def short_expansion(cur, abbrev):
database.execute(cur, ''' SELECT short_expansion FROM abbreviations WHERE abbrev = ? ;''', [abbrev])
row = cur.fetchone()
if row:
return row[0]
else:
return None
def short_expansion(cur, abbrev):
database.execute(
cur,
''' SELECT short_expansion FROM abbreviations WHERE abbrev = ? ;''',
[abbrev])
row = cur.fetchone()
if row:
return row[0]
else:
return None
def abbrevs_table(cur):
database.execute(cur, '''
SELECT abbrev
FROM abbreviations
ORDER BY abbrev
;''')
abbrevs=[]
for row in cur.fetchall():
abbrevs.append(row[0])
return "\n".join([abbrev_line(cur, abbrev) for abbrev in abbrevs])
def abbrevs_table(cur):
database.execute(
cur, '''
SELECT abbrev
FROM abbreviations
ORDER BY abbrev
;''')
abbrevs = []
for row in cur.fetchall():
abbrevs.append(row[0])
return "\n".join([abbrev_line(cur, abbrev) for abbrev in abbrevs])
def search_by(cur, cond):
'''Main search function. Take a SearchCondition object, return list of ent_seqs.
'''
if ((cond.field == 'gloss' and cond.case_sensitive)
or cond.extent in ('whole', 'partial')):
fts = False
query_s = cond.query_s[:]
else:
fts = True
query_prep = []
for cmdline_arg in cond.query:
# unify Japanese spaces, newlines etc. to a single space
re.sub('\\s+', ' ', cmdline_arg)
# if the user called
#
# myougiden 'full phrase'
#
# or
# myougiden "full phrase"
#
# we'll get a single argv with spaces in it. in this case we
# translate the argv to the string
#
# "full phrase"
#
# (including the double quotes), which is what sqlite FTS looks
# for.
if ' ' in cmdline_arg:
cmdline_arg = '"' + cmdline_arg + '"'
if cond.extent == 'beginning':
cmdline_arg += '*'
query_prep.append(cmdline_arg)
query_s = ' '.join(query_prep)
if fts:
if cond.field == 'kanji':
table = 'kanjis_fts'
elif cond.field == 'reading':
table = 'readings_fts'
elif cond.field == 'gloss':
table = 'glosses_fts'
else:
if cond.field == 'kanji':
table = 'kanjis'
elif cond.field == 'reading':
table = 'readings'
elif cond.field == 'gloss':
table = 'glosses'
where_extra = ''
if cond.regexp or (not fts and cond.extent == 'word'):
# case sensitivity set for operator in opendb()
operator = 'REGEXP ?'
if cond.extent == 'whole':
query_s = '^' + query_s + '$'
elif cond.extent == 'beginning':
query_s = '^' + query_s
elif cond.extent == 'word':
query_s = r'\b' + query_s + r'\b'
else:
if fts:
operator = 'MATCH ?'
else:
if cond.extent == 'whole':
operator = '= ?'
query_s = query_s.replace('\\', '\\\\')
if cond.case_sensitive and cond.field == 'gloss':
where_extra = 'COLLATE BINARY'
else:
# extent = 'partial
# "\" seems to be the least common character in EDICT.
operator = r"LIKE ? ESCAPE '\'"
# my editor doesn't like raw strings
# query_s = query_s.replace(r'\', r'\\')
query_s = query_s.replace('\\', '\\\\')
query_s = query_s.replace('%', r'\%')
query_s = query_s.replace('_', r'\_')
query_s = '%' + query_s + '%'
if cond.frequent:
where_extra += ' AND %s.frequent = 1' % table
database.execute(cur, '''
SELECT DISTINCT ent_seq
FROM %s
WHERE %s %s %s
;'''
% (table, cond.field, operator, where_extra),
[query_s])
res = []
for row in cur.fetchall():
res.append(row[0])
return res
def fetch_entry(cur, ent_seq):
'''Return Entry object..'''
