X-Git-Url: https://git.dlugolecki.net.pl/?a=blobdiff_plain;f=gedcom%2Fgedcom_lex_common.c;h=1da7da2ee2a8f1eef5487325ba999d1702d8a36f;hb=aa8e5450d4e93e743e848a0ac045e04c94474347;hp=5d734c44689adb6e8fe9ae0b46c9bb15594af1f9;hpb=8093e53a57e174b019f07760f5bf815271ceee9b;p=gedcom-parse.git diff --git a/gedcom/gedcom_lex_common.c b/gedcom/gedcom_lex_common.c index 5d734c4..1da7da2 100644 --- a/gedcom/gedcom_lex_common.c +++ b/gedcom/gedcom_lex_common.c @@ -1,11 +1,22 @@ -/* This program 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 2 of the License, or * - * (at your option) any later version. * - - (C) 2001 by The Genes Development Team - Original author: Peter Verthez (Peter.Verthez@advalvas.be) -*/ +/* Common lexer code. + Copyright (C) 2001, 2002 The Genes Development Team + This file is part of the Gedcom parser library. + Contributed by Peter Verthez , 2001. + + The Gedcom parser library is free software; you can redistribute it + and/or modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The Gedcom parser library 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 + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the Gedcom parser library; if not, write to the + Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + 02111-1307 USA. */ /* $Id$ */ /* $Name$ */ @@ -16,9 +27,8 @@ #include "multilex.h" #include "encoding.h" #include "gedcom.h" -#include "gedcom.tab.h" - -#define YY_NO_UNPUT +#include "gedcom.tabgen.h" +#include "compat.h" static size_t encoding_width; static int current_level = -1; @@ -32,14 +42,21 @@ static char str_buf[MAXGEDCLINELEN * UTF_FACTOR + 1]; #ifdef LEXER_TEST YYSTYPE gedcom_lval; int line_no = 1; +int compat_at = 0; int gedcom_lex(); +void message_handler(Gedcom_msg_type type, char *msg) +{ + fprintf(stderr, "(%d) %s\n", type, msg); +} + int test_loop(ENCODING enc, char* code) { int tok, res; init_encodings(); set_encoding_width(enc); + gedcom_set_message_handler(message_handler); res = open_conv_to_internal(code); if (!res) { gedcom_error("Unable to open conversion context: %s", @@ -56,8 +73,8 @@ int test_loop(ENCODING enc, char* code) case DELIM: printf("DELIM "); break; case ANYCHAR: printf("%s ", gedcom_lval.string); break; case POINTER: printf("POINTER(%s) ", gedcom_lval.string); break; - case USERTAG: printf("USERTAG(%s) ", gedcom_lval.string); break; - default: printf("TAG(%s) ", gedcom_lval.string); break; + case USERTAG: printf("USERTAG(%s) ", gedcom_lval.tag.string); break; + default: printf("TAG(%s) ", gedcom_lval.tag.string); break; } tok = gedcom_lex(); } @@ -80,7 +97,7 @@ int test_loop(ENCODING enc, char* code) { if (line_len != (size_t)-1) { \ line_len += strlen(yytext); \ if (line_len > MAXGEDCLINELEN * encoding_width) { \ - gedcom_error("Line too long, max %d characters", \ + gedcom_error(_("Line too long, max %d characters allowed"), \ MAXGEDCLINELEN); \ line_len = (size_t)-1; \ return BADTOKEN; \ @@ -90,7 +107,8 @@ int test_loop(ENCODING enc, char* code) #define MKTAGACTION(THETAG) \ { CHECK_LINE_LEN; \ - gedcom_lval.string = TO_INTERNAL(yytext, tag_buf); \ + gedcom_lval.tag.string = TO_INTERNAL(yytext, tag_buf); \ + gedcom_lval.tag.value = TAG_##THETAG; \ BEGIN(NORMAL); \ return TAG_##THETAG; \ } @@ -153,7 +171,7 @@ int test_loop(ENCODING enc, char* code) #define ACTION_0_DIGITS \ - { gedcom_error ("Level number with leading zero"); \ + { gedcom_error (_("Level number with leading zero not allowed")); \ return BADTOKEN; \ } @@ -162,7 +180,7 @@ int test_loop(ENCODING enc, char* code) { int level = atoi(TO_INTERNAL(yytext, str_buf)); \ CHECK_LINE_LEN; \ if ((level < 0) || (level > MAXGEDCLEVEL)) { \ - gedcom_error ("Level number out of range [0..