/*
**********************************************************************
-* Copyright (C) 2002-2003, International Business Machines
+* Copyright (C) 2002-2016, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
*
//
//--------------------------------------------------------------------
-#include <stdio.h>
#include "unicode/utypes.h"
#include "unicode/ucnv.h"
#include "unicode/unistr.h"
#include "unicode/rbbi.h"
#include "unicode/uclean.h"
#include "unicode/udata.h"
+#include "unicode/putil.h"
#include "uoptions.h"
#include "unewdata.h"
#include "ucmndata.h"
+#include "rbbidata.h"
+#include "cmemory.h"
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
-#define DATA_TYPE "brk"
+U_NAMESPACE_USE
static char *progName;
static UOption options[]={
{ "rules", NULL, NULL, NULL, 'r', UOPT_REQUIRES_ARG, 0 }, /* 3 */
{ "out", NULL, NULL, NULL, 'o', UOPT_REQUIRES_ARG, 0 }, /* 4 */
UOPTION_ICUDATADIR, /* 5 */
- UOPTION_DESTDIR /* 6 */
+ UOPTION_DESTDIR, /* 6 */
+ UOPTION_COPYRIGHT, /* 7 */
+ UOPTION_QUIET, /* 8 */
};
void usageAndDie(int retCode) {
- printf("Usage: %s [-v] -r rule-file -o output-file\n", progName);
+ printf("Usage: %s [-v] [-options] -r rule-file -o output-file\n", progName);
+ printf("\tRead in break iteration rules text and write out the binary data\n"
+ "options:\n"
+ "\t-h or -? or --help this usage text\n"
+ "\t-V or --version show a version message\n"
+ "\t-c or --copyright include a copyright notice\n"
+ "\t-v or --verbose turn on verbose output\n"
+ "\t-q or --quiet do not display warnings and progress\n"
+ "\t-i or --icudatadir directory for locating any needed intermediate data files,\n"
+ "\t followed by path, defaults to %s\n"
+ "\t-d or --destdir destination directory, followed by the path\n",
+ u_getDataDirectory());
exit (retCode);
}
-#if UCONFIG_NO_BREAK_ITERATION
+#if UCONFIG_NO_BREAK_ITERATION || UCONFIG_NO_FILE_IO
/* dummy UDataInfo cf. udata.h */
static UDataInfo dummyDataInfo = {
0, // reserved
{ 0x42, 0x72, 0x6b, 0x20 }, // dataFormat="Brk "
- { 2, 1, 0, 0 }, // formatVersion
- { 3, 1, 0, 0 } // dataVersion (Unicode version)
+ { 0xff, 0, 0, 0 }, // formatVersion. Filled in later with values
+ // from the RBBI rule builder. The values declared
+ // here should never appear in any real RBBI data.
+ { 4, 1, 0, 0 } // dataVersion (Unicode version)
}};
#endif
const char *ruleFileName;
const char *outFileName;
const char *outDir = NULL;
- char *outFullFileName;
- int32_t outFullFileNameLen;
+ const char *copyright = NULL;
//
// Pick up and check the command line arguments,
//
U_MAIN_INIT_ARGS(argc, argv);
progName = argv[0];
- argc=u_parseArgs(argc, argv, sizeof(options)/sizeof(options[0]), options);
+ argc=u_parseArgs(argc, argv, UPRV_LENGTHOF(options), options);
if(argc<0) {
// Unrecognized option
fprintf(stderr, "error in command line argument \"%s\"\n", argv[-argc]);
}
ruleFileName = options[3].value;
outFileName = options[4].value;
- outFullFileNameLen = strlen(outFileName);
if (options[5].doesOccur) {
u_setDataDirectory(options[5].value);
}
+ status = U_ZERO_ERROR;
+
/* Combine the directory with the file name */
if(options[6].doesOccur) {
outDir = options[6].value;
- outFullFileNameLen += strlen(outDir);
}
- outFullFileName = (char*)malloc(outFullFileNameLen + 2);
- outFullFileName[0] = 0;
- if (outDir) {
- strcpy(outFullFileName, outDir);
- strcat(outFullFileName, U_FILE_SEP_STRING);
+ if (options[7].doesOccur) {
+ copyright = U_COPYRIGHT_STRING;
}
- strcat(outFullFileName, outFileName);
-#if UCONFIG_NO_BREAK_ITERATION
+#if UCONFIG_NO_BREAK_ITERATION || UCONFIG_NO_FILE_IO
UNewDataMemory *pData;
- char msg[2048], folder[2048], name[32];
- char *basename;
- int length;
-
- /* split the outFileName into folder + name + type */
- strcpy(folder, outFileName);
