Imported anthy zipcode.t into ibus-anthy.

5 files changed, 68 insertions, 42 deletions

diff --git a/configure.ac b/configure.ac
index 60b86a5..ff5fbef 100644
--- a/configure.ac
+++ b/configure.ac
@@ -249,6 +249,7 @@ AC_ARG_WITH(anthy-zipcode,
            [ANTHY_ZIPCODE_FILE="$withval"],
            [ANTHY_ZIPCODE_FILE="$datadir/anthy/zipcode.t"])
 AC_SUBST(ANTHY_ZIPCODE_FILE)
+AM_CONDITIONAL([ENABLE_ZIPCODE], [test -f "$ANTHY_ZIPCODE_FILE" ])
 
 dnl - check file path of kasumi bin
 AC_ARG_WITH(kasumi-exec,
diff --git a/data/Makefile.am b/data/Makefile.am
index 3aa240a..f187c93 100644
--- a/data/Makefile.am
+++ b/data/Makefile.am
@@ -27,8 +27,18 @@ dicts_DATA = \
 
 dictsdir = $(pkgdatadir)/dicts
 
+if ENABLE_ZIPCODE
+dicts_DATA += zipcode.t
+
+zipcode.t: $(ANTHY_ZIPCODE_FILE)
+	$(PYTHON) zipcode-textdic.py $<
+
+CLEANFILES = zipcode.t
+endif
+
 EXTRA_DIST = \
         era.t \
         oldchar.t \
         symbol.t \
+        zipcode-textdic.py \
         $(NULL)
diff --git a/data/zipcode-textdic.py b/data/zipcode-textdic.py
new file mode 100755
index 0000000..2864f85
--- /dev/null
+++ b/data/zipcode-textdic.py
@@ -0,0 +1,54 @@
+#!/usr/bin/python
+
+# Finally decided to import anthy zipcode.t with UTF-8 into ibus-anthy
+# because if digits without hyphen is grepped by engine, it could cause
+# the timeout issue. If digits without hyphen are sent to anthy,
+# digits with hyphen also need to be sent to anthy so the lookup could
+# include too many and unnecessary candidates.
+# Also wish to install the filename of 'zipcode.t' to simplify enigne.
+
+import sys
+
+if len(sys.argv) < 2:
+    print >> sys.stderr, 'usage: %s /usr/share/anthy/zipcode.t' % sys.argv[0]
+    exit(-1)
+
+anthy_zipfile = sys.argv[1]
+
+try:
+    contents = unicode(open(anthy_zipfile).read(), 'euc_jp').encode('utf-8')
+except UnicodeDecodeError, e:
+    print >> sys.stderr, 'Your file is not eucJP? %s' % anthy_zipfile
+    contents = open(anthy_zipfile).read()
+
+output_zipfile = open('zipcode.t', 'w')
+output_zipfile.write('# copied %s with UTF-8.\n#\n' % anthy_zipfile)
+
+for line in contents.split('\n'):
+    if len(line) == 0 or line[0] == '#':
+        output_zipfile.write('%s\n' % line)
+        continue
+
+    words = line.split()
+    if len(words) < 3:
+        continue
+
+    if len(words[0]) < 1 or ord(unicode(words[0], 'utf-8')[0]) > 0xff:
+        mbcs_addr = words[0]
+    else:
+        uni_addr = ''
+        i = 0
+        for word in words[0]:
+            # Convert ASCII number char to wide number char.
+            uni_addr += unichr(0xfee0 + ord(word))
+            if i == 2:
+                # Insert wide hyphen
+                uni_addr += unichr(0x30fc)
+            i += 1
+        mbcs_addr = uni_addr.encode('utf-8')
+
+    output_zipfile.write('%s %s %s\n' % \
+            (mbcs_addr, '#T35*500', words[2]))
+
+output_zipfile.flush()
+output_zipfile.close()
diff --git a/engine/engine.py b/engine/engine.py
index a31c58b..ef160d4 100644
--- a/engine/engine.py
+++ b/engine/engine.py
@@ -1062,43 +1062,6 @@ class Engine(IBus.EngineSimple):
                     candidate = candidate.replace(key, value)
                     self.__lookup_table.append_candidate(IBus.Text.new_from_string(candidate))
 
-    def __fill_anthy_zipcode_strip(self, dict_file, id):
-        import re
-        text = self.__preedit_ja_string.get_latin()[0]
-        if text.find('-') < 0:
-            return
-        text = text.replace('-', '')
-        section = 'dict/file/' + id
-        if 'encoding' not in self.__prefs.keys(section):
-            section = 'dict/file/default'
-        encoding = self.__prefs.get_value(section, 'encoding')
-        contents = unicode(open(dict_file).read(), encoding)
-        expression = re.compile('^' + text + '[ \t]')
-
-        found = False
-        dict_dest = None
-        for line in contents.split('\n'):
-            matched = expression.search(line)
-            if matched:
-                found = True
-                dict_dest = UN(matched.string).split(' ')[2]
-                break
-        if found:
-            self.__lookup_table.append_candidate(IBus.Text.new_from_string(dict_dest))
-
-    def __fill_lookup_table_dict_mode(self):
-        if Engine.__dict_mode <= 0:
-            return
-        single_files = self.__get_single_dict_files()
-        file = single_files[Engine.__dict_mode - 1]
-        if file == None:
-            return
-        id = self.__get_dict_id_from_file(file)
-        if id == None:
-            return
-        if id == 'zipcode':
-            self.__fill_anthy_zipcode_strip(file, id)
-
     def __fill_lookup_table(self):
         if self.__convert_mode == CONV_MODE_PREDICTION:
             nr_predictions = self.__context.get_nr_predictions()
@@ -1122,7 +1085,6 @@ class Engine(IBus.EngineSimple):
             candidate = UN(buf)
             self.__lookup_table.append_candidate(IBus.Text.new_from_string(candidate))
             self.__candidate_cb(candidate)
-        self.__fill_lookup_table_dict_mode()
 
 
     def __invalidate(self):
diff --git a/setup/anthyprefs.py.in b/setup/anthyprefs.py.in
index f2a3b64..a685da1 100644
--- a/setup/anthyprefs.py.in
+++ b/setup/anthyprefs.py.in
@@ -1070,12 +1070,12 @@ _config = {
     },
 
     'dict': {
-        'zipcode': ['@ANTHY_ZIPCODE_FILE@'],
+        'zipcode': ['@pkgdatadir@/dicts/zipcode.t'],
         'symbol': ['@pkgdatadir@/dicts/symbol.t'],
         'oldchar': ['@pkgdatadir@/dicts/oldchar.t'],
         'era': ['@pkgdatadir@/dicts/era.t'],
         'files': [
-            '@ANTHY_ZIPCODE_FILE@',
+            '@pkgdatadir@/dicts/zipcode.t',
             '@pkgdatadir@/dicts/symbol.t',
             '@pkgdatadir@/dicts/oldchar.t',
             '@pkgdatadir@/dicts/era.t',
@@ -1112,9 +1112,8 @@ _config = {
         'short_label': '〒',
         'long_label': N_("Zip Code Conversion"),
         'preview_lines': 30,
-        'reverse': True,
+        'reverse': False,
         'is_system': True,
-        'encoding': 'euc_jp',
     },
 
     'dict/file/symbol': {