projects
/
yaz-moved-to-github.git
/ blobdiff
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
|
commitdiff
|
tree
raw
|
inline
| side by side
Change timeout parameter for yaz_poll.
[yaz-moved-to-github.git]
/
src
/
icu_I18N.c
diff --git
a/src/icu_I18N.c
b/src/icu_I18N.c
index
d529a19
..
5e32d6b
100644
(file)
--- a/
src/icu_I18N.c
+++ b/
src/icu_I18N.c
@@
-2,7
+2,7
@@
* Copyright (C) 1995-2007, Index Data ApS
* See the file LICENSE for details.
*
* Copyright (C) 1995-2007, Index Data ApS
* See the file LICENSE for details.
*
- * $Id: icu_I18N.c,v 1.15 2007-11-08 17:08:11 adam Exp $
+ * $Id: icu_I18N.c,v 1.19 2007-11-08 18:02:04 adam Exp $
*/
#if HAVE_CONFIG_H
*/
#if HAVE_CONFIG_H
@@
-14,8
+14,9
@@
#include <yaz/timing.h>
#endif
#include <yaz/timing.h>
#endif
-
#if YAZ_HAVE_ICU
#if YAZ_HAVE_ICU
+#include <yaz/xmalloc.h>
+
#include <yaz/icu_I18N.h>
#include <yaz/log.h>
#include <yaz/icu_I18N.h>
#include <yaz/log.h>
@@
-33,9
+34,9
@@
int icu_check_status (UErrorCode status)
{
int icu_check_status (UErrorCode status)
{
- if(U_FAILURE(status)){
- yaz_log(YLOG_WARN,
- "ICU: %d %s\n", status, u_errorName(status));
+ if (U_FAILURE(status))
+ {
+ yaz_log(YLOG_WARN, "ICU: %d %s\n", status, u_errorName(status));
return 0;
}
return 1;
return 0;
}
return 1;
@@
-47,14
+48,14
@@
int icu_check_status (UErrorCode status)
struct icu_buf_utf16 * icu_buf_utf16_create(size_t capacity)
{
struct icu_buf_utf16 * buf16
struct icu_buf_utf16 * icu_buf_utf16_create(size_t capacity)
{
struct icu_buf_utf16 * buf16
- = (struct icu_buf_utf16 *) malloc(sizeof(struct icu_buf_utf16));
+ = (struct icu_buf_utf16 *) xmalloc(sizeof(struct icu_buf_utf16));
buf16->utf16 = 0;
buf16->utf16_len = 0;
buf16->utf16_cap = 0;
if (capacity > 0){
buf16->utf16 = 0;
buf16->utf16_len = 0;
buf16->utf16_cap = 0;
if (capacity > 0){
- buf16->utf16 = (UChar *) malloc(sizeof(UChar) * capacity);
+ buf16->utf16 = (UChar *) xmalloc(sizeof(UChar) * capacity);
buf16->utf16[0] = (UChar) 0;
buf16->utf16_cap = capacity;
}
buf16->utf16[0] = (UChar) 0;
buf16->utf16_cap = capacity;
}
@@
-79,17
+80,16
@@
struct icu_buf_utf16 * icu_buf_utf16_resize(struct icu_buf_utf16 * buf16,
if (capacity > 0){
if (0 == buf16->utf16)
if (capacity > 0){
if (0 == buf16->utf16)
- buf16->utf16 = (UChar *) malloc(sizeof(UChar) * capacity);
+ buf16->utf16 = (UChar *) xmalloc(sizeof(UChar) * capacity);
else
buf16->utf16
else
buf16->utf16
- = (UChar *) realloc(buf16->utf16, sizeof(UChar) * capacity);
+ = (UChar *) xrealloc(buf16->utf16, sizeof(UChar) * capacity);
icu_buf_utf16_clear(buf16);
buf16->utf16_cap = capacity;
}
else {
icu_buf_utf16_clear(buf16);
buf16->utf16_cap = capacity;
}
else {
- if (buf16->utf16)
- free(buf16->utf16);
+ xfree(buf16->utf16);
buf16->utf16 = 0;
buf16->utf16_len = 0;
buf16->utf16_cap = 0;
buf16->utf16 = 0;
buf16->utf16_len = 0;
buf16->utf16_cap = 0;
@@
-118,29
+118,24
@@
struct icu_buf_utf16 * icu_buf_utf16_copy(struct icu_buf_utf16 * dest16,
void icu_buf_utf16_destroy(struct icu_buf_utf16 * buf16)
{
void icu_buf_utf16_destroy(struct icu_buf_utf16 * buf16)
{
- if (buf16){
- if (buf16->utf16)
- free(buf16->utf16);
