+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*
-* Copyright (C) 2009-2010, International Business Machines
+* Copyright (C) 2009-2014, International Business Machines
* Corporation and others. All Rights Reserved.
*
*******************************************************************************
* file name: gennorm2.cpp
-* encoding: US-ASCII
+* encoding: UTF-8
* tab size: 8 (not used)
* indentation:4
*
#include "unicode/utypes.h"
#include "n2builder.h"
+#include <fstream>
#include <stdio.h>
#include <stdlib.h>
+#include <string>
#include <string.h>
#include "unicode/errorcode.h"
#include "unicode/localpointer.h"
#include "unewdata.h"
#endif
-#define LENGTHOF(array) (int32_t)(sizeof(array)/sizeof((array)[0]))
-
U_NAMESPACE_BEGIN
UBool beVerbose=FALSE, haveCopyright=TRUE;
-U_DEFINE_LOCAL_OPEN_POINTER(LocalStdioFilePointer, FILE, fclose);
-
#if !UCONFIG_NO_NORMALIZATION
-void parseFile(FILE *f, Normalizer2DataBuilder &builder);
+void parseFile(std::ifstream &f, Normalizer2DataBuilder &builder);
#endif
/* -------------------------------------------------------------------------- */
SOURCEDIR,
OUTPUT_FILENAME,
UNICODE_VERSION,
+ WRITE_C_SOURCE,
+ WRITE_COMBINED_DATA,
OPT_FAST
};
UOPTION_SOURCEDIR,
UOPTION_DEF("output", 'o', UOPT_REQUIRES_ARG),
UOPTION_DEF("unicode", 'u', UOPT_REQUIRES_ARG),
+ UOPTION_DEF("csource", '\1', UOPT_NO_ARG),
+ UOPTION_DEF("combined", '\1', UOPT_NO_ARG),
UOPTION_DEF("fast", '\1', UOPT_NO_ARG)
};
/* preset then read command line options */
options[SOURCEDIR].value="";
- options[UNICODE_VERSION].value=U_UNICODE_VERSION;
argc=u_parseArgs(argc, argv, sizeof(options)/sizeof(options[HELP_H]), options);
/* error handling, printing usage message */
if( argc<2 ||
options[HELP_H].doesOccur || options[HELP_QUESTION_MARK].doesOccur
) {
- /*
- * Broken into chunks because the C89 standard says the minimum
- * required supported string length is 509 bytes.
- */
fprintf(stderr,
"Usage: %s [-options] infiles+ -o outputfilename\n"
"\n"
"Reads the infiles with normalization data and\n"
- "creates a binary file (outputfilename) with the data.\n"
+ "creates a binary file, or a C source file (--csource), with the data,\n"
+ "or writes a data file with the combined data (--combined).\n"
+ "See http://userguide.icu-project.org/transforms/normalization#TOC-Data-File-Syntax\n"
+ "\n"
+ "Alternate usage: %s [-options] a.txt b.txt minus p.txt q.txt -o outputfilename\n"
+ "\n"
+ "Computes the difference of (a, b) minus (p, q) and writes the diff data\n"
+ "in input-file syntax to the outputfilename.\n"
+ "It is then possible to build (p, q, diff) to get the same data as (a, b).\n"
+ "(Useful for computing minimal incremental mapping data files.)\n"
"\n",
- argv[0]);
+ argv[0], argv[0]);
fprintf(stderr,
"Options:\n"
"\t-h or -? or --help this usage text\n"
"\t-u or --unicode Unicode version, followed by the version like 5.2.0\n");
fprintf(stderr,
"\t-s or --sourcedir source directory, followed by the path\n"
- "\t-o or --output output filename\n");
+ "\t-o or --output output filename\n"
+ "\t --csource writes a C source file with initializers\n"
+ "\t --combined writes a .txt file (input-file syntax) with the\n"
+ "\t combined data from all of the input files\n");
fprintf(stderr,
- "\t --fast optimize the .nrm file for fast normalization,\n"
+ "\t --fast optimize the data for fast normalization,\n"
"\t which might increase its size (Writes fully decomposed\n"
"\t regular mappings instead of delta mappings.\n"
"\t You should measure the runtime speed to make sure that\n"
#else
- LocalPointer<Normalizer2DataBuilder> builder(new Normalizer2DataBuilder(errorCode));
+ LocalPointer<Normalizer2DataBuilder> b1(new Normalizer2DataBuilder(errorCode), errorCode);
+ LocalPointer<Normalizer2DataBuilder> b2;
+ LocalPointer<Normalizer2DataBuilder> diff;
+ Normalizer2DataBuilder *builder = b1.getAlias();
errorCode.assertSuccess();
- builder->setUnicodeVersion(options[UNICODE_VERSION].value);
+ if(options[UNICODE_VERSION].doesOccur) {
+ builder->setUnicodeVersion(options[UNICODE_VERSION].value);
+ }
if(options[OPT_FAST].doesOccur) {
builder->setOptimization(Normalizer2DataBuilder::OPTIMIZE_FAST);
pathLength=filename.length();
}
+ bool doMinus = false;
for(int i=1; i<argc; ++i) {
printf("gennorm2: processing %s\n", argv[i]);
+ if(strcmp(argv[i], "minus") == 0) {
+ if(doMinus) {
+ fprintf(stderr, "gennorm2 error: only one 'minus' can be specified\n");
+ exit(U_ILLEGAL_ARGUMENT_ERROR);
+ }
+ // Data from previous input files has been collected in b1.
