U_CDECL_BEGIN
static void U_CALLCONV
strprepProfileLineFn(void * /*context*/,
char *fields[][2], int32_t fieldCount,
UErrorCode *pErrorCode) {
uint32_t mapping[40];
char *end, *map;
uint32_t code;
int32_t length;
/*UBool* mapWithNorm = (UBool*) context;*/
const char* typeName;
uint32_t rangeStart=0,rangeEnd =0;
const char *s;
s = u_skipWhitespace(fields[0][0]);
if (*s == '@') {
/* a special directive introduced in 4.2 */
return;
}
if(fieldCount != 3){
*pErrorCode = U_INVALID_FORMAT_ERROR;
return;
}
typeName = fields[2][0];
map = fields[1][0];
if(uprv_strstr(typeName, usprepTypeNames[USPREP_UNASSIGNED])!=NULL){
u_parseCodePointRange(s, &rangeStart,&rangeEnd, pErrorCode);
/* store the range */
compareFlagsForRange(rangeStart,rangeEnd,USPREP_UNASSIGNED);
}else if(uprv_strstr(typeName, usprepTypeNames[USPREP_PROHIBITED])!=NULL){
u_parseCodePointRange(s, &rangeStart,&rangeEnd, pErrorCode);
/* store the range */
compareFlagsForRange(rangeStart,rangeEnd,USPREP_PROHIBITED);
}else if(uprv_strstr(typeName, usprepTypeNames[USPREP_MAP])!=NULL){
/* get the character code, field 0 */
code=(uint32_t)uprv_strtoul(s, &end, 16);
/* parse the mapping string */
length=u_parseCodePoints(map, mapping, sizeof(mapping)/4, pErrorCode);
/* store the mapping */
compareMapping(code,mapping, length,USPREP_MAP);
}else{
*pErrorCode = U_INVALID_FORMAT_ERROR;
}
}
static void U_CALLCONV
normalizationCorrectionsLineFn(void *context,
char *fields[][2], int32_t fieldCount,
UErrorCode *pErrorCode) {
uint32_t mapping[40];
char *end, *s;
uint32_t code;
int32_t length;
UVersionInfo version;
UVersionInfo thisVersion;
/* get the character code, field 0 */
code=(uint32_t)uprv_strtoul(fields[0][0], &end, 16);
if(U_FAILURE(*pErrorCode)) {
fprintf(stderr, "gensprep: error parsing NormalizationCorrections.txt mapping at %s\n", fields[0][0]);
exit(*pErrorCode);
}
/* Original (erroneous) decomposition */
s = fields[1][0];
/* parse the mapping string */
length=u_parseCodePoints(s, mapping, sizeof(mapping)/4, pErrorCode);
/* ignore corrected decomposition */
u_versionFromString(version,fields[3][0] );
u_versionFromString(thisVersion, "3.2.0");
if(U_FAILURE(*pErrorCode)) {
fprintf(stderr, "gensprep error parsing NormalizationCorrections.txt of U+%04lx - %s\n",
(long)code, u_errorName(*pErrorCode));
exit(*pErrorCode);
}
/* store the mapping */
if( version[0] > thisVersion[0] ||
((version[0]==thisVersion[0]) && (version[1] > thisVersion[1]))
){
storeMapping(code,mapping, length, USPREP_MAP, pErrorCode);
}
setUnicodeVersionNC(version);
}
static void U_CALLCONV
strprepProfileLineFn(void *context,
char *fields[][2], int32_t fieldCount,
UErrorCode *pErrorCode) {
uint32_t mapping[40];
char *end, *map;
uint32_t code;
int32_t length;
UStringPrepProfile* data = (UStringPrepProfile*) context;
const char* typeName;
uint32_t rangeStart=0,rangeEnd =0;
typeName = fields[2][0];
map = fields[1][0];
if(strstr(typeName, usprepTypeNames[USPREP_UNASSIGNED])!=NULL){
u_parseCodePointRange(fields[0][0], &rangeStart,&rangeEnd, pErrorCode);
/* store the range */
compareFlagsForRange(data, rangeStart,rangeEnd,USPREP_UNASSIGNED);
}else if(strstr(typeName, usprepTypeNames[USPREP_PROHIBITED])!=NULL){
u_parseCodePointRange(fields[0][0], &rangeStart,&rangeEnd, pErrorCode);
/* store the range */
compareFlagsForRange(data, rangeStart,rangeEnd,USPREP_PROHIBITED);
}else if(strstr(typeName, usprepTypeNames[USPREP_MAP])!=NULL){
/* get the character code, field 0 */
code=(uint32_t)uprv_strtoul(fields[0][0], &end, 16);
/* parse the mapping string */
length=u_parseCodePoints(map, mapping, sizeof(mapping)/4, pErrorCode);
/* compare the mapping */
compareMapping(data, code,mapping, length,USPREP_MAP);
}else{
*pErrorCode = U_INVALID_FORMAT_ERROR;
}
}
static void U_CALLCONV
unicodeDataLineFn(void *context,
char *fields[][2], int32_t fieldCount,