kanjis = []
readings = []
senses = []
database.execute(cur, '''SELECT
kanji_id,
kanji,
ke_inf,
frequent
FROM kanjis
WHERE ent_seq = ?;''', [ent_seq])
for row in cur.fetchall():
kanjis.append(Kanji(
kanji_id=row[0],
text=row[1],
ke_inf=row[2],
frequent=row[3],
))
database.execute(cur, '''SELECT
reading_id,
reading,
re_nokanji,
frequent,
re_inf
FROM readings
WHERE ent_seq = ?;''', [ent_seq])
for row in cur.fetchall():
reading = Reading(
reading_id=row[0],
text=row[1],
re_nokanji=row[2],
frequent=row[3],
re_inf=row[4],
)
database.execute(cur, '''SELECT re_restr
FROM reading_restrictions
WHERE reading_id = ?;''',
[reading.reading_id])
for row in cur.fetchall():
reading.re_restr.append(row[0])
readings.append(reading)
senses = []
database.execute(cur,
'''SELECT
sense_id,
pos,
field,
misc,
dial,
s_inf
FROM senses
WHERE ent_seq = ?;''',
[ent_seq]
)
for row in cur.fetchall():
sense = Sense(sense_id=row[0],
pos=row[1],
field=row[2],
misc=row[3],
dial=row[4],
s_inf=row[5])
database.execute(cur, '''
SELECT stagk
FROM sense_kanji_restrictions
WHERE sense_id = ?;
''', [sense.sense_id]
)
for row in cur.fetchall():
sense.stagk.append(row[0])
database.execute(cur, '''
SELECT stagr
FROM sense_reading_restrictions
WHERE sense_id = ?;
''', [sense.sense_id]
)
for row in cur.fetchall():
sense.stagr.append(row[0])
database.execute(cur, 'SELECT gloss FROM glosses WHERE sense_id = ?;', [sense.sense_id])
for row in cur.fetchall():
sense.glosses.append(row[0])
senses.append(sense)
database.execute(cur, 'SELECT frequent FROM entries WHERE ent_seq = ?;', [ent_seq])
frequent=cur.fetchone()[0]
return Entry(ent_seq=ent_seq,
frequent=frequent,
kanjis=kanjis,
readings=readings,
senses=senses,
)
def fetch_entry(cur, ent_seq):
'''Return Entry object..'''
kanjis = []
readings = []
senses = []
database.execute(
cur, '''SELECT
kanji_id,
kanji,
ke_inf,
frequent
FROM kanjis
WHERE ent_seq = ?;''', [ent_seq])
for row in cur.fetchall():
kanjis.append(
Kanji(
kanji_id=row[0],
text=row[1],
ke_inf=row[2],
frequent=row[3],
))
database.execute(
cur, '''SELECT
reading_id,
reading,
re_nokanji,
frequent,
re_inf
FROM readings
WHERE ent_seq = ?;''', [ent_seq])
for row in cur.fetchall():
reading = Reading(
reading_id=row[0],
text=row[1],
re_nokanji=row[2],
frequent=row[3],
re_inf=row[4],
)
database.execute(
cur, '''SELECT re_restr
FROM reading_restrictions
WHERE reading_id = ?;''', [reading.reading_id])
for row in cur.fetchall():
reading.re_restr.append(row[0])
readings.append(reading)
senses = []
database.execute(
cur, '''SELECT
sense_id,
pos,
field,
misc,
dial,
s_inf
FROM senses
WHERE ent_seq = ?;''', [ent_seq])
for row in cur.fetchall():
sense = Sense(sense_id=row[0],
pos=row[1],
field=row[2],
misc=row[3],
dial=row[4],
s_inf=row[5])
database.execute(
cur, '''
SELECT stagk
FROM sense_kanji_restrictions
WHERE sense_id = ?;
''', [sense.sense_id])
for row in cur.fetchall():
sense.stagk.append(row[0])
database.execute(
cur, '''
SELECT stagr
FROM sense_reading_restrictions
WHERE sense_id = ?;
''', [sense.sense_id])
for row in cur.fetchall():
sense.stagr.append(row[0])
database.execute(cur, 'SELECT gloss FROM glosses WHERE sense_id = ?;',
[sense.sense_id])
for row in cur.fetchall():
sense.glosses.append(row[0])
senses.append(sense)
database.execute(cur, 'SELECT frequent FROM entries WHERE ent_seq = ?;',
[ent_seq])
frequent = cur.fetchone()[0]
return Entry(
ent_seq=ent_seq,
frequent=frequent,
kanjis=kanjis,
readings=readings,
senses=senses,
)
def version(cur):
import argparse
import romkan
# import termcolor # no version
import platform
try:
import psutil
psutil_version = psutil.__version__
except ImportError:
psutil_version = None
if cur:
from myougiden import database
database.execute(
cur, '''
SELECT dbversion, jmdict_mtime
FROM versions;
''')
row = cur.fetchone()
dbversion = fmt(row[0], 'parameter')
jmdict_mtime = fmt(row[1], 'parameter')
else:
dbversion = fmt('Database not found!', 'error')
jmdict_mtime = fmt('Database not found!', 'error')
if config:
version = fmt(config.get('core', 'version'), 'parameter')
prefix = fmt(config.get('paths', 'prefix'), 'parameter')
else:
version = fmt('Config not found!', 'error')
prefix = fmt('Config not found!', 'error')
scripts = {}
for s in ('gzip', 'rsync', 'nice', 'ionice'):
path = which(s)
if path:
scripts[s] = fmt(path, 'parameter')
else:
scripts[s] = fmt(path, 'warning')
