--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/ueda/sverp/lexer.c	Mon Jul 20 00:24:37 2015 +0000
@@ -0,0 +1,215 @@
+/*
+ * This module contains the lexer underlying the parser for the minimal
+ * subset of Verilog we grok; the same lexer will be used to read and parse
+ * the primitives definition file.
+ */
+
+#include <ctype.h>
+#include <stdio.h>
+#include "lexer.h"
+
+char *parser_filename;
+FILE *parser_readF;
+int parser_lineno;
+char parser_read_word[MAXWORD+1];
+int parser_read_number;
+int pushback_token;
+
+void
+parse_error(msg)
+	char *msg;
+{
+	fprintf(stderr, "%s line %d: %s\n", parser_filename, parser_lineno,
+		msg);
+	exit(1);
+}
+
+static
+my_getchar()
+{
+	register int c;
+
+	c = getc(parser_readF);
+	if (c < 0)
+		return(c);
+	if (!isascii(c))
+		parse_error("non-ASCII character");
+	if (iscntrl(c) && c != '\n' && c != '\t')
+		parse_error("invalid control character");
+	return(c);
+}
+
+static void
+handle_trad_comment()
+{
+	register int c, flag;
+
+	for (flag = 0; ; ) {
+		c = my_getchar();
+		if (c < 0)
+			parse_error("/* comment ends in EOF");
+		if (c == '\n')
+			parser_lineno++;
+		if (c == '/' && flag)
+			return;
+		flag = (c == '*');
+	}
+}
+
+static void
+handle_line_comment()
+{
+	register int c;
+
+	for (;;) {
+		c = my_getchar();
+		if (c < 0)
+			parse_error("// comment ends in EOF");
+		if (c == '\n') {
+			parser_lineno++;
+			return;
+		}
+	}
+}
+
+static void
+handle_comment()
+{
+	int c;
+
+	c = my_getchar();
+	switch (c) {
+	case '*':
+		/* traditional C comment style */
+		handle_trad_comment();
+		return;
+	case '/':
+		/* new-fangled double slash comment style */
+		handle_line_comment();
+		return;
+	default:
+		parse_error("character after '/' is not '*' or '/'");
+		exit(1);
+	}
+}
+
+static void
+handle_num_token(first_digit)
+{
+	register int c, n;
+
+	parser_read_number = first_digit - '0';
+	for (n = 1; ; n++) {
+		c = my_getchar();
+		if (!isdigit(c))
+			break;
+		parser_read_number *= 10;
+		parser_read_number += c - '0';
+	}
+	if (c >= 0) {
+		if (isalpha(c) || c == '_')
+			parse_error(
+			"digits followed by letters: neither word nor number");
+		ungetc(c, parser_readF);
+	}
+	if (n > MAXDIGITS)
+		parse_error("number is too long (MAXDIGITS exceeded)");
+}
+
+static void
+handle_word_token(first_char)
+{
+	register int c;
+	register char *cp;
+	register int len;
+
+	cp = parser_read_word;
+	*cp++ = first_char;
+	for (len = 1; ; ) {
+		c = my_getchar();
+		if (!isalnum(c) && c != '_' && c != '$')
+			break;
+		if (len >= MAXWORD)
+			parse_error("text token is too long");
+		*cp++ = c;
+		len++;
+	}
+	*cp = '\0';
+	if (c < 0)
+		return;
+	ungetc(c, parser_readF);
+}
+
+static void
+handle_qstr()
+{
+	register int c;
+	register char *cp;
+	register int len;
+
+	cp = parser_read_word;
+	for (len = 0; ; ) {
+		c = my_getchar();
+		if (c == EOF || c == '\n')
+unterm:			parse_error("unterminated quoted string");
+		if (c == '"')
+			break;
+		if (c == '\\') {
+			c = my_getchar();
+			if (c == EOF || c == '\n')
+				goto unterm;
+		}
+		if (len >= MAXWORD)
+			parse_error("quoted string is too long");
+		*cp++ = c;
+		len++;
+	}
+	*cp = '\0';
+}
+
+get_token()
+{
+	register int c;
+
+	if (c = pushback_token) {
+		pushback_token = 0;
+		return(c);
+	}
+loop:	c = my_getchar();
+	switch (c) {
+	case EOF:
+		return(0);
+	case ' ':
+	case '\t':
+		goto loop;
+	case '\n':
+		parser_lineno++;
+		goto loop;
+	case '/':
+		handle_comment();
+		goto loop;
+	case '(':
+	case ')':
+	case ',':
+	case '.':
+	case ':':
+	case ';':
+	case '[':
+	case ']':
+		return(c);
+	case '"':
+		handle_qstr();
+		return(QSTRING);
+	}
+	if (isdigit(c)) {
+		handle_num_token(c);
+		return(NUMBER);
+	}
+	if (isalpha(c) || c == '_') {
+		handle_word_token(c);
+		return(WORD);
+	}
+	fprintf(stderr, "%s line %d: bad character \'%c\'\n", parser_filename,
+		parser_lineno, c);
+	exit(1);
+}