ICU-531.48.tar.gz

[apple/icu.git] / icuSources / tools / gensprep / gensprep.c

diff --git a/icuSources/tools/gensprep/gensprep.c b/icuSources/tools/gensprep/gensprep.c
index e8f9608ffc88fc7b7b8f8cba0d46063f2624aedb..94ba08d89b9394637584e360bdb0260d76c00e1e 100644 (file)
--- a/icuSources/tools/gensprep/gensprep.c
+++ b/icuSources/tools/gensprep/gensprep.c
@@ -1,7 +1,7 @@
 /*
 *******************************************************************************
 *
 /*
 *******************************************************************************
 *

-*   Copyright (C) 2003-2004, International Business Machines
+*   Copyright (C) 2003-2012, International Business Machines

 *   Corporation and others.  All Rights Reserved.
 *
 *******************************************************************************
 *   Corporation and others.  All Rights Reserved.
 *
 *******************************************************************************


@@ -31,6 +31,7 @@
 #include "uparse.h"
 #include "sprpimpl.h"
 
 #include "uparse.h"
 #include "sprpimpl.h"
 

+#include "unicode/uclean.h"

 #include "unicode/udata.h"
 #include "unicode/utypes.h"
 #include "unicode/putil.h"
 #include "unicode/udata.h"
 #include "unicode/utypes.h"
 #include "unicode/putil.h"


@@ -40,14 +41,15 @@ U_CDECL_BEGIN
 #include "gensprep.h"
 U_CDECL_END
 
 #include "gensprep.h"
 U_CDECL_END
 

-#ifdef WIN32
-#   pragma warning(disable: 4100)
-#endif
-

 UBool beVerbose=FALSE, haveCopyright=TRUE;
 
 #define NORM_CORRECTIONS_FILE_NAME "NormalizationCorrections.txt"
 
 UBool beVerbose=FALSE, haveCopyright=TRUE;
 
 #define NORM_CORRECTIONS_FILE_NAME "NormalizationCorrections.txt"
 

+#define NORMALIZE_DIRECTIVE "normalize"
+#define NORMALIZE_DIRECTIVE_LEN 9
+#define CHECK_BIDI_DIRECTIVE "check-bidi"
+#define CHECK_BIDI_DIRECTIVE_LEN 10
+

 /* prototypes --------------------------------------------------------------- */
 
 static void
 /* prototypes --------------------------------------------------------------- */
 
 static void


@@ -67,9 +69,9 @@ static UOption options[]={
     UOPTION_DESTDIR,
     UOPTION_SOURCEDIR,
     UOPTION_ICUDATADIR,
     UOPTION_DESTDIR,
     UOPTION_SOURCEDIR,
     UOPTION_ICUDATADIR,

-    UOPTION_PACKAGE_NAME,

     UOPTION_BUNDLE_NAME,
     { "normalization", NULL, NULL, NULL, 'n', UOPT_REQUIRES_ARG, 0 },
     UOPTION_BUNDLE_NAME,
     { "normalization", NULL, NULL, NULL, 'n', UOPT_REQUIRES_ARG, 0 },

+    { "norm-correction", NULL, NULL, NULL, 'm', UOPT_REQUIRES_ARG, 0 },

     { "check-bidi", NULL, NULL, NULL,  'k', UOPT_NO_ARG, 0},
     { "unicode", NULL, NULL, NULL, 'u', UOPT_REQUIRES_ARG, 0 },
 };
     { "check-bidi", NULL, NULL, NULL,  'k', UOPT_NO_ARG, 0},
     { "unicode", NULL, NULL, NULL, 'u', UOPT_REQUIRES_ARG, 0 },
 };


@@ -82,9 +84,9 @@ enum{
     DESTDIR,
     SOURCEDIR,
     ICUDATADIR,
     DESTDIR,
     SOURCEDIR,
     ICUDATADIR,

-    PACKAGE_NAME,

     BUNDLE_NAME,
     NORMALIZE,
     BUNDLE_NAME,
     NORMALIZE,

+    NORM_CORRECTION_DIR,

     CHECK_BIDI,
     UNICODE_VERSION
 };
     CHECK_BIDI,
     UNICODE_VERSION
 };


@@ -114,10 +116,15 @@ static int printHelp(int argc, char* argv[]){
         "\t                         followed by path, defaults to %s\n",
         u_getDataDirectory());
     fprintf(stderr,
         "\t                         followed by path, defaults to %s\n",
         u_getDataDirectory());
     fprintf(stderr,

-        "\t-p or --package-name     prepend the output data file name with the package name specified\n"

         "\t-n or --normalize        turn on the option for normalization and include mappings\n"
         "\t                         from NormalizationCorrections.txt from the given path,\n"
         "\t-n or --normalize        turn on the option for normalization and include mappings\n"
         "\t                         from NormalizationCorrections.txt from the given path,\n"

