def add_nominal_stems(ctx):
"""Add regular noun stems to the database."""
# Since there are so many nominal stems, the SQLAlchemy calls here
# are a little more low-level.
conn = ctx.engine.connect()
ins = NominalStem.__table__.insert()
gender_group = ENUM['gender_group']
buf = []
i = 0
tick = util.tick_every(5000)
for row in util.read_csv(ctx.config['NOMINAL_STEMS']):
genders_id = gender_group[row['stem_genders']]
buf.append({
'name': row['stem'],
'pos_id': Tag.NOMINAL,
'genders_id': genders_id,
})
tick(row['stem'])
i += 1
if i % 500 == 0:
conn.execute(ins, buf)
buf = []
# Add any remainder.
if buf:
conn.execute(ins, buf)
def add_nominal_stems(ctx):
"""Add regular noun stems to the database."""
# Since there are so many nominal stems, the SQLAlchemy calls here
# are a little more low-level.
conn = ctx.engine.connect()
ins = NominalStem.__table__.insert()
gender_group = ENUM['gender_group']
buf = []
i = 0
tick = util.tick_every(5000)
for row in util.read_csv(ctx.config['NOMINAL_STEMS']):
genders_id = gender_group[row['stem_genders']]
buf.append({
'name': row['stem'],
'pos_id': Tag.NOMINAL,
'genders_id': genders_id,
})
tick(row['stem'])
i += 1
if i % 500 == 0:
conn.execute(ins, buf)
buf = []
# Add any remainder.
if buf:
conn.execute(ins, buf)
def add_indeclinables(ctx):
"""Add indeclinables to the database."""
session = ctx.session
tick = util.tick_every(300)
for row in util.read_csv(ctx.config['INDECLINABLES']):
session.add(Indeclinable(name=row['name']))
tick(row['name'])
session.commit()
session.close()
def add_indeclinables(ctx):
"""Add indeclinables to the database."""
session = ctx.session
tick = util.tick_every(300)
for row in util.read_csv(ctx.config['INDECLINABLES']):
session.add(Indeclinable(name=row['name']))
tick(row['name'])
session.commit()
session.close()
def add_roots(ctx, prefix_map):
"""Populates :class:`Root` and its subclasses."""
# TODO: modified roots
session = ctx.session
e_vclass = ENUM['vclass']
e_voice = ENUM['voice']
root_map = {} # (name, hom) -> id
# First pass: Root
tick = util.tick_every(100)
for row in util.read_csv(ctx.config['UNPREFIXED_ROOTS']):
name, hom = row['root'], row['hom']
# A root can have multiple paradigms (= multiple appearances)
if (name, hom) in root_map:
continue
root = Root(name=name)
session.add(root)
session.flush()
root_map[(name, hom)] = root.id
tick(name)
# Second pass: Paradigm
for row in util.read_csv(ctx.config['UNPREFIXED_ROOTS']):
name, hom = row['root'], row['hom']
vclass, voice = row['class'], row['voice']
assert (name, hom) in root_map
root_id = root_map[(name, hom)]
paradigm = Paradigm(root_id=root_id, vclass_id=e_vclass[vclass],
voice_id=e_voice[voice])
session.add(paradigm)
session.commit()
# Prefixed roots
for i, row in enumerate(util.read_csv(ctx.config['PREFIXED_ROOTS'])):
name = row['prefixed_root']
basis = row['unprefixed_root']
hom = row['hom']
prefixes = row['prefixes'].split('-')
assert (basis, hom) in root_map
basis_id = root_map[(basis, hom)]
for prefix in prefixes:
# TODO
pass
prefixed_root = PrefixedRoot(name=name, basis_id=basis_id)
session.add(prefixed_root)
session.flush()
root_map[(name, hom)] = prefixed_root.id
tick(name)
session.commit()
session.close()
return root_map
def add_roots(ctx, prefix_map):
"""Populates :class:`Root` and its subclasses."""
# TODO: modified roots
session = ctx.session
e_vclass = ENUM['vclass']
e_voice = ENUM['voice']
root_map = {} # (name, hom) -> id
# First pass: Root
tick = util.tick_every(100)
for row in util.read_csv(ctx.config['UNPREFIXED_ROOTS']):
name, hom = row['root'], row['hom']
# A root can have multiple paradigms (= multiple appearances)
if (name, hom) in root_map:
continue
root = Root(name=name)
session.add(root)
session.flush()
root_map[(name, hom)] = root.id
tick(name)
# Second pass: Paradigm
for row in util.read_csv(ctx.config['UNPREFIXED_ROOTS']):
name, hom = row['root'], row['hom']
vclass, voice = row['class'], row['voice']
assert (name, hom) in root_map
root_id = root_map[(name, hom)]
paradigm = Paradigm(root_id=root_id,
vclass_id=e_vclass[vclass],
voice_id=e_voice[voice])
session.add(paradigm)
session.commit()
# Prefixed roots
for i, row in enumerate(util.read_csv(ctx.config['PREFIXED_ROOTS'])):
name = row['prefixed_root']
basis = row['unprefixed_root']
hom = row['hom']
prefixes = row['prefixes'].split('-')
assert (basis, hom) in root_map
basis_id = root_map[(basis, hom)]
for prefix in prefixes:
# TODO
pass
prefixed_root = PrefixedRoot(name=name, basis_id=basis_id)
session.add(prefixed_root)
session.flush()
root_map[(name, hom)] = prefixed_root.id
tick(name)
session.commit()
session.close()
return root_map