def group_best():
best_1a = ['aga', 'apa', 'apb', 'app',
'art', 'ced', 'gmd', 'ibm',
'ipm', 'pri', 'syn', 'uqb',
'val']
best_1b = ['aga', 'apa', 'apb', 'app', 'art',
'ced', 'ibm', 'kai', 'lou', 'pri',
'syn', 'tos', 'upb', 'val',
'cea', 'ceb']
best_1c = ['aga', 'apa', 'apb', 'app', 'art',
'cea', 'ceb', 'ced', 'gmd',
'ibm', 'imp', 'kai', 'lav', 'lou',
'mot', 'pri', 'sie',
'syn', 'tos', 'val']
best_dbs = best_1c
charcols = []
for db_name in best_dbs:
db_file = 'unipen_db/' + db_name + '.chardb'
charcol = CharacterCollection(db_file)
# print db_name, charcol.get_total_n_characters()
charcols.append(charcol)
charcol_best = CharacterCollection()
charcol_best.merge(charcols)
print charcol_best.get_total_n_characters()
charcol_best.save("unipen_db/best_1c.chardb")
# unipen_to_sqlite()
# group_unipen_db()
# group_best()
def get_aggregated_charcol(tuples, dbpath=None):
"""
Create a character collection out of other character collections,
character directories, tomoe dictionaries or kuchibue databases.
tuples: a list of tuples (TYPE, path list)
"""
# number of files for each character collection type
n_files = [len(t[1]) for t in tuples]
# we don't need to merge character collections if only one is provided
# this can save a lot of time for large collections
if sum(n_files) == 1 and dbpath is None:
idx = n_files.index(1)
return _get_charcol(tuples[idx][0], tuples[idx][1][0])
if dbpath is not None and dbpath.endswith(".chardb"):
if os.path.exists(dbpath):
print "%s exists already." % dbpath
print "Continuing will modify it..."
answer = raw_input("Continue anyway? (y/N)")
if answer == "y":
print "Overwrite to concatenate collections together " + \
"in a new database"
print "Don't overwrite to append new characters or " + \
"filter (-i,-e,-m) existing database"
answer = raw_input("Overwrite it? (y/N)")
if answer == "y":
os.unlink(dbpath)
else:
exit()
charcol = CharacterCollection(dbpath)
#charcol.WRITE_BACK = False
#charcol.AUTO_COMMIT = True
else:
charcol = CharacterCollection() # in memory db
charcols = [_get_charcol(typ, path) \
for typ, paths in tuples for path in paths]
charcol.merge(charcols)
return charcol
def get_aggregated_charcol(tuples, dbpath=None):
"""
Create a character collection out of other character collections,
character directories, tomoe dictionaries or kuchibue databases.
tuples: a list of tuples (TYPE, path list)
"""
# number of files for each character collection type
n_files = [len(t[1]) for t in tuples]
# we don't need to merge character collections if only one is provided
# this can save a lot of time for large collections
if sum(n_files) == 1 and dbpath is None:
idx = n_files.index(1)
return _get_charcol(tuples[idx][0], tuples[idx][1][0])
if dbpath is not None and dbpath.endswith(".chardb"):
if os.path.exists(dbpath):
print "%s exists already." % dbpath
print "Continuing will modify it..."
answer = raw_input("Continue anyway? (y/N)")
if answer == "y":
print "Overwrite to concatenate collections together " + \
"in a new database"
print "Don't overwrite to append new characters or " + \
"filter (-i,-e,-m) existing database"
answer = raw_input("Overwrite it? (y/N)")
if answer == "y":
os.unlink(dbpath)
else:
exit()
charcol = CharacterCollection(dbpath)
#charcol.WRITE_BACK = False
#charcol.AUTO_COMMIT = True
else:
charcol = CharacterCollection() # in memory db
charcols = [_get_charcol(typ, path) \
for typ, paths in tuples for path in paths]
charcol.merge(charcols)
return charcol