- basename = strrchr(folder, U_FILE_SEP_CHAR);
- if(basename == NULL) {
- basename = folder;
- } else {
- ++basename;
- }
-
- /* copy the data name and remove it from the folder */
- strcpy(name, basename);
- *basename = 0;
+ char msg[1024];
/* write message with just the name */
- sprintf(msg, "genbrk writes dummy %s because of UCONFIG_NO_BREAK_ITERATION, see uconfig.h", name);
+ sprintf(msg, "genbrk writes dummy %s because of UCONFIG_NO_BREAK_ITERATION and/or UCONFIG_NO_FILE_IO, see uconfig.h", outFileName);
fprintf(stderr, "%s\n", msg);
- /* remove the type suffix (hardcode to DATA_TYPE) */
- length = strlen(name);
- if(length > 4 && name[length - 4] == '.') {
- name[length - 4] = 0;
- }
-
/* write the dummy data file */
- pData = udata_create(folder, DATA_TYPE, name, &dummyDataInfo, NULL, &status);
+ pData = udata_create(outDir, NULL, outFileName, &dummyDataInfo, NULL, &status);
udata_writeBlock(pData, msg, strlen(msg));
udata_finish(pData, &status);
return (int)status;
#else
+ /* Initialize ICU */
+ u_init(&status);
+ if (U_FAILURE(status)) {
+ fprintf(stderr, "%s: can not initialize ICU. status = %s\n",
+ argv[0], u_errorName(status));
+ exit(1);
+ }
+ status = U_ZERO_ERROR;
//
// Read in the rule source file
//
- int result;
+ long result;
long ruleFileSize;
FILE *file;
char *ruleBufferC;
fseek(file, 0, SEEK_SET);
ruleBufferC = new char[ruleFileSize+10];
- result = fread(ruleBufferC, 1, ruleFileSize, file);
+ result = (long)fread(ruleBufferC, 1, ruleFileSize, file);
if (result != ruleFileSize) {
fprintf(stderr, "Error reading file \"%s\"\n", ruleFileName);
exit (-1);
// This will compile the rules.
//
UParseError parseError;
- parseError.line = 0;
- parseError.offset = 0;
+ parseError.line = 0;
+ parseError.offset = 0;
RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(ruleSourceS, parseError, status);
if (U_FAILURE(status)) {
fprintf(stderr, "createRuleBasedBreakIterator: ICU Error \"%s\" at line %d, column %d\n",
- u_errorName(status), parseError.line, parseError.offset);
+ u_errorName(status), (int)parseError.line, (int)parseError.offset);
exit(status);
};
const uint8_t *outData;
outData = bi->getBinaryRules(outDataSize);
+ // Copy the data format version numbers from the RBBI data header into the UDataMemory header.
+ uprv_memcpy(dh.info.formatVersion, ((RBBIDataHeader *)outData)->fFormatVersion, sizeof(dh.info.formatVersion));
//
// Create the output file
//
size_t bytesWritten;
- file = fopen(outFullFileName, "wb");
- if (file == 0) {
- fprintf(stderr, "Could not open output file \"%s\"\n", outFullFileName);
- exit(-1);
+ UNewDataMemory *pData;
+ pData = udata_create(outDir, NULL, outFileName, &(dh.info), copyright, &status);
+ if(U_FAILURE(status)) {
+ fprintf(stderr, "genbrk: Could not open output file \"%s\", \"%s\"\n",
+ outFileName, u_errorName(status));
+ exit(status);
}
- bytesWritten = fwrite(&dh, 1, sizeof(DataHeader), file);
- //
// Write the data itself.
- //
- bytesWritten = fwrite(outData, 1, outDataSize, file);
+ udata_writeBlock(pData, outData, outDataSize);
+ // finish up
+ bytesWritten = udata_finish(pData, &status);
+ if(U_FAILURE(status)) {
+ fprintf(stderr, "genbrk: error %d writing the output file\n", status);
+ exit(status);
+ }
+
if (bytesWritten != outDataSize) {
- fprintf(stderr, "Error writing to output file \"%s\"\n", outFullFileName);
+ fprintf(stderr, "Error writing to output file \"%s\"\n", outFileName);
exit(-1);
}
- fclose(file);
delete bi;
delete[] ruleSourceU;
delete[] ruleBufferC;
- free(outFullFileName);
u_cleanup();
- printf("genbrk: tool completed successfully.\n");
+ if(!options[8].doesOccur) {
+ printf("genbrk: tool completed successfully.\n");
+ }
return 0;
#endif /* #if !UCONFIG_NO_BREAK_ITERATION */
}
+
projects / apple / icu.git / blobdiff