Rename 'lex' to the now more appropriate 'parse'.

git-svn-id: http://svn.drobilla.net/resp/resp@308 ad02d1e2-f140-0410-9f75-f8b11f17cedd

1 files changed, 207 insertions, 0 deletions

diff --git a/src/parse.cpp b/src/parse.cpp
new file mode 100644
index 0000000..2ac838f
--- /dev/null
+++ b/src/parse.cpp
@@ -0,0 +1,207 @@
+/* Resp: A programming language
+ * Copyright (C) 2008-2009 David Robillard <dave@drobilla.net>
+ *
+ * Resp is free software: you can redistribute it and/or modify it under
+ * the terms of the GNU Affero General Public License as published by the
+ * Free Software Foundation, either version 3 of the License, or (at your
+ * option) any later version.
+ *
+ * Resp is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Affero General
+ * Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with Resp.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/** @file
+ * @brief Parsing (build an AST from text)
+ */
+
+#include <stdio.h>
+#include <stack>
+#include "resp.hpp"
+
+using namespace std;
+
+static inline int
+read_char(Cursor& cur, istream& in)
+{
+	int ch = in.get();
+	switch (ch) {
+	case '\n': ++cur.line; cur.col = 0; break;
+	default:   ++cur.col;
+	}
+	return ch;
+}
+
+static inline void
+skip_space(Cursor& cur, istream& in)
+{
+	while (isspace(in.peek()))
+		read_char(cur, in);
+}
+
+static inline void
+eat_char(Cursor& cur, istream& in, const char character)
+{
+	const char c = read_char(cur, in);
+	assert(c == character);
+	return;
+}
+
+static AST*
+read_string(Cursor& cur, istream& in)
+{
+	string str;
+	char c;
+	Cursor loc = cur;
+	eat_char(cur, in, '"');
+	while ((c = read_char(cur, in)) != '"') {
+		if (c == '\\') { // string escape
+			switch (c = read_char(cur, in)) {
+			case '"':
+				str.push_back('"');
+				break;
+			case '\\':
+				str.push_back('\\');
+				break;
+			default:
+				cin.putback(c);
+				throw Error(cur, string("unknown string escape `\\") + (char)c + "'");
+			}
+		} else { // any other character
+			str.push_back(c);
+		}
+	}
+	return new AString(loc, str);
+}
+
+static AST*
+read_line_comment(Cursor& cur, istream& in)
+{
+	char c;
+	while ((c = read_char(cur, in)) != '\n') {}
+	return NULL;
+}
+
+static AST*
+read_list(PEnv& penv, Cursor& cur, istream& in)
+{
+	List<ATuple, AST> list;
+
+	eat_char(cur, in, '(');
+	while (true) {
+		skip_space(cur, in);
+		if (in.peek() == ')') {
+			eat_char(cur, in, ')');
+			return list.head;
+		}
+
+		list.push_back(read_expression(penv, cur, in));
+	}
+	assert(false);
+}
+
+static AST*
+read_special(Cursor& cur, istream& in)
+{
+	eat_char(cur, in, '#');
+	switch (in.peek()) {
+	case '|':
+		while (!(read_char(cur, in) == '|' && read_char(cur, in) == '#')) {}
+		return NULL;
+	case 't':
+		eat_char(cur, in, 't');
+		return new ALiteral<bool>(T_BOOL, true, cur);
+	case 'f':
+		return new ALiteral<bool>(T_BOOL, false, cur);
+	default:
+		throw Error(cur, (format("unknown special lexeme `%1%'") % in.peek()).str());
+	}
+	assert(false);
+	return NULL;
+}
+
+static AST*
+read_number(Cursor& cur, istream& in)
+{
+	string str;
+	char c;
+	Cursor loc = cur;
+	while ((c = in.peek()) != EOF) {
+		if (isdigit(c) || c == '.')
+			str += read_char(cur, in);
+		else
+			break;
+	}
+
+	if (str.find('.') == string::npos)
+		return new ALiteral<int32_t>(T_INT32, strtol(str.c_str(), NULL, 10), loc);
+	else
+		return new ALiteral<float>(T_FLOAT, strtod(str.c_str(), NULL), loc);
+}
+
+static AST*
+read_symbol(PEnv& penv, Cursor& cur, istream& in)
+{
+	string str;
+	char c;
+	Cursor loc = cur;
+	while ((c = in.peek()) != EOF) {
+		if (!isspace(c) && c != ')' && c != '(' && c != EOF && c != -1) {
+			str += read_char(cur, in);
+		} else {
+			break;
+		}
+	}
+
+	return penv.sym(str);
+}
+	
+/// Read an expression from @a in
+AST*
+read_expression(PEnv& penv, Cursor& cur, istream& in)
+{
+	while (!cin.eof()) {
+		skip_space(cur, in);
+		const char c = in.peek();
+		switch (c) {
+		case EOF:
+			return NULL;
+		case ';':
+			read_line_comment(cur, in);
+			break;
+		case '"':
+			return read_string(cur, in);
+		case '(':
+			return read_list(penv, cur, in);
+		case ')':
+			throw Error(cur, "unexpected `)'");
+		case '#':
+		{
+			AST* ret = read_special(cur, in);
+			if (ret)
+				return ret;
+			break;
+		}
+		case '-':
+		case '+':
+			read_char(cur, in);
+			if (isdigit(in.peek())) {
+				in.putback(c);
+				return read_number(cur, in);
+			} else {
+				in.putback(c);
+				return read_symbol(penv, cur, in);
+			}
+		default:
+			if (isdigit(c))
+				return read_number(cur, in);
+			else
+				return read_symbol(penv, cur, in);
+		}
+	}
+	return NULL;
+}