri), "/0/1"));
- uri = (const uint8_t*)"C:\\Windows\\Sucks";
- assert(!strcmp((const char*)serd_uri_to_path(uri), "C:\\Windows\\Sucks"));
+ uri = (const uint8_t*)"C:\\Windows\\Sucks";
+ assert(!strcmp((const char*)serd_uri_to_path(uri), "C:\\Windows\\Sucks"));
- uri = (const uint8_t*)"C|/Windows/Sucks";
- assert(!strcmp((const char*)serd_uri_to_path(uri), "C|/Windows/Sucks"));
+ uri = (const uint8_t*)"C|/Windows/Sucks";
+ assert(!strcmp((const char*)serd_uri_to_path(uri), "C|/Windows/Sucks"));
- uri = (const uint8_t*)"http://example.org/path";
- assert(!serd_uri_to_path(uri));
+ uri = (const uint8_t*)"http://example.org/path";
+ assert(!serd_uri_to_path(uri));
}
#if defined(__GNUC__)
-# pragma GCC diagnostic pop
+# pragma GCC diagnostic pop
#endif
static void
test_uri_parsing(void)
{
- test_file_uri(NULL, "C:/My 100%", true,
- "file:///C:/My%20100%%", NULL);
- test_file_uri("ahost", "C:\\Pointless Space", true,
- "file://ahost/C:/Pointless%20Space", "C:/Pointless Space");
- test_file_uri(NULL, "/foo/bar", true,
- "file:///foo/bar", NULL);
- test_file_uri("bhost", "/foo/bar", true,
- "file://bhost/foo/bar", NULL);
- test_file_uri(NULL, "a/relative path", false,
- "a/relative path", NULL);
- test_file_uri(NULL, "a/relative <path>", true,
- "a/relative%20%3Cpath%3E", NULL);
-
- // Test tolerance of parsing junk URI escapes
-
- uint8_t* out_path = serd_file_uri_parse(USTR("file:///foo/%0Xbar"), NULL);
- assert(!strcmp((const char*)out_path, "/foo/bar"));
- serd_free(out_path);
+ test_file_uri(NULL, "C:/My 100%", true, "file:///C:/My%20100%%", NULL);
+ test_file_uri("ahost",
+ "C:\\Pointless Space",
+ true,
+ "file://ahost/C:/Pointless%20Space",
+ "C:/Pointless Space");
+ test_file_uri(NULL, "/foo/bar", true, "file:///foo/bar", NULL);
+ test_file_uri("bhost", "/foo/bar", true, "file://bhost/foo/bar", NULL);
+ test_file_uri(NULL, "a/relative path", false, "a/relative path", NULL);
+ test_file_uri(
+ NULL, "a/relative <path>", true, "a/relative%20%3Cpath%3E", NULL);
+
+ // Test tolerance of parsing junk URI escapes
+
+ uint8_t* out_path = serd_file_uri_parse(USTR("file:///foo/%0Xbar"), NULL);
+ assert(!strcmp((const char*)out_path, "/foo/bar"));
+ serd_free(out_path);
}
static void
test_uri_from_string(void)
{
- SerdNode nonsense = serd_node_new_uri_from_string(NULL, NULL, NULL);
- assert(nonsense.type == SERD_NOTHING);
-
- SerdURI base_uri;
- SerdNode base = serd_node_new_uri_from_string(USTR("http://example.org/"),
- NULL, &base_uri);
- SerdNode nil = serd_node_new_uri_from_string(NULL, &base_uri, NULL);
- SerdNode nil2 = serd_node_new_uri_from_string(USTR(""), &base_uri, NULL);
- assert(nil.type == SERD_URI);
- assert(!strcmp((const char*)nil.buf, (const char*)base.buf));
- assert(nil2.type == SERD_URI);
- assert(!strcmp((const char*)nil2.buf, (const char*)base.buf));
- serd_node_free(&nil);
- serd_node_free(&nil2);
-
- serd_node_free(&base);
+ SerdNode nonsense = serd_node_new_uri_from_string(NULL, NULL, NULL);
+ assert(nonsense.type == SERD_NOTHING);
+
+ SerdURI base_uri;
+ SerdNode base =
