X-Git-Url: https://git.dlugolecki.net.pl/?a=blobdiff_plain;f=gedcom%2Fmultilex.c;h=393516adc8fdde69a30ad81e005b708cd1dedc33;hb=60282b9f2f1326231ea5019e239d00bdccd6608b;hp=420bfcb8b1c88b7a684328313f85055601df6af2;hpb=32da62601457ba994c6b71d71470ae066fc3969b;p=gedcom-parse.git diff --git a/gedcom/multilex.c b/gedcom/multilex.c index 420bfcb..393516a 100644 --- a/gedcom/multilex.c +++ b/gedcom/multilex.c @@ -24,6 +24,7 @@ #include "gedcom_internal.h" #include "multilex.h" #include "encoding.h" +#include "encoding_state.h" #include "xref.h" int line_no = 0; @@ -38,19 +39,19 @@ int lexer_init(Encoding enc, FILE* f) if (enc == ONE_BYTE) { lf = &gedcom_1byte_lex; gedcom_1byte_myinit(f); - set_encoding_width(enc); + set_read_encoding_width(enc); return open_conv_to_internal("ASCII"); } else if (enc == TWO_BYTE_HILO) { lf = &gedcom_hilo_lex; gedcom_hilo_myinit(f); - set_encoding_width(enc); + set_read_encoding_width(enc); return open_conv_to_internal("UNICODE"); } else if (enc == TWO_BYTE_LOHI) { lf = &gedcom_lohi_lex; gedcom_lohi_myinit(f); - set_encoding_width(enc); + set_read_encoding_width(enc); return open_conv_to_internal("UNICODE"); } else { @@ -79,6 +80,7 @@ int determine_encoding(FILE* f) char first[2]; int read; + set_read_encoding_bom(WITHOUT_BOM); read = fread(first, 1, 2, f); if (read != 2) { gedcom_warning(_("Error reading from input file: %s"), strerror(errno)); @@ -86,26 +88,28 @@ int determine_encoding(FILE* f) return ONE_BYTE; } else if ((first[0] == '0') && (first[1] == ' ')) { - gedcom_debug_print(_("One-byte encoding")); + gedcom_debug_print("One-byte encoding"); rewind_file(f); return ONE_BYTE; } else if ((first[0] == '\0') && (first[1] == '0')) { - gedcom_debug_print(_("Two-byte encoding, high-low")); + gedcom_debug_print("Two-byte encoding, high-low"); rewind_file(f); return TWO_BYTE_HILO; } else if ((first[0] == '\xFE') && (first[1] == '\xFF')) { - gedcom_debug_print(_("Two-byte encoding, high-low, with BOM")); + gedcom_debug_print("Two-byte encoding, high-low, with BOM"); + set_read_encoding_bom(WITH_BOM); return TWO_BYTE_HILO; } else if ((first[0] == '0') && (first[1] == '\0')) { - gedcom_debug_print(_("Two-byte encoding, low-high")); + gedcom_debug_print("Two-byte encoding, low-high"); rewind_file(f); return TWO_BYTE_LOHI; } else if ((first[0] == '\xFF') && (first[1] == '\xFE')) { - gedcom_debug_print(_("Two-byte encoding, low-high, with BOM")); + gedcom_debug_print("Two-byte encoding, low-high, with BOM"); + set_read_encoding_bom(WITH_BOM); return TWO_BYTE_LOHI; } else if ((first[0] == '\xEF') && (first[1] == '\xBB')) { @@ -115,7 +119,12 @@ int determine_encoding(FILE* f) rewind_file(f); } else if (first[0] == '\xBF') { - gedcom_debug_print(_("UTF-8 encoding, with BOM")); + set_read_encoding_bom(WITH_BOM); + gedcom_debug_print("UTF-8 encoding, with BOM"); + } + else { + gedcom_warning(_("Unknown encoding, falling back to one-byte")); + rewind_file(f); } return ONE_BYTE; } @@ -128,6 +137,25 @@ int determine_encoding(FILE* f) int init_called = 0; +/** This function initializes the Gedcom parser library and must be called + before any other function in this library. + + The function also initializes locale handling by calling + setlocale(LC_ALL, "") , in case the application would not do this + (it doesn't hurt for the application to do the same). + + \attention This function should be called as early as possible. The + requirement + is that it should come before the first call to \c iconv_open (part of the + generic character set conversion feature) in the program, either by your + program itself, or indirectly by the library calls it makes. + \attention Practically, + it should e.g. come before any calls to any GTK functions, because GTK + uses \c iconv_open in its initialization. + + \retval 0 in case of success + \retval nonzero in case of failure (e.g. failure to set locale) + */ int gedcom_init() { init_called = 1; @@ -141,6 +169,28 @@ int gedcom_init() return 0; } +/** This function parses the given file. By itself, it doesn't provide any + other information than the parse result. + + The function also empties the cross-reference table before parsing, and + checks the validity of the + cross-references if the parse was successful. + The following conditions can occur in the cross-reference table: + - An xref was defined, but not used (warning) + - An xref was used, but not defined (error) + - An xref was used as a different type than the defined type (error) + + \param file_name The name of the Gedcom file to parse + + \retval 0 if the parse was successful and no errors were found in the + cross-reference table + \retval nonzero on errors, which can include: + - \ref gedcom_init() was not called + - The given file was not found + - The parse of the given file failed + - There were errors found in the cross-reference table + */ + int gedcom_parse_file(const char* file_name) { Encoding enc; @@ -176,6 +226,15 @@ int gedcom_parse_file(const char* file_name) return result; } +/** This function starts a new model. It does this by parsing the \c new.ged + file in the data directory of the library (\c $PREFIX/share/gedcom-parse). + This can be used to start from an empty model, and to build up the model + by adding new records yourself. + + \retval 0 on success + \retval nonzero on errors (mainly the errors from \ref gedcom_parse_file()). + */ + int gedcom_new_model() { int result = 1;