- free(buf16);
- }
+ if (buf16)
+ xfree(buf16->utf16);
+ xfree(buf16);
}
}
-
-
-
struct icu_buf_utf8 * icu_buf_utf8_create(size_t capacity)
{
struct icu_buf_utf8 * buf8
struct icu_buf_utf8 * icu_buf_utf8_create(size_t capacity)
{
struct icu_buf_utf8 * buf8
- = (struct icu_buf_utf8 *) malloc(sizeof(struct icu_buf_utf8));
+ = (struct icu_buf_utf8 *) xmalloc(sizeof(struct icu_buf_utf8));
buf8->utf8 = 0;
buf8->utf8_len = 0;
buf8->utf8_cap = 0;
if (capacity > 0){
buf8->utf8 = 0;
buf8->utf8_len = 0;
buf8->utf8_cap = 0;
if (capacity > 0){
- buf8->utf8 = (uint8_t *) malloc(sizeof(uint8_t) * capacity);
+ buf8->utf8 = (uint8_t *) xmalloc(sizeof(uint8_t) * capacity);
buf8->utf8[0] = (uint8_t) 0;
buf8->utf8_cap = capacity;
}
buf8->utf8[0] = (uint8_t) 0;
buf8->utf8_cap = capacity;
}
@@
-167,17
+162,16
@@
struct icu_buf_utf8 * icu_buf_utf8_resize(struct icu_buf_utf8 * buf8,
if (capacity > 0){
if (0 == buf8->utf8)
if (capacity > 0){
if (0 == buf8->utf8)
- buf8->utf8 = (uint8_t *) malloc(sizeof(uint8_t) * capacity);
+ buf8->utf8 = (uint8_t *) xmalloc(sizeof(uint8_t) * capacity);
else
buf8->utf8
else
buf8->utf8
- = (uint8_t *) realloc(buf8->utf8, sizeof(uint8_t) * capacity);
+ = (uint8_t *) xrealloc(buf8->utf8, sizeof(uint8_t) * capacity);
icu_buf_utf8_clear(buf8);
buf8->utf8_cap = capacity;
}
else {
icu_buf_utf8_clear(buf8);
buf8->utf8_cap = capacity;
}
else {
- if (buf8->utf8)
- free(buf8->utf8);
+ xfree(buf8->utf8);
buf8->utf8 = 0;
buf8->utf8_len = 0;
buf8->utf8_cap = 0;
buf8->utf8 = 0;
buf8->utf8_len = 0;
buf8->utf8_cap = 0;
@@
-220,11
+214,9
@@
const char *icu_buf_utf8_to_cstr(struct icu_buf_utf8 *src8)
void icu_buf_utf8_destroy(struct icu_buf_utf8 * buf8)
{
void icu_buf_utf8_destroy(struct icu_buf_utf8 * buf8)
{
- if (buf8){
- if (buf8->utf8)
- free(buf8->utf8);
- free(buf8);
- }
+ if (buf8)
+ xfree(buf8->utf8);
+ xfree(buf8);
}
}
@@
-331,7
+323,7
@@
UErrorCode icu_utf16_to_utf8(struct icu_buf_utf8 * dest8,
struct icu_casemap * icu_casemap_create(char action, UErrorCode *status)
{
struct icu_casemap * casemap
struct icu_casemap * icu_casemap_create(char action, UErrorCode *status)
{
struct icu_casemap * casemap
- = (struct icu_casemap *) malloc(sizeof(struct icu_casemap));
+ = (struct icu_casemap *) xmalloc(sizeof(struct icu_casemap));
casemap->action = action;
switch(casemap->action) {
casemap->action = action;
switch(casemap->action) {
@@
-354,8
+346,7
@@
struct icu_casemap * icu_casemap_create(char action, UErrorCode *status)
void icu_casemap_destroy(struct icu_casemap * casemap)
{
void icu_casemap_destroy(struct icu_casemap * casemap)
{
- if (casemap)
- free(casemap);
+ xfree(casemap);
}
}
@@
-381,7
+372,7
@@
int icu_utf16_casemap(struct icu_buf_utf16 * dest16,
int32_t dest16_len = 0;
int32_t dest16_len = 0;
- if (!src16->utf16_len){ //guarding for empty source string
+ if (!src16->utf16_len){ /* guarding for empty source string */
if (dest16->utf16)
dest16->utf16[0] = (UChar) 0;
dest16->utf16_len = 0;
if (dest16->utf16)
dest16->utf16[0] = (UChar) 0;
dest16->utf16_len = 0;
@@
-507,9
+498,8
@@
struct icu_tokenizer * icu_tokenizer_create(const char *locale, char action,