-        "\t                         e.g: /test/icu/source/data/unidata\n"
+        "\t                         e.g: /test/icu/source/data/unidata\n");
+    fprintf(stderr,
+        "\t-m or --norm-correction  use NormalizationCorrections.txt from the given path\n"
+        "\t                         when the input file contains a normalization directive.\n"
+        "\t                         unlike -n/--normalize, this option does not force the\n"
+        "\t                         normalization.\n");
+    fprintf(stderr,

         "\t-k or --check-bidi       turn on the option for checking for BiDi in the profile\n"
         "\t-u or --unicode          version of Unicode to be used with this profile followed by the version\n"
         );
         "\t-k or --check-bidi       turn on the option for checking for BiDi in the profile\n"
         "\t-u or --unicode          version of Unicode to be used with this profile followed by the version\n"
         );


@@ -131,7 +138,7 @@ main(int argc, char* argv[]) {
     char* filename = NULL;
 #endif
     const char *srcDir=NULL, *destDir=NULL, *icuUniDataDir=NULL;
     char* filename = NULL;
 #endif
     const char *srcDir=NULL, *destDir=NULL, *icuUniDataDir=NULL;

-    const char *packageName=NULL, *bundleName=NULL, *inputFileName = NULL;
+    const char *bundleName=NULL, *inputFileName = NULL;

     char *basename=NULL;
     int32_t sprepOptions = 0;
 
     char *basename=NULL;
     int32_t sprepOptions = 0;
 


@@ -144,7 +151,6 @@ main(int argc, char* argv[]) {
     options[SOURCEDIR].value="";
     options[UNICODE_VERSION].value="0"; /* don't assume the unicode version */
     options[BUNDLE_NAME].value = DATA_NAME;
     options[SOURCEDIR].value="";
     options[UNICODE_VERSION].value="0"; /* don't assume the unicode version */
     options[BUNDLE_NAME].value = DATA_NAME;

-    options[PACKAGE_NAME].value = NULL;

     options[NORMALIZE].value = "";
 
     argc=u_parseArgs(argc, argv, sizeof(options)/sizeof(options[0]), options);
     options[NORMALIZE].value = "";
 
     argc=u_parseArgs(argc, argv, sizeof(options)/sizeof(options[0]), options);


@@ -165,9 +171,12 @@ main(int argc, char* argv[]) {
     haveCopyright=options[COPYRIGHT].doesOccur;
     srcDir=options[SOURCEDIR].value;
     destDir=options[DESTDIR].value;
     haveCopyright=options[COPYRIGHT].doesOccur;
     srcDir=options[SOURCEDIR].value;
     destDir=options[DESTDIR].value;

-    packageName = options[PACKAGE_NAME].value;

     bundleName = options[BUNDLE_NAME].value;
     bundleName = options[BUNDLE_NAME].value;

-    icuUniDataDir = options[NORMALIZE].value;
+    if(options[NORMALIZE].doesOccur) {
+        icuUniDataDir = options[NORMALIZE].value;
+    } else {
+        icuUniDataDir = options[NORM_CORRECTION_DIR].value;
+    }

 
     if(argc<2) {
         /* print the help message */
 
     if(argc<2) {
         /* print the help message */


@@ -187,7 +196,7 @@ main(int argc, char* argv[]) {
         "gensprep writes dummy " U_ICUDATA_NAME "_" DATA_NAME "." DATA_TYPE
         " because UCONFIG_NO_IDNA is set, \n"
         "see icu/source/common/unicode/uconfig.h\n");
         "gensprep writes dummy " U_ICUDATA_NAME "_" DATA_NAME "." DATA_TYPE
         " because UCONFIG_NO_IDNA is set, \n"
         "see icu/source/common/unicode/uconfig.h\n");

-    generateData(destDir, packageName, bundleName);
+    generateData(destDir, bundleName);

 
 #else
 
 
 #else
 


@@ -219,7 +228,7 @@ main(int argc, char* argv[]) {
         return errorCode;
     }
     
         return errorCode;
     }
     

-    if(options[NORMALIZE].doesOccur){
+    if(options[NORMALIZE].doesOccur){ /* this option might be set by @normalize;; in the source file */

         /* set up directory for NormalizationCorrections.txt */
         uprv_strcpy(filename,icuUniDataDir);
         basename=filename+uprv_strlen(filename);
         /* set up directory for NormalizationCorrections.txt */
         uprv_strcpy(filename,icuUniDataDir);
         basename=filename+uprv_strlen(filename);