UErrorCode *pErrorCode) {
uint32_t decomp[40];
Norm norm;
const char *s;
char *end;
uint32_t code, value;
int32_t length;
UBool isCompat, something=FALSE;
/* ignore First and Last entries for ranges */
if( *fields[1][0]=='<' &&
(length=(int32_t)(fields[1][1]-fields[1][0]))>=9 &&
(0==uprv_memcmp(", First>", fields[1][1]-8, 8) || 0==uprv_memcmp(", Last>", fields[1][1]-7, 7))
) {
return;
}
/* reset the properties */
uprv_memset(&norm, 0, sizeof(Norm));
/* 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]) {
fprintf(stderr, "gennorm: syntax error in field 0 at %s\n", fields[0][0]);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
/* get canonical combining class, field 3 */
value=(uint32_t)uprv_strtoul(fields[3][0], &end, 10);
if(end<=fields[3][0] || end!=fields[3][1] || value>0xff) {
fprintf(stderr, "gennorm: syntax error in field 3 at %s\n", fields[0][0]);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
if(value>0) {
norm.udataCC=(uint8_t)value;
something=TRUE;
}
/* get the decomposition, field 5 */
if(fields[5][0]<fields[5][1]) {
if(*(s=fields[5][0])=='<') {
++s;
isCompat=TRUE;
/* skip and ignore the compatibility type name */
do {
if(s==fields[5][1]) {
/* missing '>' */
fprintf(stderr, "gennorm: syntax error in field 5 at %s\n", fields[0][0]);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
} while(*s++!='>');
} else {
isCompat=FALSE;
}
/* parse the decomposition string */
length=u_parseCodePoints(s, decomp, sizeof(decomp)/4, pErrorCode);
if(U_FAILURE(*pErrorCode)) {
fprintf(stderr, "gennorm error parsing UnicodeData.txt decomposition of U+%04lx - %s\n",
(long)code, u_errorName(*pErrorCode));
exit(*pErrorCode);
}
/* store the string */
if(length>0) {
something=TRUE;
if(isCompat) {
norm.lenNFKD=(uint8_t)length;
norm.nfkd=decomp;
} else {
if(length>2) {
fprintf(stderr, "gennorm: error - length of NFD(U+%04lx) = %ld >2 in UnicodeData - illegal\n",
(long)code, (long)length);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
norm.lenNFD=(uint8_t)length;
norm.nfd=decomp;
}
}
}
/* check for non-character code points */
if((code&0xfffe)==0xfffe || (uint32_t)(code-0xfdd0)<0x20 || code>0x10ffff) {
fprintf(stderr, "gennorm: error - properties for non-character code point U+%04lx\n",
(long)code);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
if(something) {
/* there are normalization values, so store them */
#if 0
if(beVerbose) {
printf("store values for U+%04lx: cc=%d, lenNFD=%ld, lenNFKD=%ld\n",
(long)code, norm.udataCC, (long)norm.lenNFD, (long)norm.lenNFKD);
}
#endif
storeNorm(code, &norm);
}
}
static void U_CALLCONV
strprepProfileLineFn(void *context,
char *fields[][2], int32_t fieldCount,
UErrorCode *pErrorCode) {
uint32_t mapping[40];
char *end, *map;
uint32_t code;
int32_t length;
/*UBool* mapWithNorm = (UBool*) 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){
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;
}
/* store the range */
storeRange(rangeStart,rangeEnd,USPREP_UNASSIGNED, pErrorCode);
}else if(uprv_strstr(typeName, usprepTypeNames[USPREP_PROHIBITED])!=NULL){
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;
}
/* store the range */
storeRange(rangeStart,rangeEnd,USPREP_PROHIBITED, pErrorCode);
}else if(uprv_strstr(typeName, usprepTypeNames[USPREP_MAP])!=NULL){
/* get the character code, field 0 */
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);
}
/* parse the mapping string */
length=u_parseCodePoints(map, mapping, sizeof(mapping)/4, pErrorCode);
/* store the mapping */
storeMapping(code,mapping, length,USPREP_MAP, pErrorCode);
}else{
*pErrorCode = U_INVALID_FORMAT_ERROR;
}
if(U_FAILURE(*pErrorCode)) {
fprintf(stderr, "gensprep error parsing %s line %s at %s. Error: %s\n",filename,
fields[0][0],fields[2][0],u_errorName(*pErrorCode));
exit(*pErrorCode);
}
}