UErrorCode *status)
{
struct icu_tokenizer * tokenizer
UErrorCode *status)
{
struct icu_tokenizer * tokenizer
- = (struct icu_tokenizer *) malloc(sizeof(struct icu_tokenizer));
+ = (struct icu_tokenizer *) xmalloc(sizeof(struct icu_tokenizer));
- strcpy(tokenizer->locale, locale);
tokenizer->action = action;
tokenizer->bi = 0;
tokenizer->buf16 = 0;
tokenizer->action = action;
tokenizer->bi = 0;
tokenizer->buf16 = 0;
@@
-522,33
+512,23
@@
struct icu_tokenizer * icu_tokenizer_create(const char *locale, char action,
switch(tokenizer->action) {
case 'l':
case 'L':
switch(tokenizer->action) {
case 'l':
case 'L':
- tokenizer->bi
- = ubrk_open(UBRK_LINE, tokenizer->locale,
- 0, 0, status);
+ tokenizer->bi = ubrk_open(UBRK_LINE, locale, 0, 0, status);
break;
case 's':
case 'S':
break;
case 's':
case 'S':
- tokenizer->bi
- = ubrk_open(UBRK_SENTENCE, tokenizer->locale,
- 0, 0, status);
+ tokenizer->bi = ubrk_open(UBRK_SENTENCE, locale, 0, 0, status);
break;
case 'w':
case 'W':
break;
case 'w':
case 'W':
- tokenizer->bi
- = ubrk_open(UBRK_WORD, tokenizer->locale,
- 0, 0, status);
+ tokenizer->bi = ubrk_open(UBRK_WORD, locale, 0, 0, status);
break;
case 'c':
case 'C':
break;
case 'c':
case 'C':
- tokenizer->bi
- = ubrk_open(UBRK_CHARACTER, tokenizer->locale,
- 0, 0, status);
+ tokenizer->bi = ubrk_open(UBRK_CHARACTER, locale, 0, 0, status);
break;
case 't':
case 'T':
break;
case 't':
case 'T':
- tokenizer->bi
- = ubrk_open(UBRK_TITLE, tokenizer->locale,
- 0, 0, status);
+ tokenizer->bi = ubrk_open(UBRK_TITLE, locale, 0, 0, status);
break;
default:
*status = U_UNSUPPORTED_ERROR;
break;
default:
*status = U_UNSUPPORTED_ERROR;
@@
-570,7
+550,7
@@
void icu_tokenizer_destroy(struct icu_tokenizer * tokenizer)
if (tokenizer) {
if (tokenizer->bi)
ubrk_close(tokenizer->bi);
if (tokenizer) {
if (tokenizer->bi)
ubrk_close(tokenizer->bi);
- free(tokenizer);
+ xfree(tokenizer);
}
}
}
}
@@
-694,7
+674,7
@@
struct icu_normalizer * icu_normalizer_create(const char *rules, char action,
{
struct icu_normalizer * normalizer
{
struct icu_normalizer * normalizer
- = (struct icu_normalizer *) malloc(sizeof(struct icu_normalizer));
+ = (struct icu_normalizer *) xmalloc(sizeof(struct icu_normalizer));
normalizer->action = action;
normalizer->trans = 0;
normalizer->action = action;
normalizer->trans = 0;
@@
-741,7
+721,7
@@
void icu_normalizer_destroy(struct icu_normalizer * normalizer){
icu_buf_utf16_destroy(normalizer->rules16);
if (normalizer->trans)
utrans_close(normalizer->trans);
icu_buf_utf16_destroy(normalizer->rules16);
if (normalizer->trans)
utrans_close(normalizer->trans);
- free(normalizer);
+ xfree(normalizer);
}
}
}
}
@@
-757,7
+737,7
@@
int icu_normalizer_normalize(struct icu_normalizer * normalizer,
|| !dest16)
return 0;
|| !dest16)
return 0;
- if (!src16->utf16_len){ //guarding for empty source string
+ if (!src16->utf16_len){ /* guarding for empty source string */
icu_buf_utf16_clear(dest16);
return 0;
}
icu_buf_utf16_clear(dest16);
return 0;
}
@@
-791,7
+771,7
@@
struct icu_chain_step * icu_chain_step_create(struct icu_chain * chain,
if(!chain || !type || !rule)
