-/* $Id: test_icu_I18N.c,v 1.16 2007-05-11 08:27:29 marc Exp $
+/* $Id: test_icu_I18N.c,v 1.25 2007-05-24 10:35:21 adam Exp $
Copyright (c) 2006-2007, Index Data.
This file is part of Pazpar2.
printf("\n");
}
-
+
+
+ for( i = 0; i < src_list_len; i++)
+ free(list[i]);
+
+
ucol_close(coll);
icu_buf_utf8_destroy(buf8);
icu_buf_utf16_destroy(buf16);
-
-
return success;
}
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
-void test_icu_I18N_transliterator(int argc, char **argv)
+
+
+
+int test_icu_normalizer(const char * rules8cstr,
+ const char * src8cstr,
+ const char * chk8cstr)
{
+ int success = 0;
UErrorCode status = U_ZERO_ERROR;
- UParseError parse_error[256];
-
- const char * rules8cstr = "[:Punctuation:] Any-Remove";
- //const char * rules8cstr = "[:Control:] Any-Remove";
- //const char * rules8cstr = "[:Decimal_Number:] Any-Remove\n";
- //const char * rules8cstr = "[:Final_Punctuation:] Any-Remove";
- //const char * rules8cstr = "Lower; [:^Letter:] Remove";
- //const char * rules8cstr = "[:^Number:] Remove";
- //const char * rules8cstr = "Lower;[[:WhiteSpace:][:Punctuation:]] Remove";
- //const char * rules8cstr = "NFD; [:Nonspacing Mark:] Remove; NFC";
-
-
-
-
- const char * src8cstr = "Genesis 1\n"
- "The Beginning\n"
- "1 In the beginning God created the heavens and the earth.\n"
- "2 Now the earth was formless and empty, darkness was over "
- "the surface of the deep, and the Spirit of God was hovering "
- "over the waters.\n"
- "3 And God said, Let there be light, and there was light.\n";
-
-
- struct icu_buf_utf16 * rules16 = icu_buf_utf16_create(0);
struct icu_buf_utf16 * src16 = icu_buf_utf16_create(0);
struct icu_buf_utf16 * dest16 = icu_buf_utf16_create(0);
struct icu_buf_utf8 * dest8 = icu_buf_utf8_create(0);
-
- icu_utf16_from_utf8_cstr(rules16, rules8cstr, &status);
+ struct icu_normalizer * normalizer
+ = icu_normalizer_create(rules8cstr, 'f', &status);
icu_check_status(status);
-
+
icu_utf16_from_utf8_cstr(src16, src8cstr, &status);
icu_check_status(status);
- UTransliterator * trans
- = utrans_openU(rules16->utf16, rules16->utf16_len,
- UTRANS_FORWARD,
- 0, 0,
- parse_error, &status);
-
- //= utrans_openU(0, 0, UTRANS_FORWARD,
- // rules16->utf16, rules16->utf16_len,
- // parse_error, &status);
-
+ icu_normalizer_normalize(normalizer, dest16, src16, &status);
icu_check_status(status);
- if(U_FAILURE(status)) {
- printf("Parse Error: \n line %d offset %d \n '%s'\n",
- parse_error->line, parse_error->offset,
- rules8cstr);
- }
-
- utrans_transUChars (trans, src16->utf16, &(src16->utf16_len),
- src16->utf16_cap,
- 0, &(src16->utf16_len), &status);
- icu_utf16_to_utf8(dest8, src16, &status);
+ icu_utf16_to_utf8(dest8, dest16, &status);
icu_check_status(status);
- printf("Transliterator:\n%s\n", dest8->utf8);
+ if(!strcmp((const char *) dest8->utf8,
+ (const char *) chk8cstr))
+ success = 1;
+ else {
+ success = 0;
+ printf("Normalization\n");
+ printf("Rules: '%s'\n", rules8cstr);
+ printf("Input: '%s'\n", src8cstr);
+ printf("Normalized: '%s'\n", dest8->utf8);
+ printf("Expected: '%s'\n", chk8cstr);
+ }
+
- utrans_close (trans);
- icu_buf_utf16_destroy(rules16);
+ icu_normalizer_destroy(normalizer);
icu_buf_utf16_destroy(src16);
icu_buf_utf16_destroy(dest16);
icu_buf_utf8_destroy(dest8);
+ return success;
+};
-#if 0
-
- icu_check_status(status);
+// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
+
+void test_icu_I18N_normalizer(int argc, char **argv)
+{
- int32_t ustr16_lim = *ustr16_len;
- /* Transliterate a segment of a UChar* string */
+ YAZ_CHECK(test_icu_normalizer("[:Punctuation:] Any-Remove",
+ "Don't shoot!",
+ "Dont shoot"));
- utrans_transUChars (trans, ustr16, &*ustr16_len,
