def test_crud(db):
migration = Connection(DBSession)
assert len(list(migration.select(common.Identifier))) == 0
pk = migration.insert(common.Identifier,
id='iso-csw',
name='csw',
type=common.IdentifierType.iso.value)
assert migration.pk(common.Identifier, 'iso-csw') == pk
assert len(list(migration.select(common.Identifier))) == 1
identifier = DBSession.query(common.Identifier)\
.options(undefer('*')).get(pk)
assert identifier.active
assert identifier.version == 1
assert identifier.created
assert identifier.updated
migration.update(common.Identifier, [('name', 'cea')], pk=pk)
DBSession.refresh(identifier)
assert identifier.name == 'cea'
migration.delete(common.Identifier, pk=pk)
with pytest.raises(InvalidRequestError):
DBSession.refresh(identifier)
def test_crud(self):
from clld.db.migration import Connection
migration = Connection(DBSession)
assert len(list(migration.select(common.Identifier))) == 0
pk = migration.insert(common.Identifier,
id='iso-csw',
name='csw',
type=common.IdentifierType.iso.value)
assert migration.pk(common.Identifier, 'iso-csw') == pk
assert len(list(migration.select(common.Identifier))) == 1
identifier = DBSession.query(common.Identifier).get(pk)
assert identifier.active
assert identifier.version == 1
assert identifier.created
assert identifier.updated
migration.update(common.Identifier, [('name', 'cea')], pk=pk)
DBSession.refresh(identifier)
assert identifier.name == 'cea'
migration.delete(common.Identifier, pk=pk)
self.assertRaises(InvalidRequestError, DBSession.refresh, identifier)
def upgrade():
conn = Connection(op.get_bind())
example_map = {}
sid = 204
for example in jsonload(data_file('lingala_examples.json')):
sid += 1
kw = {
'id': '60-%s' % sid,
'language_pk': conn.pk(Language, '60'),
'name': example['Text'],
'description': example['Translation'],
'gloss': '\t'.join(example['Gloss'].split()),
'analyzed': '\t'.join(example['Text'].split()),
'type': example['Type'].strip().lower(),
'jsondata': {'sort': int(example['Order_number']), 'alt_translation': None}
}
example_map[example['Example_number']] = conn.insert(Sentence, **kw)
for ve in jsonload(data_file('lingala_value_examples.json')):
vspk = conn.pk(ValueSet, '60-%s' % ve['Features::Feature_number'])
vpk = conn.pk(Value, vspk, attr='valueset_pk')
conn.insert(
ValueSentence, value_pk=vpk, sentence_pk=example_map[ve['Example_number']])
for i, comment in enumerate(reader(data_file('lingala_valueset_comments.tab'), delimiter='\t', dicts=True)):
vspk = conn.pk(ValueSet, '60-%s' % comment['Features::Feature_number'])
comment['Comments_on_value_assignment'] = comment['Comments_on_value_assignment'].replace('\x0b', '\n')
conn.update(
ValueSet,
{
'description': comment['Comments_on_value_assignment'],
'markup_description': None,
},
pk=vspk)
def upgrade():
csv = path(phoible.__file__).dirname().joinpath("..", "data", "InventoryID-InternetArchive.csv")
ia_urls = {row[0]: row[1] for row in reader(csv) if row[1] != "NA"}
conn = Connection(op.get_bind())
for id_, url in ia_urls.items():
pk = conn.pk(Contribution, id_)
conn.update(Inventory, dict(internetarchive_url=url), pk=pk)
def upgrade():
csv = Path(phoible.__file__).parent.joinpath(
