1 /* The lexer multiplexer for Gedcom.
2 Copyright (C) 2001,2002 The Genes Development Team
3 This file is part of the Gedcom parser library.
4 Contributed by Peter Verthez <Peter.Verthez@advalvas.be>, 2001.
6 The Gedcom parser library is free software; you can redistribute it
7 and/or modify it under the terms of the GNU Lesser General Public
8 License as published by the Free Software Foundation; either
9 version 2.1 of the License, or (at your option) any later version.
11 The Gedcom parser library is distributed in the hope that it will be
12 useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 Lesser General Public License for more details.
16 You should have received a copy of the GNU Lesser General Public
17 License along with the Gedcom parser library; if not, write to the
18 Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
24 #include "gedcom_internal.h"
31 typedef int (*lex_func)(void);
34 #define NEW_MODEL_FILE "new.ged"
36 int lexer_init(Encoding enc, FILE* f)
38 if (enc == ONE_BYTE) {
39 lf = &gedcom_1byte_lex;
40 gedcom_1byte_myinit(f);
41 set_encoding_width(enc);
42 return open_conv_to_internal("ASCII");
44 else if (enc == TWO_BYTE_HILO) {
45 lf = &gedcom_hilo_lex;
46 gedcom_hilo_myinit(f);
47 set_encoding_width(enc);
48 return open_conv_to_internal("UNICODE");
50 else if (enc == TWO_BYTE_LOHI) {
51 lf = &gedcom_lohi_lex;
52 gedcom_lohi_myinit(f);
53 set_encoding_width(enc);
54 return open_conv_to_internal("UNICODE");
63 close_conv_to_internal();
71 void rewind_file(FILE* f)
73 if (fseek(f, 0, 0) != 0)
74 gedcom_warning(_("Error positioning input file: %s"), strerror(errno));
77 int determine_encoding(FILE* f)
82 read = fread(first, 1, 2, f);
84 gedcom_warning(_("Error reading from input file: %s"), strerror(errno));
88 else if ((first[0] == '0') && (first[1] == ' ')) {
89 gedcom_debug_print(_("One-byte encoding"));
93 else if ((first[0] == '\0') && (first[1] == '0')) {
94 gedcom_debug_print(_("Two-byte encoding, high-low"));
98 else if ((first[0] == '\xFE') && (first[1] == '\xFF')) {
99 gedcom_debug_print(_("Two-byte encoding, high-low, with BOM"));
100 return TWO_BYTE_HILO;
102 else if ((first[0] == '0') && (first[1] == '\0')) {
103 gedcom_debug_print(_("Two-byte encoding, low-high"));
105 return TWO_BYTE_LOHI;
107 else if ((first[0] == '\xFF') && (first[1] == '\xFE')) {
108 gedcom_debug_print(_("Two-byte encoding, low-high, with BOM"));
109 return TWO_BYTE_LOHI;
111 else if ((first[0] == '\xEF') && (first[1] == '\xBB')) {
112 read = fread(first, 1, 1, f);
114 gedcom_warning(_("Error reading from input file: %s"), strerror(errno));
117 else if (first[0] == '\xBF') {
118 gedcom_debug_print(_("UTF-8 encoding, with BOM"));
121 gedcom_warning(_("Unknown encoding, falling back to one-byte"));
127 gedcom_warning(_("Unknown encoding, falling back to one-byte"));
138 update_gconv_search_path();
140 if (!setlocale(LC_ALL, "")
141 || ! bindtextdomain(PACKAGE, LOCALEDIR)
142 || ! bind_textdomain_codeset(PACKAGE, INTERNAL_ENCODING))
148 int gedcom_parse_file(const char* file_name)
155 gedcom_error(_("Internal error: GEDCOM parser not initialized"));
158 file = fopen(file_name, "r");
160 gedcom_error(_("Could not open file '%s': %s"),
161 file_name, strerror(errno));
165 enc = determine_encoding(file);
167 if (lexer_init(enc, file)) {
170 result = gedcom_parse();
173 result = check_xref_table();
183 int gedcom_new_model()
188 file = fopen(NEW_MODEL_FILE, "r");
191 result = gedcom_parse_file(NEW_MODEL_FILE);
194 char* filename = (char*) malloc(strlen(PKGDATADIR) + strlen(NEW_MODEL_FILE)
199 sprintf(filename, "%s/%s", PKGDATADIR, NEW_MODEL_FILE);
200 result = gedcom_parse_file(filename);
207 int gedcom_check_version(int major, int minor, int patch)
209 if (major < GEDCOM_PARSE_VERSION_MAJOR)
211 else if (major > GEDCOM_PARSE_VERSION_MAJOR)
213 else if (minor < GEDCOM_PARSE_VERSION_MINOR)
215 else if (minor > GEDCOM_PARSE_VERSION_MINOR)
217 else if (patch <= GEDCOM_PARSE_VERSION_PATCH)