return 0;
if(!chain || !type || !rule)
return 0;
- step = (struct icu_chain_step *) malloc(sizeof(struct icu_chain_step));
+ step = (struct icu_chain_step *) xmalloc(sizeof(struct icu_chain_step));
step->type = type;
step->type = type;
@@
-844,23
+824,20
@@
void icu_chain_step_destroy(struct icu_chain_step * step){
default:
break;
}
default:
break;
}
- free(step);
+ xfree(step);
}
}
-struct icu_chain * icu_chain_create(const char *locale,
- int sort,
+struct icu_chain * icu_chain_create(const char *locale, int sort,
UErrorCode * status)
{
UErrorCode * status)
{
-
struct icu_chain * chain
struct icu_chain * chain
- = (struct icu_chain *) malloc(sizeof(struct icu_chain));
+ = (struct icu_chain *) xmalloc(sizeof(struct icu_chain));
*status = U_ZERO_ERROR;
*status = U_ZERO_ERROR;
- strncpy((char *) chain->locale, (const char *) locale, 16);
- chain->locale[16 - 1] = '\0';
+ chain->locale = xstrdup(locale);
chain->sort = sort;
chain->sort = sort;
@@
-869,7
+846,6
@@
struct icu_chain * icu_chain_create(const char *locale,
if (U_FAILURE(*status))
return 0;
if (U_FAILURE(*status))
return 0;
-
chain->token_count = 0;
chain->src8cstr = 0;
chain->token_count = 0;
chain->src8cstr = 0;
@@
-888,8
+864,8
@@
struct icu_chain * icu_chain_create(const char *locale,
void icu_chain_destroy(struct icu_chain * chain)
{
void icu_chain_destroy(struct icu_chain * chain)
{
- if (chain){
-
+ if (chain)
+ {
if (chain->coll)
ucol_close(chain->coll);
if (chain->coll)
ucol_close(chain->coll);
@@
-900,7
+876,8
@@
void icu_chain_destroy(struct icu_chain * chain)
icu_buf_utf16_destroy(chain->src16);
icu_chain_step_destroy(chain->steps);
icu_buf_utf16_destroy(chain->src16);
icu_chain_step_destroy(chain->steps);
- free(chain);
+ xfree(chain->locale);
+ xfree(chain);
}
}
}
}
@@
-942,35
+919,27
@@
struct icu_chain * icu_chain_xml_config(const xmlNode *xml_node,
xml_rule = xmlGetProp(node, (xmlChar *) "rule");
xml_rule = xmlGetProp(node, (xmlChar *) "rule");
- if (!strcmp((const char *) node->name,
- (const char *) "casemap")){
+ if (!strcmp((const char *) node->name, "casemap"))
step = icu_chain_insert_step(chain, ICU_chain_step_type_casemap,
(const uint8_t *) xml_rule, status);
step = icu_chain_insert_step(chain, ICU_chain_step_type_casemap,
(const uint8_t *) xml_rule, status);
- }
- else if (!strcmp((const char *) node->name,
- (const char *) "normalize")){
+ else if (!strcmp((const char *) node->name, "normalize"))
step = icu_chain_insert_step(chain, ICU_chain_step_type_normalize,
(const uint8_t *) xml_rule, status);
step = icu_chain_insert_step(chain, ICU_chain_step_type_normalize,
(const uint8_t *) xml_rule, status);
- }
- else if (!strcmp((const char *) node->name,
- (const char *) "tokenize")){
+ else if (!strcmp((const char *) node->name, "tokenize"))
step = icu_chain_insert_step(chain, ICU_chain_step_type_tokenize,
(const uint8_t *) xml_rule, status);
step = icu_chain_insert_step(chain, ICU_chain_step_type_tokenize,
(const uint8_t *) xml_rule, status);
- }
- else if (!strcmp((const char *) node->name,
- (const char *) "display")){
+ else if (!strcmp((const char *) node->name, "display"))