'..', 'data', 'InventoryID-InternetArchive.csv')
ia_urls = {row[0]: row[1] for row in reader(csv) if row[1] != 'NA'}
conn = Connection(op.get_bind())
for id_, url in ia_urls.items():
pk = conn.pk(Contribution, id_)
conn.update(Inventory, dict(internetarchive_url=url), pk=pk)
def test_crud(self):
from clld.db.migration import Connection
migration = Connection(DBSession)
assert len(list(migration.select(common.Identifier))) == 0
pk = migration.insert(
common.Identifier,
id='iso-csw', name='csw', type=common.IdentifierType.iso.value)
assert migration.pk(common.Identifier, 'iso-csw') == pk
assert len(list(migration.select(common.Identifier))) == 1
identifier = DBSession.query(common.Identifier)\
.options(undefer('*')).get(pk)
assert identifier.active
assert identifier.version == 1
assert identifier.created
assert identifier.updated
migration.update(common.Identifier, [('name', 'cea')], pk=pk)
DBSession.refresh(identifier)
assert identifier.name == 'cea'
migration.delete(common.Identifier, pk=pk)
self.assertRaises(InvalidRequestError, DBSession.refresh, identifier)
def upgrade():
conn = Connection(op.get_bind())
lpk = conn.pk(Language, 'jiar1239')
for lid in list(conn.select(LanguageIdentifier, language_pk=lpk)):
id_ = list(conn.select(Identifier, pk=lid.identifier_pk))[0]
if id_.type == 'iso639-3':
conn.delete(LanguageIdentifier, pk=lid.pk)
conn.delete(Identifier, pk=id_.pk)
def upgrade():
conn = Connection(op.get_bind())
for hid, countries in REMOVE.items():
l = conn.pk(Languoid, hid, attr='hid')
for country in set(countries):
c = conn.pk(Country, country)
assert conn.first(Languoidcountry, country_pk=c, languoid_pk=l)
conn.delete(Languoidcountry, country_pk=c, languoid_pk=l)
for hid, countries in ADD.items():
l = conn.pk(Languoid, hid, attr='hid')
for c in set(countries):
c = conn.pk(Country, c)
conn.insert(Languoidcountry, country_pk=c, languoid_pk=l)
def upgrade():
conn = Connection(op.get_bind())
spk = conn.pk()
conn.insert()
source = table(
'source', column('id', String), column('url', Unicode), column('note', Unicode))
op.execute(
source.update()
.where(source.c.id == op.inline_literal('heathetal2015'))
.values({
'url': op.inline_literal(
'http://cdstar.shh.mpg.de/bitstreams/EAEA0-C97A-A1D2-2E76-0/a.xls'),
'note': op.inline_literal('')})
)
def test_set_glottocode(db):
c = Connection(DBSession)
lpk = c.insert(common.Language, id='l', name='Language')
c.set_glottocode('l', 'abcd1234')
c.set_glottocode('l', 'abcd1234')
l = DBSession.query(common.Language).get(lpk)
assert l.glottocode == 'abcd1234'
c.set_glottocode('l', 'dcba1234')
DBSession.expire_all()
l = DBSession.query(common.Language).get(lpk)
assert l.glottocode == 'dcba1234'
c.set_glottocode('l', 'abcd1234')
def upgrade():
conn = Connection(op.get_bind())
spk = conn.pk(Source, 'heathetal2015')
conn.insert(
Source_files,
jsondata={
"thumbnail": None,
"web": None,
"size": 7531008,
"objid": "EAEA0-C97A-A1D2-2E76-0",
"original": "a.xls"},
id='heathetal2015-1',
name='Dogon.comp.vocab.UNICODE.xls',
ord=1,
mime_type='application/vnd.ms-excel',
object_pk=spk)
def upgrade():
"""Bajau (Semporna) should be renamed Sama (Southern)."""
