modified tables.symbol_rule{} and unichar_half_to_full()

Japanese zenkaku symbol convert rule:

(1) original tables.symbol_rule{}
(2) ibus.unichar_half_to_full()
(3) rule for Windows, Locale: "EUC-JISX0213", "UTF-8"
(4) rule for Mac, Locale: "EUC-JP"

       (1)     (2)     (3)     (4)
'"' - U+201D, U+FF02, U+201D, U+201D
"'" - U+2019, U+FF07, U+2019, U+2019
"-" - U+30FC, U+FF0D, U+FF0D, U+2212
"`" - U+2018, U+FF40, U+2018, U+2018
"~" - U+301C, U+FF5E, U+FF5E, U+301C

I selected (3).

4 files changed, 20 insertions, 25 deletions

diff --git a/engine/engine.py b/engine/engine.py
index aaf6531..6ccbabf 100644
--- a/engine/engine.py
+++ b/engine/engine.py
@@ -29,6 +29,7 @@ from tables import *
 from ibus import keysyms
 from ibus import modifier
 import jastring
+from segment import unichar_half_to_full
 
 sys.path.append(path.join(os.getenv('IBUS_ANTHY_PKGDATADIR'), 'setup'))
 from anthyprefs import AnthyPrefs
@@ -829,18 +830,7 @@ class Engine(ibus.EngineBase):
         elif mode == 2:
             if self.__convert_mode == CONV_MODE_HALF_WIDTH_KATAKANA:
                 return True
-            if self.__convert_mode == CONV_MODE_HIRAGANA or \
-                self.__convert_mode == CONV_MODE_KATAKANA or \
-                self.__convert_mode == CONV_MODE_OFF or \
-                self.__convert_mode == CONV_MODE_ANTHY:
-                self.__convert_mode = CONV_MODE_HALF_WIDTH_KATAKANA
-            else:
-                if CONV_MODE_LATIN_0 <= self.__convert_mode <= CONV_MODE_LATIN_3:
-                    self.__convert_mode += 1
-                    if self.__convert_mode > CONV_MODE_LATIN_3:
-                        self.__convert_mode = CONV_MODE_LATIN_1
-                else:
-                    self.__convert_mode = CONV_MODE_LATIN_0
+            self.__convert_mode = CONV_MODE_HALF_WIDTH_KATAKANA
         elif mode == 3:
             if CONV_MODE_WIDE_LATIN_0 <= self.__convert_mode <= CONV_MODE_WIDE_LATIN_3:
                 self.__convert_mode += 1
@@ -874,7 +864,7 @@ class Engine(ibus.EngineBase):
             char = unichr(keyval)
             wide_char = None#symbol_rule.get(char, None)
             if wide_char == None:
-                wide_char = ibus.unichar_half_to_full(char)
+                wide_char = unichar_half_to_full(char)
             self.__commit_string(wide_char)
             return True
 
@@ -1079,7 +1069,7 @@ class Engine(ibus.EngineBase):
         char = unichr(keysyms.space)
         wide_char = symbol_rule.get(char, None)
         if wide_char == None:
-            wide_char = ibus.unichar_half_to_full(char)
+            wide_char = unichar_half_to_full(char)
         self.__commit_string(wide_char)
         return True
 
diff --git a/engine/jastring.py b/engine/jastring.py
index c9cd734..f913d0d 100644
--- a/engine/jastring.py
+++ b/engine/jastring.py
@@ -22,7 +22,7 @@
 import romaji
 import kana
 
-from ibus import unichar_half_to_full
+from segment import unichar_half_to_full
 
 SymbolTable = {}
 for i in range(32, 127):
diff --git a/engine/segment.py b/engine/segment.py
index 7d968fe..a91607a 100644
--- a/engine/segment.py
+++ b/engine/segment.py
@@ -19,9 +19,14 @@
 # along with this program; if not, write to the Free Software
 # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 
-from ibus import unichar_half_to_full
+#from ibus import unichar_half_to_full
+from ibus import unichar_half_to_full as h_to_f
 from tables import *
 
+def unichar_half_to_full(c):
+    tdl = {'"': u'\u201d', "'": u'\u2019', '`': u'\u2018'}
+    return tdl[c] if c in tdl else h_to_f(c)
+
 class Segment(object):
     def __init__(self, enchars=u"", jachars=u""):
         self._enchars = enchars
diff --git a/engine/tables.py b/engine/tables.py
index dcccf69..107a4e9 100644
--- a/engine/tables.py
+++ b/engine/tables.py
@@ -237,21 +237,21 @@ symbol_rule = {
     u","  : u"、",
     u"."  : u"。",
     u"!"  : u"！",
-    u"\"" : u"＂",
+    u"\"" : u"\u201d",
     u"#"  : u"＃",
     u"$"  : u"＄",
     u"%"  : u"％",
     u"&"  : u"＆",
-    u"'"  : u"＇",
+    u"'"  : u"\u2019",
     u"("  : u"（",
     u")"  : u"）",
-    u"~"  : u"～",
-    u"-"  : u"－",
+    u"~"  : u"\uff5e",
+    u"-"  : u"ー",
     u"="  : u"＝",
     u"^"  : u"＾",
     u"\\" : u"＼",
     u"|"  : u"｜",
-    u"`"  : u"｀",
+    u"`"  : u"\u2018",
     u"@"  : u"＠",
     u"{"  : u"｛",
     u"["  : u"「",
@@ -584,20 +584,20 @@ hiragana_katakana_table = {
     u"、" : (u"、", u"､"),
     u"。" : (u"。", u"｡"),
     u"！" : (u"！", u"!"),
-    u"＂" : (u"＂", u"\""),
+    u"\u201d"  : (u"\u201d", u"\""),
     u"＃" : (u"＃", u"#"),
     u"＄" : (u"＄", u"$"),
     u"％" : (u"％", u"%"),
     u"＆" : (u"＆", u"&"),
-    u"＇" : (u"＇", u"'"),
+    u"\u2019"  : (u"\u2019", u"'"),
     u"（" : (u"（", u"("),
     u"）" : (u"）", u")"),
-    u"～" : (u"～", u"~"),
+    u"\uff5e" : (u"\uff5e", u"~"),
     u"＝" : (u"＝", u"="),
     u"＾" : (u"＾", u"^"),
     u"＼" : (u"＼", u"\\"),
     u"｜" : (u"｜", u"|"),
-    u"｀" : (u"｀", u"`"),
+    u"\u2018"  : (u"\u2018", u"`"),
     u"＠" : (u"＠", u"@"),
     u"｛" : (u"｛", u"{"),
     u"「" : (u"「", u"｢"),