-/* $Id: test_icu_I18N.c,v 1.11 2007-05-09 14:01:21 marc Exp $
+/* $Id: test_icu_I18N.c,v 1.17 2007-05-11 22:23:33 marc Exp $
Copyright (c) 2006-2007, Index Data.
This file is part of Pazpar2.
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
-void test_icu_I18N_normmap(int argc, char **argv)
+int test_icu_transliterator(const char * rules8cstr,
+ const char * src8cstr,
+ const char * chk8cstr)
{
+ int success = 0;
+
+ UErrorCode status = U_ZERO_ERROR;
+ UParseError parse_error[256];
+
+
+ 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);
+ 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_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_check_status(status);
+
+ if(!strcmp((const char *) dest8->utf8,
+ (const char *) chk8cstr))
+ success = 1;
+ else {
+ success = 0;
+ printf("Normaliozation;");
+ 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_buf_utf16_destroy(src16);
+ icu_buf_utf16_destroy(dest16);
+ icu_buf_utf8_destroy(dest8);
+
+ return success;
}
+#if 0
+ 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
+
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
-void test_icu_I18N_tokenizer(int argc, char **argv)
+void test_icu_I18N_transliterator(int argc, char **argv)
{
- const char * src8cstr
- = "Though I am not naturally honest, I am so sometimes by chance.";
+ YAZ_CHECK(test_icu_transliterator("[:Punctuation:] Any-Remove",
+ "Don't shoot!",
+ "Dont shoot"));
+
+ YAZ_CHECK(test_icu_transliterator("[:Control:] Any-Remove",
+ "Don't\n shoot!",
+ "Don't shoot!"));
+
+ YAZ_CHECK(test_icu_transliterator("[:Decimal_Number:] Any-Remove",
+ "This is 4 you!",
+ "This is you!"));
+
+ YAZ_CHECK(test_icu_transliterator("Lower; [:^Letter:] Remove",
+ "Don't shoot!",
+ "dontshoot"));
+
+ YAZ_CHECK(test_icu_transliterator("[:^Number:] Remove",
+ "Monday 15th of April",
+ "15"));
+
+ YAZ_CHECK(test_icu_transliterator("Lower;"
+ "[[:WhiteSpace:][:Punctuation:]] Remove",
+ " word4you? ",
+ "word4you"));
+
+
+ YAZ_CHECK(test_icu_transliterator("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,
+ const char * src8cstr, int count)
+{
+ int success = 1;
UErrorCode status = U_ZERO_ERROR;
struct icu_buf_utf16 * src16 = icu_buf_utf16_create(0);
struct icu_buf_utf16 * tkn16 = icu_buf_utf16_create(0);
struct icu_buf_utf8 * tkn8 = icu_buf_utf8_create(0);
- printf("Input: '%s'\n", src8cstr);
+ //printf("Input: '%s'\n", src8cstr);
// transforming to UTF16
icu_utf16_from_utf8_cstr(src16, src8cstr, &status);
// set up tokenizer
struct icu_tokenizer * tokenizer
- = icu_tokenizer_create("en", 's', &status);
+ = icu_tokenizer_create(locale, action, &status);
icu_check_status(status);
YAZ_CHECK(tokenizer);
YAZ_CHECK(tokenizer->bi);
// perform work on tokens
- printf("Tokens: ");
+ //printf("Tokens: ");
while(icu_tokenizer_next_token(tokenizer, tkn16, &status)){
icu_check_status(status);
// converting to UTF8
icu_utf16_to_utf8(tkn8, tkn16, &status);
- printf("'%s' ", tkn8->utf8);
+ //printf("(%d)'%s' ", icu_tokenizer_token_id(tokenizer), tkn8->utf8);
//printf("token %d %d %d %d '%s'\n",
- // icu_tokenizer_token_id(tokenizer),
+ //
// icu_tokenizer_token_start(tokenizer),
// icu_tokenizer_token_end(tokenizer),
// icu_tokenizer_token_length(tokenizer),
// tkn8->utf8);
}
- printf(" (%d)(%d)\n", icu_tokenizer_token_id(tokenizer),