@@ -238,7 +247,7 @@ main(int argc, char* argv[]) {
         sprepOptions |= _SPREP_NORMALIZATION_ON;
     }
     
         sprepOptions |= _SPREP_NORMALIZATION_ON;
     }
     

-    if(options[CHECK_BIDI].doesOccur){
+    if(options[CHECK_BIDI].doesOccur){ /* this option might be set by @check-bidi;; in the source file */

         sprepOptions |= _SPREP_CHECK_BIDI_ON;
     }
 
         sprepOptions |= _SPREP_CHECK_BIDI_ON;
     }
 


@@ -247,13 +256,15 @@ main(int argc, char* argv[]) {
     /* process parsed data */
     if(U_SUCCESS(errorCode)) {
         /* write the data file */
     /* process parsed data */
     if(U_SUCCESS(errorCode)) {
         /* write the data file */

-       generateData(destDir, packageName, bundleName);
+        generateData(destDir, bundleName);

 
 

-       cleanUpData();
+        cleanUpData();

     }
 
     uprv_free(filename);
 
     }
 
     uprv_free(filename);
 

+    u_cleanup();
+

 #endif
 
     return errorCode;
 #endif
 
     return errorCode;


@@ -336,13 +347,34 @@ strprepProfileLineFn(void *context,
     const char* typeName;
     uint32_t rangeStart=0,rangeEnd =0;
     const char* filename = (const char*) context;
     const char* typeName;
     uint32_t rangeStart=0,rangeEnd =0;
     const char* filename = (const char*) context;

- 
+    const char *s;
+
+    s = u_skipWhitespace(fields[0][0]);
+    if (*s == '@') {
+        /* special directive */
+        s++;
+        length = fields[0][1] - s;
+        if (length >= NORMALIZE_DIRECTIVE_LEN
+            && uprv_strncmp(s, NORMALIZE_DIRECTIVE, NORMALIZE_DIRECTIVE_LEN) == 0) {
+            options[NORMALIZE].doesOccur = TRUE;
+            return;
+        }
+        else if (length >= CHECK_BIDI_DIRECTIVE_LEN
+            && uprv_strncmp(s, CHECK_BIDI_DIRECTIVE, CHECK_BIDI_DIRECTIVE_LEN) == 0) {
+            options[CHECK_BIDI].doesOccur = TRUE;
+            return;
+        }
+        else {
+            fprintf(stderr, "gensprep error parsing a directive %s.", fields[0][0]);
+        }
+    }
+

     typeName = fields[2][0];
     map = fields[1][0];
 
     if(uprv_strstr(typeName, usprepTypeNames[USPREP_UNASSIGNED])!=NULL){
 
     typeName = fields[2][0];
     map = fields[1][0];
 
     if(uprv_strstr(typeName, usprepTypeNames[USPREP_UNASSIGNED])!=NULL){
 

-        u_parseCodePointRange(fields[0][0], &rangeStart,&rangeEnd, pErrorCode);
+        u_parseCodePointRange(s, &rangeStart,&rangeEnd, pErrorCode);

         if(U_FAILURE(*pErrorCode)){
             fprintf(stderr, "Could not parse code point range. Error: %s\n",u_errorName(*pErrorCode));
             return;
         if(U_FAILURE(*pErrorCode)){
             fprintf(stderr, "Could not parse code point range. Error: %s\n",u_errorName(*pErrorCode));
             return;


@@ -353,7 +385,7 @@ strprepProfileLineFn(void *context,
 
     }else if(uprv_strstr(typeName, usprepTypeNames[USPREP_PROHIBITED])!=NULL){
 
 
     }else if(uprv_strstr(typeName, usprepTypeNames[USPREP_PROHIBITED])!=NULL){
 

-        u_parseCodePointRange(fields[0][0], &rangeStart,&rangeEnd, pErrorCode);
+        u_parseCodePointRange(s, &rangeStart,&rangeEnd, pErrorCode);

         if(U_FAILURE(*pErrorCode)){
             fprintf(stderr, "Could not parse code point range. Error: %s\n",u_errorName(*pErrorCode));
             return;
         if(U_FAILURE(*pErrorCode)){
             fprintf(stderr, "Could not parse code point range. Error: %s\n",u_errorName(*pErrorCode));
             return;


@@ -365,8 +397,8 @@ strprepProfileLineFn(void *context,
     }else if(uprv_strstr(typeName, usprepTypeNames[USPREP_MAP])!=NULL){
 
         /* get the character code, field 0 */
     }else if(uprv_strstr(typeName, usprepTypeNames[USPREP_MAP])!=NULL){
 
         /* get the character code, field 0 */

-        code=(uint32_t)uprv_strtoul(fields[0][0], &end, 16);
-        if(end<=fields[0][0] || end!=fields[0][1]) {
+        code=(uint32_t)uprv_strtoul(s, &end, 16);
+        if(end<=s || end!=fields[0][1]) {

             fprintf(stderr, "gensprep: syntax error in field 0 at %s\n", fields[0][0]);
             *pErrorCode=U_PARSE_ERROR;
             exit(U_PARSE_ERROR);
             fprintf(stderr, "gensprep: syntax error in field 0 at %s\n", fields[0][0]);
             *pErrorCode=U_PARSE_ERROR;
             exit(U_PARSE_ERROR);