def words_by_reading(self, char, reading, **kwargs):
"""Returns a list of Word objects composed of every word in the
solutions database that contains char with reading."""
r_id = reading_query.get_id(char, reading)
return self.words_by_reading_id(r_id, **kwargs)
def db_populate_kanji_readings():
print "Filling database with kanji/reading data..."
conn = get_connection()
c = conn.cursor()
kd_conn = sqlite3.connect(KANJIDIC_PATH)
kd_conn.row_factory = sqlite3.Row
kd_c = kd_conn.cursor()
reading_l = []
start = time.time()
s = "SELECT * FROM reading WHERE r_type='ja_on' OR r_type='ja_kun'"
readings = kd_c.execute(s).fetchall()
bases_used = {}
lastchar = None
for r in readings:
char = r['character_literal']
if char != lastchar:
bases_used.clear()
reading = r['reading']
if reading[-1] == u"-":
reading = reading[:-1]
elif reading[0] == u"-":
reading = reading[1:]
(base,s,oku) = reading.partition('.')
if base not in bases_used:
bases_used[base] = True
reading_l.append([char, base, r['r_type']])
reading_l.append([char, reading, r['r_type']])
lastchar = char
c.executemany('''INSERT INTO reading(character, reading, type)
VALUES (?,?,?)''', reading_l)
conn.commit()
conn = get_connection()
c = conn.cursor()
reading_l = []
#Now add our own, non-kanjidic entries
f = codecs.open(OTHER_READINGS_PATH, encoding='utf-8')
for line in f:
line = line.strip('\n')
(char, s, reading) = line.partition(",")
#Search the existing entries for the same entry, which might be in
#a different character set (so db constraint won't pick up on it).
#We don't want to add any duplicates.
id = reading_query.get_id(char, reading)
if id is None:
reading_l.append([char, reading, 'other'])
c.executemany('''INSERT INTO reading(character, reading, type)
VALUES (?,?,?)''', reading_l)
conn.commit()
print 'Filling database with kanji/reading data took '\
'%s seconds' % (time.time() - start)
in_keb = '%%'
if count > 0:
for n in range(0, count):
in_keb += '%s%%' % char
s = 'select word, reading from word where word.word like ?'
results = self.conn.execute(s, [in_keb]).fetchall()
for r in results:
word_list.append(Word(self.conn, r['word'], r['reading']))
return word_list
def clear_words(self):
"""Remove all existing words from the database."""
self.conn.execute('delete from word')
if __name__ == '__main__':
#dbpath = 'dbs/test_query.sqlite'
dbpath = 'dbs/jmdict_solutions.sqlite'
#word_db.create_db(dbpath)
q = WordQuery(dbpath)
# words = q.contains_char(u'帰')
# for w in words:
# print w.word, w.reading
id = reading_query.get_id(u'帰', u'かえ.る')
words = q.words_by_reading_id(id)
for w in words:
print w.word, w.reading
print 'found', len(words)