- icu_tokenizer_token_count(tokenizer));
+ //printf("\nTokens: %d\n", icu_tokenizer_token_count(tokenizer));
+
+
+ if (count != icu_tokenizer_token_count(tokenizer)){
+ success = 0;
+ printf("\nTokenizer '%s:%c' Error: \n", locale, action);
+ printf("Input: '%s'\n", src8cstr);
+ printf("Tokens: %d", icu_tokenizer_token_count(tokenizer));
+ printf(", expected: %d\n", count);
+ }
icu_tokenizer_destroy(tokenizer);
icu_buf_utf16_destroy(src16);
icu_buf_utf16_destroy(tkn16);
icu_buf_utf8_destroy(tkn8);
+
+ return success;
+}
+
+
+// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
+
+void test_icu_I18N_tokenizer(int argc, char **argv)
+{
+
+
+ const char * en_str
+ = "O Romeo, Romeo! wherefore art thou Romeo?";
+
+ YAZ_CHECK(test_icu_tokenizer("en", 's', en_str, 2));
+ YAZ_CHECK(test_icu_tokenizer("en", 'l', en_str, 7));
+ YAZ_CHECK(test_icu_tokenizer("en", 'w', en_str, 16));
+ YAZ_CHECK(test_icu_tokenizer("en", 'c', en_str, 41));
+
+
+
+ const char * da_str
+ = "Blåbærtærte. Denne kage stammer fra Finland. "
+ "Den er med blåbær, men alle sommerens forskellige bær kan bruges.";
+
+ YAZ_CHECK(test_icu_tokenizer("da", 's', da_str, 3));
+ YAZ_CHECK(test_icu_tokenizer("dar", 'l', da_str, 17));
+ YAZ_CHECK(test_icu_tokenizer("da", 'w', da_str, 37));
+ YAZ_CHECK(test_icu_tokenizer("da", 'c', da_str, 110));
+
}
//test_icu_I18N_casemap_failures(argc, argv);
test_icu_I18N_casemap(argc, argv);
test_icu_I18N_sortmap(argc, argv);
- test_icu_I18N_normmap(argc, argv);
+ test_icu_I18N_transliterator(argc, argv);
test_icu_I18N_tokenizer(argc, argv);
#else // HAVE_ICU
-// CRAP to follow
-#if 0
-
-// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
-
-void test_icu_I18N_casemap_failures(int argc, char **argv)
-{
-
- size_t buf_cap = 128;
- char buf[buf_cap];
- size_t dest8_len = 0;
- NMEM nmem = nmem_create();
- char * dest8 = 0;
-
- const char * src8 = "A ReD fOx hunTS sQUirriLs";
- //size_t src8_len = strlen(src8);
-
- //printf("original string: '%s' (%d)\n", src8, (int) src8_len);
-
- // some calling error needs investigation
- dest8 = icu_casemap(nmem, buf, buf_cap, &dest8_len,
- src8, "en", 't');
- YAZ_CHECK(0 == dest8_len);
- //printf("icu_casemap 'en:t' '%s' (%d)\n", dest8, (int) dest8_len);
-
-
- // attention: does not fail even if no locale 'xy_zz' defined
- // it seems to default to english locale
- dest8 = icu_casemap(nmem, buf, buf_cap, &dest8_len,
- src8, "zz_abc", 'l');
- YAZ_CHECK(dest8_len);
- //printf("icu_casemap 'zz:l' '%s' (%d)\n", dest8, (int) dest8_len);
-
-
- // shall fail - no buf buffer defined
- dest8 = icu_casemap(nmem, 0, buf_cap, &dest8_len,
- src8, "en", 'l');
- YAZ_CHECK(0 == dest8_len);
- //printf("icu_casemap 'en:l' '%s' (%d)\n", dest8, (int) dest8_len);
-
- // shall fail - no buf_cap defined
- dest8 = icu_casemap(nmem, buf, 0, &dest8_len,
- src8, "en", 'l');
- YAZ_CHECK(0 == dest8_len);
- //printf("icu_casemap 'en:l' '%s' (%d)\n", dest8, (int) dest8_len);
-
- // shall fail - no action 'x' defined
- dest8 = icu_casemap(nmem, buf, buf_cap, &dest8_len,
- src8, "en", 'x');
- YAZ_CHECK(0 == dest8_len);
- //printf("icu_casemap 'en:x' '%s' (%d)\n", dest8, (int) dest8_len);
-
- nmem_destroy(nmem);
-}
-
-
-
-#endif
-
-
-
/*
* Local variables:
* c-basic-offset: 4
projects / pazpar2-moved-to-github.git / blobdiff