step = icu_chain_insert_step(chain, ICU_chain_step_type_display,
(const uint8_t *) "", status);
step = icu_chain_insert_step(chain, ICU_chain_step_type_display,
(const uint8_t *) "", status);
- }
xmlFree(xml_rule);
xmlFree(xml_rule);
- if (!step || U_FAILURE(*status)){
+ if (!step || U_FAILURE(*status))
+ {
icu_chain_destroy(chain);
return 0;
}
}
icu_chain_destroy(chain);
return 0;
}
}
-
return chain;
}
return chain;
}
@@
-998,7
+967,8
@@
struct icu_chain_step * icu_chain_insert_step(struct icu_chain * chain,
/* create utf16 destination buffers as needed, or */
/* create utf16 destination buffers as needed, or */
- switch(type) {
+ switch(type)
+ {
case ICU_chain_step_type_display:
buf16 = src16;
break;
case ICU_chain_step_type_display:
buf16 = src16;
break;
@@
-1038,7
+1008,8
@@
int icu_chain_step_next_token(struct icu_chain * chain,
/* assign utf16 src buffers as neeed, advance in previous steps
tokens until non-zero token met, and setting stop condition */
/* assign utf16 src buffers as neeed, advance in previous steps
tokens until non-zero token met, and setting stop condition */
- if (step->previous){
+ if (step->previous)
+ {
src16 = step->previous->buf16;
/* tokens might be killed in previous steps, therefore looping */
src16 = step->previous->buf16;
/* tokens might be killed in previous steps, therefore looping */
@@
-1048,7
+1019,8
@@
int icu_chain_step_next_token(struct icu_chain * chain,
got_new_token
= icu_chain_step_next_token(chain, step->previous, status);
}
got_new_token
= icu_chain_step_next_token(chain, step->previous, status);
}
- else { /* first step can only work once on chain->src16 input buffer */
+ else
+ { /* first step can only work once on chain->src16 input buffer */
src16 = chain->src16;
step->more_tokens = 0;
got_new_token = 1;
src16 = chain->src16;
step->more_tokens = 0;
got_new_token = 1;
@@
-1058,7
+1030,8
@@
int icu_chain_step_next_token(struct icu_chain * chain,
return 0;
/* stop if nothing to process */
return 0;
/* stop if nothing to process */
- if (step->need_new_token && !got_new_token){
+ if (step->need_new_token && !got_new_token)
+ {
step->more_tokens = 0;
return 0;
}
step->more_tokens = 0;
return 0;
}
@@
-1067,7
+1040,8
@@
int icu_chain_step_next_token(struct icu_chain * chain,
perform the work, eventually put this steps output in
step->buf16 or the chains UTF8 output buffers */
perform the work, eventually put this steps output in
step->buf16 or the chains UTF8 output buffers */
- switch(step->type) {
+ switch(step->type)
+ {
case ICU_chain_step_type_display:
icu_utf16_to_utf8(chain->display8, src16, status);
break;
case ICU_chain_step_type_display:
icu_utf16_to_utf8(chain->display8, src16, status);
break;
@@
-1082,12
+1056,12
@@
int icu_chain_step_next_token(struct icu_chain * chain,
break;
case ICU_chain_step_type_tokenize:
/* attach to new src16 token only first time during splitting */
break;
case ICU_chain_step_type_tokenize:
/* attach to new src16 token only first time during splitting */
- if (step->need_new_token){
+ if (step->need_new_token)
+ {
