aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2020-06-21Add stack overflow testsDavid Robillard2-1/+72
2020-06-21Improve node test coverageDavid Robillard1-0/+6
2020-06-21Add test for bad blank graph labelsDavid Robillard2-0/+9
2020-06-21Fix writer flags typeDavid Robillard1-18/+18
2020-06-21Simplify reader stack pushing codeDavid Robillard1-8/+10
2020-06-21Ensure serd_new_resolved_uri returns an absolute URI or NULLDavid Robillard3-8/+17
2020-06-21Use simpler names for statement flagsDavid Robillard3-26/+24
2020-06-21Simplify streaming API and improve pretty printingDavid Robillard9-168/+327
This removes the obligation from the caller to correctly maintain flags to describe the current anonymous context, instead making the writer handle this itself as much as possible. Flags remain for the cases the writer can not infer from context: the start of anonymous subject and object nodes.
2020-06-21Cache commonly used nodes in the worldDavid Robillard3-14/+32
2020-06-21Add SerdNodes class for storing a cache of nodesDavid Robillard8-4/+306
2020-06-21Add serd_env_copy() and serd_env_equals()David Robillard3-0/+66
2020-06-21Separate ByteSink from WriterDavid Robillard4-17/+25
2020-06-21Expose SerdByteSink in public APIDavid Robillard5-102/+146
2020-06-21Remove serd_reader_set_default_graph()David Robillard4-34/+2
2020-06-21Add zix data structures for use in modelDavid Robillard7-0/+1621
2020-06-21Add rooted URI resolution testDavid Robillard1-1/+2
2020-06-21Add test for reading chunksDavid Robillard2-1/+115
2020-06-21Make reader return failure when no statement is readDavid Robillard2-3/+3
2020-06-21Fix warningsDavid Robillard2-4/+4
2020-06-21Simplify writer style optionsDavid Robillard6-40/+35
2020-06-21Avoid writing invalid prefixed name prefixesDavid Robillard1-1/+2
2020-06-21Separate style enum from style flags typesDavid Robillard4-6/+11
2020-06-21Automatically finish reader on destructionDavid Robillard2-0/+6
2020-06-21Make SerdSink opaqueDavid Robillard7-27/+158
2020-06-21Simplify SerdEnv APIDavid Robillard5-80/+118
2020-06-21Add cursor to statementDavid Robillard5-3/+21
This allows the file and position of statements to be used in sinks, for example for more helpful error reporting outside the parser.
2020-06-21Add SerdCursor to public APIDavid Robillard10-37/+292
2020-06-21Remove escape parameter from serd_new_file_uriDavid Robillard4-16/+12
Since characters are escaped because they are not valid characters in a URI, any use of this function without escaping is problematic at best.
2020-06-21Shorten node constructor namesDavid Robillard7-179/+298
2020-06-21Make statement sink take a statement rather than nodesDavid Robillard5-41/+34
This makes the interface more extensible, towards associating more information with statements. The serd_sink_write_nodes wrapper remains so that user code does not need to allocate in order to write statement.
2020-06-21Add sink interfaceDavid Robillard3-0/+123
2020-06-21Add SerdStatementDavid Robillard4-0/+129
2020-06-21Expand datatype of literal nodes in serd_env_expand_nodeDavid Robillard1-0/+12
2020-06-21Move syntax name/extension utilities to public APIDavid Robillard4-49/+109
2020-06-21Move SerdField to public APIDavid Robillard2-46/+50
2020-06-21Clean up reader error handlingDavid Robillard4-347/+372
2020-06-21Improve push_byte() performanceDavid Robillard1-4/+4
2020-06-21Add debug checks for node paddingDavid Robillard2-8/+36
2020-06-21Zero node padding before passing to reader sinksDavid Robillard5-3/+41
2020-06-21Remove datatype and language from reader contextDavid Robillard2-30/+17
2020-06-21Simplify stack management by popping in bulk at higher levelsDavid Robillard4-104/+70
Since all memory used by the reader is POD in the stack, there is no benefit to forcing code to explicitly pop everything pushed to the stack, since any function can record an offset and pop back down to it regardless of what its callers pushed if it knows that it does not need those items.
2020-06-21Use a fixed-size reader stackDavid Robillard12-230/+293
This improves performance, and makes the reader more suitable for embedded or network-facing applications, at the cost of requiring the user to specify a maximum stack size.
2020-06-21Set datatypes on integer, decimal, and base64 nodesDavid Robillard4-37/+128
2020-06-21Add serd_world_get_blank()David Robillard4-1/+50
2020-06-21Avoid copying nodes when equivalent prefixes are setDavid Robillard1-3/+5
2020-06-21Separate base64 implementationDavid Robillard5-89/+181
2020-06-21Use SerdNodeFlags type everywhereDavid Robillard1-9/+9
2020-06-21Clean up node construction APIDavid Robillard8-232/+322
2020-06-21Clean up includes and improve source file separationDavid Robillard10-11/+19
2020-06-21Move system utilities to separate source filesDavid Robillard11-61/+95