+ // Collect data from further input files in b2.
+ b2.adoptInsteadAndCheckErrorCode(new Normalizer2DataBuilder(errorCode), errorCode);
+ diff.adoptInsteadAndCheckErrorCode(new Normalizer2DataBuilder(errorCode), errorCode);
+ errorCode.assertSuccess();
+ builder = b2.getAlias();
+ if(options[UNICODE_VERSION].doesOccur) {
+ builder->setUnicodeVersion(options[UNICODE_VERSION].value);
+ }
+ if(options[OPT_FAST].doesOccur) {
+ builder->setOptimization(Normalizer2DataBuilder::OPTIMIZE_FAST);
+ }
+ doMinus = true;
+ continue;
+ }
filename.append(argv[i], errorCode);
- LocalStdioFilePointer f(fopen(filename.data(), "r"));
- if(f==NULL) {
+ std::ifstream f(filename.data());
+ if(f.fail()) {
fprintf(stderr, "gennorm2 error: unable to open %s\n", filename.data());
exit(U_FILE_ACCESS_ERROR);
}
builder->setOverrideHandling(Normalizer2DataBuilder::OVERRIDE_PREVIOUS);
- parseFile(f.getAlias(), *builder);
+ parseFile(f, *builder);
filename.truncate(pathLength);
}
- builder->writeBinaryFile(options[OUTPUT_FILENAME].value);
+ if(doMinus) {
+ Normalizer2DataBuilder::computeDiff(*b1, *b2, *diff);
+ diff->writeDataFile(options[OUTPUT_FILENAME].value, /* writeRemoved= */ true);
+ } else if(options[WRITE_COMBINED_DATA].doesOccur) {
+ builder->writeDataFile(options[OUTPUT_FILENAME].value, /* writeRemoved= */ false);
+ } else if(options[WRITE_C_SOURCE].doesOccur) {
+ builder->writeCSourceFile(options[OUTPUT_FILENAME].value);
+ } else {
+ builder->writeBinaryFile(options[OUTPUT_FILENAME].value);
+ }
return errorCode.get();
#if !UCONFIG_NO_NORMALIZATION
-void parseFile(FILE *f, Normalizer2DataBuilder &builder) {
+void parseFile(std::ifstream &f, Normalizer2DataBuilder &builder) {
IcuToolErrorCode errorCode("gennorm2/parseFile()");
- char line[300];
+ std::string lineString;
uint32_t startCP, endCP;
- while(NULL!=fgets(line, (int)sizeof(line), f)) {
+ while(std::getline(f, lineString)) {
+ if (lineString.empty()) {
+ continue; // skip empty lines.
+ }
+#if (U_CPLUSPLUS_VERSION >= 11)
+ char *line = &lineString.front();
+#else
+ char *line = &lineString.at(0);
+#endif
char *comment=(char *)strchr(line, '#');
if(comment!=NULL) {
*comment=0;
continue; // skip empty and comment-only lines
}
if(line[0]=='*') {
+ const char *s=u_skipWhitespace(line+1);
+ if(0==strncmp(s, "Unicode", 7)) {
+ s=u_skipWhitespace(s+7);
+ builder.setUnicodeVersion(s);
+ }
continue; // reserved syntax
}
const char *delimiter;
fprintf(stderr, "gennorm2 error: parsing code point range from %s\n", line);
exit(errorCode.reset());
}
+ if (endCP >= 0xd800 && startCP <= 0xdfff) {
+ fprintf(stderr, "gennorm2 error: value or mapping for surrogate code points: %s\n",
+ line);
+ exit(U_ILLEGAL_ARGUMENT_ERROR);
+ }
delimiter=u_skipWhitespace(delimiter);
if(*delimiter==':') {
const char *s=u_skipWhitespace(delimiter+1);
}
if(*delimiter=='=' || *delimiter=='>') {
UChar uchars[Normalizer2Impl::MAPPING_LENGTH_MASK];
- int32_t length=u_parseString(delimiter+1, uchars, LENGTHOF(uchars), NULL, errorCode);
+ int32_t length=u_parseString(delimiter+1, uchars, UPRV_LENGTHOF(uchars), NULL, errorCode);
if(errorCode.isFailure()) {
fprintf(stderr, "gennorm2 error: parsing mapping string from %s\n", line);
exit(errorCode.reset());
projects / apple / icu.git / blobdiff