conn = Connection(op.get_bind())
bjs = conn.pk(Language, 'bjs')
conn.update(Language, dict(name='Sama (Southern)'), pk=bjs)
conn.update(WalsLanguage, dict(ascii_name='samasouthern'), pk=bjs)
def test_set_glottocode(self):
from clld.db.migration import Connection
c = Connection(DBSession)
lpk = c.insert(common.Language, id='l', name='Language')
c.set_glottocode('l', 'abcd1234')
c.set_glottocode('l', 'abcd1234')
l = DBSession.query(common.Language).get(lpk)
assert l.glottocode == 'abcd1234'
c.set_glottocode('l', 'dcba1234')
DBSession.expire_all()
l = DBSession.query(common.Language).get(lpk)
assert l.glottocode == 'dcba1234'
c.set_glottocode('l', 'abcd1234')
def upgrade():
conn = Connection(op.get_bind())
pk = conn.insert(Contributor, id='gt', name='Güldemann, Tom', jsondata={},
polymorphic_type='custom',
address='Humboldt University of Berlin')
pke = conn.insert(Editor, jsondata={}, polymorphic_type='custom',
dataset_pk=1, contributor_pk=pk)
conn.insert(TsammalexEditor, pk=pke)
conn.insert(TsammalexContributor, pk=pk, sections='', research_project='')
def test_set_glottocode(db):
c = Connection(DBSession)
lpk = c.insert(common.Language, id='l', name='Language')
c.set_glottocode('l', 'abcd1234')
c.set_glottocode('l', 'abcd1234')
l = DBSession.query(common.Language).get(lpk)
assert l.glottocode == 'abcd1234'
c.set_glottocode('l', 'dcba1234')
DBSession.expire_all()
l = DBSession.query(common.Language).get(lpk)
assert l.glottocode == 'dcba1234'
c.set_glottocode('l', 'abcd1234')
def upgrade():
conn = Connection(op.get_bind())
for fid, lid, sid in VALUESETREFERENCES:
vs = conn.pk(ValueSet, '%s-%s' % (fid, lid))
s = conn.pk(Source, sid)
conn.update(ValueSetReference,
dict(source_pk=s),
valueset_pk=vs,
source_pk=None)
assert not list(conn.select(ValueSetReference, source_pk=None))
def upgrade():
conn = Connection(op.get_bind())
for fid, lid, sid in VALUESETREFERENCES:
vs = conn.pk(ValueSet, '%s-%s' % (fid, lid))
s = conn.pk(Source, sid)
conn.update(ValueSetReference, dict(source_pk=s), valueset_pk=vs, source_pk=None)
assert not list(conn.select(ValueSetReference, source_pk=None))
def test_set_glottocode(self):
from clld.db.migration import Connection
c = Connection(DBSession)
lpk = c.insert(common.Language, id='l', name='Language')
c.set_glottocode('l', 'abcd1234')
c.set_glottocode('l', 'abcd1234')
l = DBSession.query(common.Language).get(lpk)
assert l.glottocode == 'abcd1234'
c.set_glottocode('l', 'dcba1234')
DBSession.expire_all()
l = DBSession.query(common.Language).get(lpk)
assert l.glottocode == 'dcba1234'
c.set_glottocode('l', 'abcd1234')
def upgrade():
conn = Connection(op.get_bind())
pk = conn.pk(Language, 'cea')
for li in conn.all(LanguageIdentifier, language_pk=pk):
id_ = conn.first(Identifier, pk=li.identifier_pk)
if (id_.name, id_.type) in [
('East Cree', 'name'),
('crj', 'iso639-3'),
('Cree, Southern East', 'name'),
]:
conn.delete(LanguageIdentifier, pk=li.pk)
def upgrade():
"""
Raji should be taken out
of the Bodic genus and placed in a new genus Raji-Raute, but still part
of the Tibeto-Burman subfamily of the Sino-Tibetan family.
"""
conn = Connection(op.get_bind())
pk = conn.pk(Family, 'sinotibetan')
dizoid = conn.insert(Genus,
id='rajiraute',
name='Raji-Raute',
family_pk=pk,
subfamily='Tibeto-Burman',
icon='tffff00')
pk = conn.pk(Language, 'rji')
conn.update(WalsLanguage, [('genus_pk', dizoid)], pk=pk)
def upgrade():
"""
Raji should be taken out
of the Bodic genus and placed in a new genus Raji-Raute, but still part
of the Tibeto-Burman subfamily of the Sino-Tibetan family.
"""
conn = Connection(op.get_bind())
pk = conn.pk(Family, 'sinotibetan')
dizoid = conn.insert(
Genus,
id='rajiraute',
name='Raji-Raute',
family_pk=pk,
subfamily='Tibeto-Burman',
icon='tffff00')
pk = conn.pk(Language, 'rji')
conn.update(WalsLanguage, [('genus_pk', dizoid)], pk=pk)
def upgrade():
conn = Connection(op.get_bind())
for hid, countries in REMOVE.items():
l = conn.pk(Languoid, hid, attr='hid')
for country in set(countries):
c = conn.pk(Country, country)
assert conn.first(Languoidcountry, country_pk=c, languoid_pk=l)
conn.delete(Languoidcountry, country_pk=c, languoid_pk=l)
for hid, countries in ADD.items():
l = conn.pk(Languoid, hid, attr='hid')
for c in set(countries):
c = conn.pk(Country, c)
conn.insert(Languoidcountry, country_pk=c, languoid_pk=l)
def upgrade():
conn = Connection(op.get_bind())
