def load(id_, data):
d = Dictionary(path(__file__).dirname().joinpath('yalalag.bak'), encoding='latin1')
d.entries = filter(lambda r: r.get('lx') and r.get('ge'), d.entries)
lang = data.add(common.Language, id_, id=id_, name='Yalálag Zapotec',
latitude=17.18574, longitude=-96.17891)
#
# TODO:
#
# iso code zpu
contrib = data.add(
common.Contributor, 'heriberto', id='avelinoheriberto', name='Heriberto Avelino')
DBSession.flush()
number = max([int(dd.id) for dd in data['Dictionary'].values()] + [0])
vocab = data.add(
models.Dictionary, id_, id=str(number + 1),
name='Yalálag Zapotec Dictionary',
language=lang,
published=date(2013, 5, 5))
DBSession.flush()
DBSession.add(common.Contribution_files(
object_pk=vocab.pk,
name='description',
file=common.File(
name='yalalag.pdf',
mime_type='application/pdf',
content=open(path(__file__).dirname().joinpath('yalalag.pdf')).read())))
old_db = create_engine(DB)
for row in old_db.execute("select * from meaning"):
data.add(
common.ValueSet, '%s-%s' % (row['label'].lower(), id_),
id='%s-%s' % (id_, row['id'].replace('.', '-')),
language=lang,
contribution=vocab,
parameter=data['Meaning'][row['id']])
DBSession.flush()
ue = common.UnitParameter(id='ue', name='usage')
DBSession.add(ue)
DBSession.flush()
for name in d.values('ue'):
p = data.add(common.UnitDomainElement, name, id='ue-'+slug(name), name=name)
p.unitparameter_pk = ue.pk
for i, row in enumerate(d.entries):
w = data.add(
models.Word, row.get('lx'),
id='%s-%s' % (id_, i),
name=row.get('lx'),
description='; '.join(row.getall('ge')),
dictionary=vocab)
w.language = lang
DBSession.flush()
for marker in [
'bw', 'ce', 'cf', 'de', 'dn', 'et', 'gv', 'lc', 'mr', 'nt',
'ph', 're', 'rn', 'sc', 'se', 'un', 'va',
]:
for k, name in enumerate(row.getall(marker)):
DBSession.add(
common.Unit_data(key=marker, value=name, ord=k, object_pk=w.pk))
for j, name in enumerate(row.getall('ue')):
DBSession.add(common.UnitValue(
id='ue-%s-%s' % (i, j),
unit=w,
unitparameter=ue,
unitdomainelement=data['UnitDomainElement'][name],
contribution=vocab,
))
meaning_prefix = ''
for j, name in enumerate(row.getall('ps')):
if POS_MAP[name] == 'verb' or ' verb' in POS_MAP[name]:
meaning_prefix = 'to '
elif POS_MAP[name] == 'noun':
meaning_prefix = 'the '
if j > 0:
# only one part-of-speech value per entry!
raise ValueError
DBSession.add(common.UnitValue(
id='pos-%s-%s' % (id_, i),
unit=w,
unitparameter=data['UnitParameter']['pos'],
unitdomainelement=data['UnitDomainElement'][POS_MAP[name]],
contribution=vocab,
))
for j, name in enumerate(row.getall('ge')):
if name.startswith(meaning_prefix):
meaning_prefix = ''
key = '%s%s-%s' % (meaning_prefix, name.lower(), id_)
if key in data['ValueSet']:
value = data.add(
models.Counterpart, '%s-%s' % (i, j),
id='%s-%s-%s' % (id_, i, j),
name=row.get('lx'),
valueset=data['ValueSet'][key],
word=w)
if row.get('xv'):
ex = data.add(
common.Sentence, i,
id='%s-%s' % (id_, i),
name=row.get('xv'),
description=row.get('xe', default=''))
DBSession.add(models.WordSentence(word=w, sentence=ex))
DBSession.flush()
DBSession.add(common.ContributionContributor(
ord=1,
primary=True,
contributor_pk=contrib.pk,
contribution_pk=vocab.pk))
DBSession.flush()
def load(id_, data, files_dir, data_dir):
d = Dictionary(data_dir.joinpath("Hoocak_lex_ld100.lex"))
d.entries = filter(lambda r: r.get("lx"), d.entries)
lang = data["Language"][id_]
vocab = data["Dictionary"][id_]
sd = common.UnitParameter(id="sd", name="semantic domain")
DBSession.add(sd)
DBSession.flush()
for name in d.values("sd"):
if name.startswith("??"):
continue
p = data.add(common.UnitDomainElement, name, id="sd-" + slug(name), name=name)
p.unitparameter_pk = sd.pk
DBSession.flush()
for i, row in enumerate(d.entries):
w = data.add(
models.Word,
row.get("lx"),
id="%s-%s" % (id_, i),
name=row.get("lx"),
description="; ".join(row.getall("me")),
dictionary=vocab,
)
w.language = lang
if row.get("hm"):
try:
w.number = int(row.get("hm"))
except:
print "---->", row.get("hm")
DBSession.flush()
for marker, label in [
("al", "alternative form"),
("cf", "conjugated form"),
("cc", "conjugation class"),
("mp", "metaphony"),
("is", "internal structure"),
]:
for k, name in enumerate(row.getall(marker)):
DBSession.add(common.Unit_data(key=label, value=name, ord=k, object_pk=w.pk))
for marker in ["pc", "sf"]:
for l, spec in enumerate(row.getall(marker)):
try:
p, mimetype = spec
except:
p = spec
mimetype = "image/jpeg" if marker == "pc" else "audio/mpeg"
p = data_dir.joinpath(*p.split("\\"))
with open(p, "rb") as fp:
f = common.Unit_files(
name=p.basename(), id=mimetype.split("/")[0], mime_type=mimetype, ord=l + 1, object_pk=w.pk
)
DBSession.add(f)
DBSession.flush()
DBSession.refresh(f)
f.create(files_dir, fp.read())
for j, name in enumerate(row.getall("sd")):
if name.startswith("??"):
continue
DBSession.add(
common.UnitValue(
id="sd-%s-%s" % (i, j),
unit=w,
unitparameter=sd,
unitdomainelement=data["UnitDomainElement"][name],
contribution=vocab,
)
)
meaning_prefix = ""
for j, name in enumerate(row.getall("ps")):
if name.startswith("??"):
continue
if name == "v.inact":
name += "."
if name.startswith("v."):
meaning_prefix = "to "
elif name.startswith("n."):
meaning_prefix = "the "
if j > 0:
# only one part-of-speech value per entry!
raise ValueError
DBSession.add(
common.UnitValue(
id="pos-%s-%s" % (id_, i),
unit=w,
unitparameter=data["UnitParameter"]["pos"],
unitdomainelement=data["UnitDomainElement"][POS_MAP[name]],
contribution=vocab,
)
)
for j, name in enumerate(row.getall("me")):
key = "%s%s" % (meaning_prefix, name.lower())
if key in data["Meaning"]:
meaning = data["Meaning"][key]
vsid = ("%s-%s" % (key, id_),)
if vsid in data["ValueSet"]:
vs = data["ValueSet"][vsid]
else:
vs = data.add(
common.ValueSet,
vsid,
id="%s-%s" % (id_, meaning.id),
language=lang,
contribution=vocab,
parameter=meaning,
)
DBSession.add(models.Counterpart(id="%s-%s-%s" % (id_, i, j), name=row.get("lx"), valueset=vs, word=w))
DBSession.flush()