Adding in contents of enamdict

Former-commit-id: 2738889da33182d941c2179afbc5f3d776dc886a

build_dict.sh

@@ -2,7 +2,8 @@
 
 KANJIDIC=util/kanjidic
 EDICT=util/edict
+ENAMDICT=util/enamdict
 DICT=yomi_base/japanese/dictionary.db
 
 [ -f $DICT ] && rm $DICT
-util/compile.py --kanjidic $KANJIDIC --edict $EDICT $DICT
+util/compile.py --kanjidic $KANJIDIC --edict $EDICT $DICT --enamdict $ENAMDICT

util/compile.py

@@ -40,21 +40,31 @@ PARSED_TAGS = {
     'aux',     # auxiliary
     'aux-adj', # auxiliary adjective
     'aux-v',   # auxiliary verb
+    'c',       # company name
     'conj',    # conjunction
     'ctr',     # counter
     'exp',     # Expressions (phrases, clauses, etc.)
+    'f',       # female given name
+    'g',       # given name, as-yet not classified by sex
+    'h',       # full (usually family plus given) name of a particular person
     'int',     # interjection (kandoushi)
     'iv',      # irregular verb
+    'm',       # male given name
     'n',       # noun (common) (futsuumeishi)
     'n-adv',   # adverbial noun (fukushitekimeishi)
     'n-pref',  # noun, used as a prefix
     'n-suf',   # noun, used as a suffix
     'n-t',     # noun (temporal) (jisoumeishi)
     'num',     # numeric
+    'p',       # place-name
     'pn',      # pronoun
+    'pr',      # product name
     'pref' ,   # prefix
     'prt',     # particle
+    's',       # surname
+    'st',      # stations
     'suf',     # suffix
+    'u',       # person name, either given or surname, as-yet unclassified
     'v1',      # Ichidan verb
     'v2a-s',   # Nidan verb with 'u' ending (archaic)
     'v4h',     # Yodan verb with `hu/fu' ending (archaic)
@@ -172,18 +182,21 @@ def writeEdict(cursor, values):
     cursor.executemany('INSERT INTO Terms VALUES(?, ?, ?, ?)', values)
 
 
-def build(path, kanjidic, kradfile, edict):
+def build(path, kanjidic, kradfile, edict, enamdict):
     with sqlite3.connect(path) as db:
-        cursor = db.cursor()
-
         if kanjidic is not None:
-            writeKanjiDic(cursor, parseKanjiDic(kanjidic))
+            writeKanjiDic(db, parseKanjiDic(kanjidic))
 
         if kradfile is not None:
-            writeKradFile(cursor, parseKradFile(kradfile))
+            writeKradFile(db, parseKradFile(kradfile))
 
+        terms = []
         if edict is not None:
-            writeEdict(cursor, parseEdict(edict))
+            terms += parseEdict(edict)
+        if enamdict is not None:
+            terms += parseEdict(enamdict)
+        if len(terms) > 0:
+            writeEdict(db, terms)
 
 
 def main():
@@ -191,13 +204,20 @@ def main():
     parser.add_option('--kanjidic', dest='kanjidic')
     parser.add_option('--kradfile', dest='kradfile')
     parser.add_option('--edict', dest='edict')
+    parser.add_option('--enamdict', dest='enamdict')
 
     options, args = parser.parse_args()
 
     if len(args) == 0:
         parser.print_help()
     else:
-        build(args[0], options.kanjidic, options.kradfile, options.edict)
+        build(
+            args[0],
+            options.kanjidic,
+            options.kradfile,
+            options.edict,
+            options.enamdict
+        )
 
 
 if __name__ == '__main__':

yomi_base/constants.py

@@ -17,5 +17,5 @@
 
 
 c = {
-    'appVersion': '0.14b',
+    'appVersion': '0.15b',
 }

yomi_base/japanese/dictionary.db

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76e8f49f623012ca646b235b660c244070f0fff53811b77c429b907878c28585
-size 20201472
+oid sha256:7944b0babc6812a36c01265b8df517f15f5358dee415ece53736c5c2136aa5cb
+size 57806848

yomi_base/reader_util.py

@@ -240,10 +240,9 @@ def extractKindleDeck(filename):
     words = list()
 
     try:
-        connection = sqlite3.connect(unicode(filename))
-        for row in connection.execute('select word from WORDS'):
+        with sqlite3.connect(unicode(filename)) as db:
+            for row in db.execute('select word from WORDS'):
                 words.append(row[0])
-        connection.close()
     except sqlite3.OperationalError:
         pass