- ustr16_cap,
- 0, &ustr16_lim, &status);
+ YAZ_CHECK(test_icu_normalizer("[:Control:] Any-Remove",
+ "Don't\n shoot!",
+ "Don't shoot!"));
+
+ YAZ_CHECK(test_icu_normalizer("[:Decimal_Number:] Any-Remove",
+ "This is 4 you!",
+ "This is you!"));
+
+ YAZ_CHECK(test_icu_normalizer("Lower; [:^Letter:] Remove",
+ "Don't shoot!",
+ "dontshoot"));
- utrans_close (trans);
-
- printf("\n\nUnicode Set Patterns:\n"
- " Pattern Description\n"
- " Ranges [a-z] The lower case letters a through z\n"
- " Named Chars [abc123] The six characters a,b,c,1,2 and 3\n"
- " String [abc{def}] chars a, b and c, and string 'def'\n"
- " Categories [\\p{Letter}] Perl General Category 'Letter'.\n"
- " Categories [:Letter:] Posix General Category 'Letter'.\n"
- "\n"
- " Combination Example\n"
- " Union [[:Greek:] [:letter:]]\n"
- " Intersection [[:Greek:] & [:letter:]]\n"
- " Set Complement [[:Greek:] - [:letter:]]\n"
- " Complement [^[:Greek:] [:letter:]]\n"
- "\n"
- "see: http://icu.sourceforge.net/userguide/unicodeSet.html\n"
- "\n"
- "Examples:\n"
- " [:Punctuation:] Any-Remove\n"
- " [:Cased-Letter:] Any-Upper\n"
- " [:Control:] Any-Remove\n"
- " [:Decimal_Number:] Any-Remove\n"
- " [:Final_Punctuation:] Any-Remove\n"
- " [:Georgian:] Any-Upper\n"
- " [:Katakana:] Any-Remove\n"
- " [:Arabic:] Any-Remove\n"
- " [:Punctuation:] Remove\n"
- " [[:Punctuation:]-[.,]] Remove\n"
- " [:Line_Separator:] Any-Remove\n"
- " [:Math_Symbol:] Any-Remove\n"
- " Lower; [:^Letter:] Remove (word tokenization)\n"
- " [:^Number:] Remove (numeric tokenization)\n"
- " [:^Katagana:] Remove (remove everything except Katagana)\n"
- " Lower;[[:WhiteSpace:][:Punctuation:]] Remove (word tokenization)\n"
- " NFD; [:Nonspacing Mark:] Remove; NFC (removes accents from characters)\n"
- " [A-Za-z]; Lower(); Latin-Katakana; Katakana-Hiragana (transforms latin and katagana to hiragana)\n"
- " [[:separator:][:start punctuation:][:initial punctuation:]] Remove \n"
- "\n"
- "see http://icu.sourceforge.net/userguide/Transform.html\n"
- " http://www.unicode.org/Public/UNIDATA/UCD.html\n"
- " http://icu.sourceforge.net/userguide/Transform.html\n"
- " http://icu.sourceforge.net/userguide/TransformRule.html\n"
- );
-#endif
+ YAZ_CHECK(test_icu_normalizer("[:^Number:] Remove",
+ "Monday 15th of April",
+ "15"));
+
+ YAZ_CHECK(test_icu_normalizer("Lower;"
+ "[[:WhiteSpace:][:Punctuation:]] Remove",
+ " word4you? ",
+ "word4you"));
+
+
+ YAZ_CHECK(test_icu_normalizer("NFD; [:Nonspacing Mark:] Remove; NFC",
+ "à côté de l'alcôve ovoïde",
+ "a cote de l'alcove ovoide"));
}
+
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
int test_icu_tokenizer(const char * locale, char action,
// converting to UTF8
icu_utf16_to_utf8(tkn8, tkn16, &status);
- //printf("(%d)'%s' ", icu_tokenizer_token_id(tokenizer), tkn8->utf8);
-
//printf("token %d %d %d %d '%s'\n",
//
// icu_tokenizer_token_start(tokenizer),
// icu_tokenizer_token_length(tokenizer),
// tkn8->utf8);
}
- //printf("\nTokens: %d\n", icu_tokenizer_token_count(tokenizer));
-
if (count != icu_tokenizer_token_count(tokenizer)){
success = 0;
}
+void test_icu_I18N_chain(int argc, char **argv)
+{
+ const char * en_str
+ = "O Romeo, Romeo! wherefore art thou\t Romeo?";
+
+ printf("ICU chain:\ninput: '%s'\n", en_str);
+ UErrorCode status = U_ZERO_ERROR;
+ //struct icu_chain_step * step = 0;
+ struct icu_chain * chain = 0;
+
+ const char * xml_str = "<icu_chain id=\"en:word\" locale=\"en\">"
+ "<normalize rule=\"[:Control:] Any-Remove\"/>"
+ "<tokenize rule=\"l\"/>"
+ "<normalize rule=\"[[:WhiteSpace:][:Punctuation:]] Remove\"/>"
+ "<display/>"
+ "<casemap rule=\"l\"/>"
+ "<normal/>"
+ "<sort/>"
+ "</icu_chain>";
+
+