+ serd_node_new_uri_from_string(USTR("http://example.org/"), NULL, &base_uri);
+ SerdNode nil = serd_node_new_uri_from_string(NULL, &base_uri, NULL);
+ SerdNode nil2 = serd_node_new_uri_from_string(USTR(""), &base_uri, NULL);
+ assert(nil.type == SERD_URI);
+ assert(!strcmp((const char*)nil.buf, (const char*)base.buf));
+ assert(nil2.type == SERD_URI);
+ assert(!strcmp((const char*)nil2.buf, (const char*)base.buf));
+ serd_node_free(&nil);
+ serd_node_free(&nil2);
+
+ serd_node_free(&base);
}
static void
test_relative_uri(void)
{
- SerdURI base_uri;
- SerdNode base = serd_node_new_uri_from_string(USTR("http://example.org/"),
- NULL, &base_uri);
-
- SerdNode abs = serd_node_from_string(SERD_URI, USTR("http://example.org/foo/bar"));
- SerdURI abs_uri;
- serd_uri_parse(abs.buf, &abs_uri);
-
- SerdURI rel_uri;
- SerdNode rel = serd_node_new_relative_uri(&abs_uri, &base_uri, NULL, &rel_uri);
- assert(!strcmp((const char*)rel.buf, "/foo/bar"));
-
- SerdNode up = serd_node_new_relative_uri(&base_uri, &abs_uri, NULL, NULL);
- assert(!strcmp((const char*)up.buf, "../"));
-
- SerdNode noup = serd_node_new_relative_uri(&base_uri, &abs_uri, &abs_uri, NULL);
- assert(!strcmp((const char*)noup.buf, "http://example.org/"));
-
- SerdNode x = serd_node_from_string(SERD_URI, USTR("http://example.org/foo/x"));
- SerdURI x_uri;
- serd_uri_parse(x.buf, &x_uri);
-
- SerdNode x_rel = serd_node_new_relative_uri(&x_uri, &abs_uri, &abs_uri, NULL);
- assert(!strcmp((const char*)x_rel.buf, "x"));
-
- serd_node_free(&x_rel);
- serd_node_free(&noup);
- serd_node_free(&up);
- serd_node_free(&rel);
- serd_node_free(&base);
+ SerdURI base_uri;
+ SerdNode base =
+ serd_node_new_uri_from_string(USTR("http://example.org/"), NULL, &base_uri);
+
+ SerdNode abs =
+ serd_node_from_string(SERD_URI, USTR("http://example.org/foo/bar"));
+ SerdURI abs_uri;
+ serd_uri_parse(abs.buf, &abs_uri);
+
+ SerdURI rel_uri;
+ SerdNode rel =
+ serd_node_new_relative_uri(&abs_uri, &base_uri, NULL, &rel_uri);
+ assert(!strcmp((const char*)rel.buf, "/foo/bar"));
+
+ SerdNode up = serd_node_new_relative_uri(&base_uri, &abs_uri, NULL, NULL);
+ assert(!strcmp((const char*)up.buf, "../"));
+
+ SerdNode noup =
+ serd_node_new_relative_uri(&base_uri, &abs_uri, &abs_uri, NULL);
+ assert(!strcmp((const char*)noup.buf, "http://example.org/"));
+
+ SerdNode x =
+ serd_node_from_string(SERD_URI, USTR("http://example.org/foo/x"));
+ SerdURI x_uri;
+ serd_uri_parse(x.buf, &x_uri);
+
+ SerdNode x_rel = serd_node_new_relative_uri(&x_uri, &abs_uri, &abs_uri, NULL);
+ assert(!strcmp((const char*)x_rel.buf, "x"));
+
+ serd_node_free(&x_rel);
+ serd_node_free(&noup);
+ serd_node_free(&up);
+ serd_node_free(&rel);
+ serd_node_free(&base);
}
int
main(void)
{
- test_uri_to_path();
- test_uri_parsing();
- test_uri_from_string();
- test_relative_uri();
+ test_uri_to_path();
+ test_uri_parsing();
+ test_uri_from_string();
+ test_relative_uri();
- printf("Success\n");
- return 0;
+ printf("Success\n");
+ return 0;
}
generated by cgit v1.2.1 (git 2.18.0) at 2024-11-06 03:19:49 +0000