icu_tokenizer_attach(step->u.tokenizer, src16, status);
step->need_new_token = 0;
}
icu_tokenizer_attach(step->u.tokenizer, src16, status);
step->need_new_token = 0;
}
-
/* splitting one src16 token into multiple buf16 tokens */
step->more_tokens
= icu_tokenizer_next_token(step->u.tokenizer,
/* splitting one src16 token into multiple buf16 tokens */
step->more_tokens
= icu_tokenizer_next_token(step->u.tokenizer,
@@
-1096,11
+1070,11
@@
int icu_chain_step_next_token(struct icu_chain * chain,
/* make sure to get new previous token if this one had been used up
by recursive call to _same_ step */
/* make sure to get new previous token if this one had been used up
by recursive call to _same_ step */
- if (!step->more_tokens){
+ if (!step->more_tokens)
+ {
step->more_tokens = icu_chain_step_next_token(chain, step, status);
step->more_tokens = icu_chain_step_next_token(chain, step, status);
- return step->more_tokens; // avoid one token count too much!
+ return step->more_tokens; /* avoid one token count too much! */
}
}
-
break;
default:
return 0;
break;
default:
return 0;
@@
-1135,7
+1109,8
@@
int icu_chain_assign_cstr(struct icu_chain * chain,
chain->token_count = 0;
/* clear all steps stop states */
chain->token_count = 0;
/* clear all steps stop states */
- while (stp){
+ while (stp)
+ {
stp->more_tokens = 1;
stp->need_new_token = 1;
stp = stp->previous;
stp->more_tokens = 1;
stp->need_new_token = 1;
stp = stp->previous;
@@
-1164,10
+1139,12
@@
int icu_chain_next_token(struct icu_chain * chain,
return 0;
/* special case with no steps - same as index type binary */
return 0;
/* special case with no steps - same as index type binary */
- if (!chain->steps){
+ if (!chain->steps)
+ {
if (chain->token_count)
return 0;
if (chain->token_count)
return 0;
- else {
+ else
+ {
chain->token_count++;
if (chain->sort)
chain->token_count++;
if (chain->sort)
@@
-1178,12
+1155,13
@@
int icu_chain_next_token(struct icu_chain * chain,
}
}
/* usual case, one or more icu chain steps existing */
}
}
/* usual case, one or more icu chain steps existing */
- else {
-
+ else
+ {
while(!got_token && chain->steps && chain->steps->more_tokens)
got_token = icu_chain_step_next_token(chain, chain->steps, status);
while(!got_token && chain->steps && chain->steps->more_tokens)
got_token = icu_chain_step_next_token(chain, chain->steps, status);
- if (got_token){
+ if (got_token)
+ {
chain->token_count++;
icu_utf16_to_utf8(chain->norm8, chain->steps->buf16, status);
chain->token_count++;
icu_utf16_to_utf8(chain->norm8, chain->steps->buf16, status);
@@
-1192,7
+1170,7
@@
int icu_chain_next_token(struct icu_chain * chain,
icu_sortkey8_from_utf16(chain->coll,
chain->sort8, chain->steps->buf16,
status);
icu_sortkey8_from_utf16(chain->coll,
chain->sort8, chain->steps->buf16,
status);
-
+
return chain->token_count;
}
}
return chain->token_count;
}
}
@@
-1241,12
+1219,8
@@
const UCollator * icu_chain_get_coll(struct icu_chain * chain)
return chain->coll;
}
return chain->coll;
}
-
#endif /* YAZ_HAVE_ICU */
#endif /* YAZ_HAVE_ICU */
-
-
-
/*
* Local variables:
* c-basic-offset: 4
/*
* Local variables:
* c-basic-offset: 4