[Groonga-commit] groonga/groonga at 987f80f [master] NormalizerNFKC100: fix unify_kana conversion for n-sounds

Back to archive index

Kouhei Sutou null+****@clear*****
Wed May 2 13:04:44 JST 2018


Kouhei Sutou	2018-05-02 13:04:44 +0900 (Wed, 02 May 2018)

  New Revision: 987f80f09a8f7fdb5b2459a3a5830d7215851635
  https://github.com/groonga/groonga/commit/987f80f09a8f7fdb5b2459a3a5830d7215851635

  Message:
    NormalizerNFKC100: fix unify_kana conversion for n-sounds

  Modified files:
    lib/normalizer.c
    test/command/suite/normalizers/nfkc100/unify_kana.expected
    test/command/suite/normalizers/nfkc100/unify_kana.test

  Modified: lib/normalizer.c (+5 -1)
===================================================================
--- lib/normalizer.c    2018-05-01 17:37:46 +0900 (6ea189872)
+++ lib/normalizer.c    2018-05-02 13:04:44 +0900 (ea218b0ac)
@@ -660,7 +660,11 @@ utf8_normalize_unify_kana(const unsigned char *utf8_char,
        (utf8_char[1] == 0x83 && (0xbd <= utf8_char[2] &&
                                  utf8_char[2] <= 0xbe)))) {
     unified[0] = utf8_char[0];
-    unified[1] = utf8_char[1] - 1;
+    if (utf8_char[2] & 0x20) {
+      unified[1] = utf8_char[1] - 1;
+    } else {
+      unified[1] = utf8_char[1] - 2;
+    }
     unified[2] = utf8_char[2] ^ 0x20;
     return unified;
   }

  Modified: test/command/suite/normalizers/nfkc100/unify_kana.expected (+12 -2)
===================================================================
--- test/command/suite/normalizers/nfkc100/unify_kana.expected    2018-05-01 17:37:46 +0900 (d1f315517)
+++ test/command/suite/normalizers/nfkc100/unify_kana.expected    2018-05-02 13:04:44 +0900 (1e5befaf4)
@@ -1,4 +1,4 @@
-normalize   'NormalizerNFKC100("unify_kana", true)'   "あイウェおヽヾ"   WITH_TYPES
+normalize   'NormalizerNFKC100("unify_kana", true)'   "あイウェおざジたチなニぱピまミヽヾ"   WITH_TYPES
 [
   [
     0,
@@ -6,7 +6,7 @@ normalize   'NormalizerNFKC100("unify_kana", true)'   "あイウェおヽヾ"
     0.0
   ],
   {
-    "normalized": "あいうぇおゝゞ",
+    "normalized": "あいうぇおざじたちなにぱぴまみゝゞ",
     "types": [
       "hiragana",
       "hiragana",
@@ -14,6 +14,16 @@ normalize   'NormalizerNFKC100("unify_kana", true)'   "あイウェおヽヾ"
       "hiragana",
       "hiragana",
       "hiragana",
+      "hiragana",
+      "hiragana",
+      "hiragana",
+      "hiragana",
+      "hiragana",
+      "hiragana",
+      "hiragana",
+      "hiragana",
+      "hiragana",
+      "hiragana",
       "hiragana"
     ],
     "checks": [

  Modified: test/command/suite/normalizers/nfkc100/unify_kana.test (+1 -1)
===================================================================
--- test/command/suite/normalizers/nfkc100/unify_kana.test    2018-05-01 17:37:46 +0900 (818bc56d3)
+++ test/command/suite/normalizers/nfkc100/unify_kana.test    2018-05-02 13:04:44 +0900 (b797d6b6e)
@@ -1,4 +1,4 @@
 normalize \
   'NormalizerNFKC100("unify_kana", true)' \
-  "あイウェおヽヾ" \
+  "あイウェおざジたチなニぱピまミヽヾ" \
   WITH_TYPES
-------------- next part --------------
HTML����������������������������...
URL: https://lists.osdn.me/mailman/archives/groonga-commit/attachments/20180502/409f6f73/attachment-0001.htm 



More information about the Groonga-commit mailing list
Back to archive index