Implement parser for git-style config files.

* am/grecs.m4: New option: parser-git
* doc/GRECS_SETUP.3: Document new options.
* src/git-parser.c: New file.
* src/Make.am [GRECS_COND_GIT_PARSER]: Define GRECS_PARSER_GIT.
(GRECS_SRC): Add GRECS_PARSER_GIT.
* src/grecs.h (grecs_git_parser): New proto.
* src/txtacc.c (grecs_txtacc_free): Ignore NULL argument.

4 files changed, 308 insertions, 3 deletions

diff --git a/src/Make.am b/src/Make.am
index e05afc5..312c734 100644
--- a/src/Make.am
+++ b/src/Make.am
@@ -24,6 +24,10 @@ if GRECS_COND_BIND_PARSER
   GRECS_EXTRA_BIND = bind-gram.h
 endif
 
+if GRECS_COND_GIT_PARSER
+  GRECS_PARSER_GIT = git-parser.c
+endif
+
 GRECS_SRC = \
  asprintf.c\
  diag.c\
@@ -45,6 +49,7 @@ GRECS_SRC = \
  version.c\
  wordsplit.c\
  $(GRECS_PARSER_BIND)\
+ $(GRECS_PARSER_GIT)\
  $(GRECS_PARSER_META1)
 
 noinst_HEADERS =
diff --git a/src/git-parser.c b/src/git-parser.c
new file mode 100644
index 0000000..e0675ef
--- /dev/null
+++ b/src/git-parser.c
@@ -0,0 +1,297 @@
+/* Git-style configuration file parser for Grecs.
+   Copyright (C) 2011 Sergey Poznyakoff
+
+   Grecs is free software; you can redistribute it and/or modify it
+   under the terms of the GNU General Public License as published by the
+   Free Software Foundation; either version 3 of the License, or (at your
+   option) any later version.
+
+   Grecs 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 General Public License for more details.
+
+   You should have received a copy of the GNU General Public License along
+   with Grecs. If not, see <http://www.gnu.org/licenses/>. */
+
+#ifdef HAVE_CONFIG_H
+# include <config.h>
+#endif
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+#include <grecs.h>
+
+static FILE *infile;
+static int input_char;
+static struct grecs_txtacc *acc;
+
+#define TOK_EOF 0
+#define TOK_EQ '='
+#define TOK_SECTION 256
+#define TOK_KEYWORD 257
+#define TOK_VALUE   258
+#define TOK_ERR -1
+
+struct token {
+        int type;
+        char *buf;
+        char *tag;
+        char chbuf[2];
+        int putback;
+} tok;
+
+#define ISSPACE(c) (strchr(" \t\r\f\n", c) != NULL)
+#define ISIDENT(c) ((isascii(c) && isalnum(c)) || (c) == '_')
+#define ISINITIAL(c) ((isascii(c) && isalpha(c)) || (c) == '_')
+
+static int
+input()
+{
+        if (!infile || feof(infile))
+                return 0;
+        input_char = fgetc(infile);
+        if (input_char == '\n')
+                grecs_current_locus.line++;
+        return input_char;
+}
+
+static void
+unput()
+{
+        if (!input_char)
+                return;
+        if (input_char == '\n')
+                grecs_current_locus.line--;
+        ungetc(input_char, infile);
+}
+
+static void
+error_recovery()
+{
+        while (input() && input_char != '\n')
+                ;
+}
+
+static void
+collect_tag()
+{
+        while (input() &&
+               !(ISSPACE(input_char) || input_char == ']'))
+                grecs_txtacc_grow_char(acc, input_char);
+        grecs_txtacc_grow_char(acc, 0);
+        tok.tag = grecs_txtacc_finish(acc, 0);
+}
+
+static void
+collect_string()
+{
+        while (input()) {
+                if (input_char == '\\') {
+                        if (!input()) {
+                                grecs_error(&grecs_current_locus, 0,
+                                            "unexpected EOF in string");
+                                break;
+                        }
+                } else if (input_char == '"')
+                        break;
+                grecs_txtacc_grow_char(acc, input_char);
+        }
+        grecs_txtacc_grow_char(acc, 0);
+        tok.tag = grecs_txtacc_finish(acc, 0);
+}
+
+static void
+gettoken(void)
+{
+        int putback = tok.putback;
+        tok.putback = 0;
+        if (putback)
+                return;
+
+        tok.buf = tok.tag = NULL;
+        /* Skip whitespace */
+        while (input() && ISSPACE(input_char))
+                ;
+
+        if (input_char <= 0) {
+                tok.type = TOK_EOF;
+                return;
+        }
+        
+        if (input_char == '[') {
+                tok.type = TOK_SECTION;
+                while (input() &&
+                       !(ISSPACE(input_char) || input_char == ']'))
+                        grecs_txtacc_grow_char(acc, input_char);
+                grecs_txtacc_grow_char(acc, 0);
+                tok.buf = grecs_txtacc_finish(acc, 0);
+                if (input_char != ']') {
+                        while (input() && ISSPACE(input_char))
+                                ;
+                        if (input_char == '"')
+                                collect_string();
+                        else if (input_char != ']')
+                                collect_tag();
+                }
+                if (input_char != ']') {
+                        while (input() && ISSPACE(input_char))
+                                ;
+                        if (input_char != ']') {
+                                if (isascii(input_char) && isprint(input_char))
+                                        grecs_error(&grecs_current_locus, 0,
+                                               "expected `]' but found `%c'",
+                                                    input_char);
+                                else if (!input_char)
+                                        grecs_error(&grecs_current_locus, 0,
+                                               "expected `]' but found EOF");
+                                else
+                                        grecs_error(&grecs_current_locus, 0,
+                                               "expected `]' but found \\%03o",
+                                                    (unsigned char)input_char);
+                                tok.type = TOK_ERR;
+                        }
+                }
+                return;
+        }
+
+        if (ISINITIAL(input_char)) {
+                tok.type = TOK_KEYWORD;
+                do 
+                        grecs_txtacc_grow_char(acc, input_char);
+                while (input() && ISIDENT(input_char));
+                unput();
+                grecs_txtacc_grow_char(acc, 0);
+                tok.buf = grecs_txtacc_finish(acc, 0);
+                return;
+        }
+
+        tok.chbuf[0] = input_char;
+        tok.chbuf[1] = 0;
+        tok.buf = tok.chbuf;
+        tok.type = input_char;
+}
+
+static struct grecs_value *
+getvalue()
+{
+        int len;
+        struct grecs_value *val = grecs_malloc(sizeof(*val));
+        while (input() && ISSPACE(input_char) && input_char != '\n')
+                ;
+        if (input_char != '\n')
+                do
+                        grecs_txtacc_grow_char(acc, input_char);
+                while (input() && input_char != '\n');
+        grecs_txtacc_grow_char(acc, 0);
+        tok.type = TOK_VALUE;
+        tok.buf = grecs_txtacc_finish(acc, 1);
+        len = strlen(tok.buf);
+        while (len > 0 && ISSPACE(tok.buf[len-1]))
+                tok.buf[--len] = 0;
+        val->type = GRECS_TYPE_STRING;
+        val->v.string = tok.buf;
+        return val;
+}
+
+static int
+read_statement(struct grecs_node *parent)
+{
+        struct grecs_node *node;
+        
+        gettoken();
+        if (tok.type == TOK_EOF || tok.type == TOK_SECTION) {
+                tok.putback = 1;
+                return 0;
+        }
+        if (tok.type != TOK_KEYWORD) {
+                grecs_error(&grecs_current_locus, 0, "syntax error");
+                error_recovery();