return ('''
myougiden version %s , database v. %s
JMdict last modified %s
Python version %s (%s)
Prefix: %s
Libraries:
romkan: %s
argparse: %s
psutil: %s
External programs:
gzip: %s
rsync: %s
nice: %s
ionice: %s
'''.strip() % (
version,
dbversion,
jmdict_mtime,
fmt(platform.python_version(), 'parameter'),
fmt(platform.platform(), 'parameter'),
prefix,
fmt(romkan.__version__, 'parameter'),
# fmt(termcolor.__version__, 'parameter'),
fmt(argparse.__version__, 'parameter'),
fmt(psutil_version, 'parameter'),
scripts['gzip'],
scripts['rsync'],
scripts['nice'],
scripts['ionice'],
))
def version(cur):
import argparse
import romkan
# import termcolor # no version
import platform
try:
import psutil
psutil_version = psutil.__version__
except ImportError:
psutil_version = None
if cur:
from myougiden import database
database.execute(cur, '''
SELECT dbversion, jmdict_mtime
FROM versions;
''')
row = cur.fetchone()
dbversion = fmt(row[0], 'parameter')
jmdict_mtime = fmt(row[1], 'parameter')
else:
dbversion = fmt('Database not found!', 'error')
jmdict_mtime = fmt('Database not found!', 'error')
if config:
version = fmt(config.get('core','version'), 'parameter')
prefix = fmt(config.get('paths','prefix'), 'parameter')
else:
version = fmt('Config not found!', 'error')
prefix = fmt('Config not found!', 'error')
scripts = {}
for s in ('gzip', 'rsync', 'nice', 'ionice'):
path = which(s)
if path:
scripts[s] = fmt(path, 'parameter')
else:
scripts[s] = fmt(path, 'warning')
return ('''
myougiden version %s , database v. %s
JMdict last modified %s
Python version %s (%s)
Prefix: %s
Libraries:
romkan: %s
argparse: %s
psutil: %s
External programs:
gzip: %s
rsync: %s
nice: %s
ionice: %s
'''.strip() % (
version,
dbversion,
jmdict_mtime,
fmt(platform.python_version(), 'parameter'),
fmt(platform.platform(), 'parameter'),
prefix,
fmt(romkan.__version__, 'parameter'),
# fmt(termcolor.__version__, 'parameter'),
fmt(argparse.__version__, 'parameter'),
fmt(psutil_version, 'parameter'),
scripts['gzip'],
scripts['rsync'],
scripts['nice'],
scripts['ionice'],
))
def search_by(cur, cond):
'''Main search function. Take a SearchCondition object, return list of ent_seqs.
'''
if ((cond.field == 'gloss' and cond.case_sensitive)
or cond.extent in ('whole', 'partial')):
fts=False
query_s = cond.query_s[:]
else:
fts=True
query_prep = []
for cmdline_arg in cond.query:
# unify Japanese spaces, newlines etc. to a single space
re.sub('\\s+', ' ', cmdline_arg)
# if the user called
#
# myougiden 'full phrase'
#
# or
# myougiden "full phrase"
#
# we'll get a single argv with spaces in it. in this case we
# translate the argv to the string
#
# "full phrase"
#
# (including the double quotes), which is what sqlite FTS looks
# for.
if ' ' in cmdline_arg:
cmdline_arg = '"' + cmdline_arg + '"'
if cond.extent == 'beginning':
cmdline_arg += '*'
query_prep.append(cmdline_arg)
query_s = ' '.join(query_prep)
if fts:
if cond.field == 'kanji':
table = 'kanjis_fts'
elif cond.field == 'reading':
table = 'readings_fts'
elif cond.field == 'gloss':
table = 'glosses_fts'
else:
if cond.field == 'kanji':
table = 'kanjis'
elif cond.field == 'reading':
table = 'readings'
elif cond.field == 'gloss':
table = 'glosses'
where_extra = ''
if cond.regexp or (not fts and cond.extent == 'word'):
# case sensitivity set for operator in opendb()
operator = 'REGEXP ?'
if cond.extent == 'whole':
query_s = '^' + query_s + '$'
elif cond.extent == 'beginning':
query_s = '^' + query_s
elif cond.extent == 'word':
query_s = r'\b' + query_s + r'\b'
else:
if fts:
operator = 'MATCH ?'
else:
if cond.extent == 'whole':
operator = '= ?'
query_s = query_s.replace('\\', '\\\\')
if cond.case_sensitive and cond.field == 'gloss':
where_extra = 'COLLATE BINARY';
else:
# extent = 'partial
# "\" seems to be the least common character in EDICT.
operator = r"LIKE ? ESCAPE '\'"
# my editor doesn't like raw strings
# query_s = query_s.replace(r'\', r'\\')
query_s = query_s.replace('\\', '\\\\')
query_s = query_s.replace('%', r'\%')
query_s = query_s.replace('_', r'\_')
query_s = '%' + query_s + '%'
if cond.frequent:
where_extra += ' AND %s.frequent = 1' % table
database.execute(cur, '''
SELECT DISTINCT ent_seq
FROM %s
WHERE %s %s %s
;'''
% (table, cond.field, operator, where_extra),
[query_s])
res = []
for row in cur.fetchall():
res.append(row[0])
return res