%d]", \ + gedcom_error (_("Level number out of range [0..%d]"), \ MAXGEDCLEVEL); \ return BADTOKEN; \ } \ @@ -180,8 +198,7 @@ int test_loop(ENCODING enc, char* code) } \ else { \ /* should never happen (error to GEDCOM spec) */ \ - gedcom_error ("GEDCOM level number is %d higher than " \ - "previous", \ + gedcom_error (_("GEDCOM level number is %d higher than previous"), \ level_diff); \ return BADTOKEN; \ } \ @@ -190,12 +207,13 @@ int test_loop(ENCODING enc, char* code) #define ACTION_ALPHANUM \ { if (strlen(yytext) > MAXGEDCTAGLEN * encoding_width) { \ - gedcom_error("Tag '%s' too long, max %d characters", \ + gedcom_error(_("Tag '%s' too long, max %d characters allowed"), \ yytext, MAXGEDCTAGLEN); \ return BADTOKEN; \ } \ CHECK_LINE_LEN; \ - gedcom_lval.string = TO_INTERNAL(yytext, tag_buf); \ + gedcom_lval.tag.string = TO_INTERNAL(yytext, tag_buf); \ + gedcom_lval.tag.value = USERTAG; \ BEGIN(NORMAL); \ return USERTAG; \ } @@ -209,16 +227,26 @@ int test_loop(ENCODING enc, char* code) #define ACTION_ANY \ - { CHECK_LINE_LEN; \ - gedcom_lval.string = TO_INTERNAL(yytext, str_buf); \ - /* Due to character conversions, it is possible that the current \ - character will be combined with the next, and so now we don't have a \ - character yet... \ - In principle, this is only applicable to the 1byte case (e.g. ANSEL), \ - but it doesn't harm the unicode case. \ - */ \ - if (strlen(gedcom_lval.string) > 0) \ - return ANYCHAR; \ + { char* tmp; \ + CHECK_LINE_LEN; \ + tmp = TO_INTERNAL(yytext, str_buf); \ + if (!tmp) { \ + /* Something went wrong during conversion... */ \ + gedcom_error(_("Invalid character for encoding: '%s' (0x%02x)"), \ + yytext, yytext[0]); \ + return BADTOKEN; \ + } \ + else { \ + gedcom_lval.string = tmp; \ + /* Due to character conversions, it is possible that the current \ + character will be combined with the next, and so now we don't have a \ + character yet... \ + In principle, this is only applicable to the 1byte case (e.g. ANSEL),\ + but it doesn't harm the unicode case. \ + */ \ + if (strlen(gedcom_lval.string) > 0) \ + return ANYCHAR; \ + } \ } @@ -232,7 +260,7 @@ int test_loop(ENCODING enc, char* code) #define ACTION_POINTER \ { CHECK_LINE_LEN; \ if (strlen(yytext) > MAXGEDCPTRLEN * encoding_width) { \ - gedcom_error("Pointer '%s' too long, max %d characters", \ + gedcom_error(_("Pointer '%s' too long, max %d characters allowed"), \ yytext, MAXGEDCPTRLEN); \ return BADTOKEN; \ } \ @@ -266,17 +294,40 @@ int test_loop(ENCODING enc, char* code) return CLOSE; \ } \ else { \ + char* ptr; int size; \ /* Reset our state */ \ current_level = -1; \ level_diff = MAXGEDCLEVEL; \ /* ... then terminate lex */ \ yyterminate(); \ + /* Get rid of f*cking compiler warning from lex generated code */ \ + /* yyterminate does return(), so program will never come here */ \ + yy_flex_realloc(ptr, size); \ } \ } +#define ACTION_NORMAL_AT \ + { if (compat_at) { \ + int i, j; \ + char *yycopy = strdup(yytext); \ + if (yycopy) { \ + for (i = 0; i < 2; i++) \ + for (j = yyleng - 1; j >= 0; --j) \ + unput(yycopy[j]); \ + free(yycopy); \ + } \ + else { \ + MEMORY_ERROR; \ + } \ + } \ + else { \ + gedcom_error(_("'@' character should be written as '@@' in values")); \ + return BADTOKEN; \ + } \ + } #define ACTION_UNEXPECTED \ - { gedcom_error("Unexpected character: '%s' (0x%02x)", \ + { gedcom_error(_("Unexpected character: '%s' (0x%02x)"), \ yytext, yytext[0]); \ return BADTOKEN; \ }