+ xmlDoc *doc = xmlParseMemory(xml_str, strlen(xml_str));
+ xmlNode *xml_node = xmlDocGetRootElement(doc);
+ YAZ_CHECK(xml_node);
+
+
+ chain = icu_chain_xml_config(xml_node, &status);
+
+#if 0
+ chain = icu_chain_create((uint8_t *) "en:word", (uint8_t *) "en");
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_normalize,
+ (const uint8_t *) "[:Control:] Any-Remove",
+ &status);
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_tokenize,
+ (const uint8_t *) "s",
+ &status);
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_tokenize,
+ (const uint8_t *) "l",
+ &status);
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_normalize,
+ (const uint8_t *)
+ "[[:WhiteSpace:][:Punctuation:]] Any-Remove",
+ &status);
+ 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_normalize, */
+/* (const uint8_t *) "Lower", */
+/* &status); */
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_casemap,
+ (const uint8_t *) "l",
+ &status);
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_norm,
+ (const uint8_t *)"",
+ &status);
+/* step = icu_chain_insert_step(chain, ICU_chain_step_type_sort, */
+/* (const uint8_t *)"", */
+/* &status); */
+
+#endif
+
+ xmlFreeDoc(doc);
+ YAZ_CHECK(chain);
+
+ YAZ_CHECK(icu_chain_assign_cstr(chain, en_str, &status));
+
+ while (icu_chain_next_token(chain, &status)){
+ printf("%d '%s' '%s'\n",
+ icu_chain_get_token_count(chain),
+ icu_chain_get_norm(chain),
+ icu_chain_get_display(chain));
+ }
+
+ YAZ_CHECK_EQ(icu_chain_get_token_count(chain), 7);
+
+
+ YAZ_CHECK(icu_chain_assign_cstr(chain, "what is this?", &status));
+
+ while (icu_chain_next_token(chain, &status)){
+ printf("%d '%s' '%s'\n",
+ icu_chain_get_token_count(chain),
+ icu_chain_get_norm(chain),
+ icu_chain_get_display(chain));
+ }
+
+
+ YAZ_CHECK_EQ(icu_chain_get_token_count(chain), 3);
+
+ icu_chain_destroy(chain);
+}
+
+
+void test_bug_1140(void)
+{
+ const char * en_str
+ = "O Romeo, Romeo! wherefore art thou\t Romeo?";
+
+ printf("ICU chain:\ninput: '%s'\n", en_str);
+
+ UErrorCode status = U_ZERO_ERROR;
+ //struct icu_chain_step * step = 0;
+ struct icu_chain * chain = 0;
+
+ const char * xml_str = "<icu_chain id=\"en:word\" locale=\"en\">"
+
+ /* if the first rule is normalize instead. Then it works */
+#if 0
+ "<normalize rule=\"[:Control:] Any-Remove\"/>"
+#endif
+ "<tokenize rule=\"l\"/>"
+ "<normalize rule=\"[[:WhiteSpace:][:Punctuation:]] Remove\"/>"
+ "<display/>"
+ "<casemap rule=\"l\"/>"
+ "<normal/>"
+ "<sort/>"
+ "</icu_chain>";
+
+
+ xmlDoc *doc = xmlParseMemory(xml_str, strlen(xml_str));
+ xmlNode *xml_node = xmlDocGetRootElement(doc);
+ YAZ_CHECK(xml_node);
+
+ chain = icu_chain_xml_config(xml_node, &status);
+
+ xmlFreeDoc(doc);
+ YAZ_CHECK(chain);
+
+ YAZ_CHECK(icu_chain_assign_cstr(
+ chain, "O Romeo, Romeo! wherefore art thou\t Romeo?",
+ &status));
+
+ while (icu_chain_next_token(chain, &status))
+ ;
+
+ YAZ_CHECK_EQ(icu_chain_get_token_count(chain), 7);
+
+ YAZ_CHECK(icu_chain_assign_cstr(chain, "what is this?", &status));
+
+ while (icu_chain_next_token(chain, &status)){
+ printf("%d '%s' '%s'\n",
+ icu_chain_get_token_count(chain),
+ icu_chain_get_norm(chain),
+ icu_chain_get_display(chain));
+ }
+
+ /* we expect 'what' 'is' 'this', i.e. 3 tokens */
+ YAZ_CHECK_EQ(icu_chain_get_token_count(chain), 3);
+
+ icu_chain_destroy(chain);
+}
#endif // HAVE_ICU
//test_icu_I18N_casemap_failures(argc, argv);
test_icu_I18N_casemap(argc, argv);
test_icu_I18N_sortmap(argc, argv);
- test_icu_I18N_transliterator(argc, argv);
+ test_icu_I18N_normalizer(argc, argv);
test_icu_I18N_tokenizer(argc, argv);
+ test_icu_I18N_chain(argc, argv);
+#if 0
+ /* currently fails */
+ test_bug_1140();
+#endif
#else // HAVE_ICU