# We need to change the name of Kaliai-Kove (kkv) to Lusi (which is the Glottolog name).
pk = conn.pk(Language, 'kkv')
conn.update(Language, [('name', 'Lusi')], pk=pk)
conn.update(WalsLanguage, [('ascii_name', 'lusi'), ('iso_codes', 'khl')], pk=pk)
#wals3=# select li.pk, i.name, i.type, i.description from languageidentifier as li, identifier as i where li.language_pk = 2163 and li.identifier_pk = i.pk;
# pk | name | type | description
#-------+--------------+-----------+-------------
# 1302 | Kove | name | ruhlen
# 2749 | Kove-Kaliai | name | routledge
# 4202 | Kaliai | name | other
# 4406 | Kandoka-Lusi | name | other
# 4901 | Kaliai Kove | name | other
# 7797 | khl | iso639-3 | Lusi
# 7798 | kvc | iso639-3 | Kove
# 10488 | Lusi | name | ethnologue
# 10489 | Kove | name | ethnologue
# 13658 | kali1298 | glottolog |
conn.update(Identifier, [('name', 'lusi1240')], name='kali1298')
remove = []
for row in conn.execute("""
select li.pk, i.name from languageidentifier as li, identifier as i
where li.language_pk = %s and li.identifier_pk = i.pk""" % pk):
if 'Kove' in row['name'] or row['name'] == 'kvc':
remove.append(row['pk'])
for pk in remove:
conn.delete(LanguageIdentifier, pk=pk)
# The language Dizi (wals code diz) needs to be removed from North Omotic and placed in a
# separate (new) genus called Dizoid, within the Omotic subfamily of Afro-Asiatic.
# (Glottolog actually treats Dizoid as a separate family altogether, which is good reason
# to conclude that it is at least a separate genus.)
pk = conn.pk(Family, 'afroasiatic')
dizoid = conn.insert(
Genus,
id='dizoid',
name='Dizoid',
family_pk=pk,
subfamily='Omotic',
icon='tdd0000')
pk = conn.pk(Language, 'diz')
conn.update(WalsLanguage, [('genus_pk', dizoid)], pk=pk)
# The name for what is now Siraiya (sry) should be changed to Siraya. (Siraya is also the Glottolog name)
pk = conn.pk(Language, 'sry')
conn.update(Language, [('name', 'Siraya')], pk=pk)
conn.update(WalsLanguage, [('ascii_name', 'siraya')], pk=pk)
# I want to modify the assignment of languages to genera for the Austronesian languages of Taiwan.
# (This change brings WALS in line with Glottolog.) Namely, I want to split up the current Paiwanic genus as follows:
#
# - Amis and Siraya should be moved into a new genus called East Formosan.
# - Thao should be moved into a new genus called Western Plains Austronesian.
#
paiwanic = conn.first(Genus, id='paiwanic')
eastformosan = conn.insert(
Genus,
id='eastformosan',
name='East Formosan',
family_pk=paiwanic['family_pk'],
icon='fdd0000')
westernplainsaustronesian = conn.insert(
Genus,
id='westernplainsaustronesian',
name='Western Plains Austronesian',
family_pk=paiwanic['family_pk'],
icon='f0000dd')
for lid in ['ami', 'sry']:
pk = conn.pk(Language, lid)
conn.update(WalsLanguage, [('genus_pk', eastformosan)], pk=pk)
pk = conn.pk(Language, 'thw')
conn.update(WalsLanguage, [('genus_pk', westernplainsaustronesian)], pk=pk)
# The sole remaining language in Paiwanic is Paiwan, but since this genus now contains only one language,
# we should rename the genus Paiwan.
conn.update(Genus, [('name', 'Paiwan')], id='paiwanic')
# For similar reasons, the genus Tsouic should just be called Tsou.
conn.update(Genus, [('name', 'Tsou')], id='tsouic')
# The genus currently called Southern Atlantic should be renamed Mel. (This brings it in line with Glottolog.)
conn.update(Genus, [('name', 'Mel')], id='southernatlantic')
for id in ['Ashton-1947', 'Ashton-1969']:
pk = conn.pk(Source, id)
conn.update(Source, [('author', 'Ethel O. Ashton')], pk=pk)
pk = conn.pk(Source, 'Poldaufem-1971')
conn.update(
Source,
[
('name', 'Poldauf 1971'),
('author', 'Poldauf, Ivan'),
('description', 'Anglicko-český a česko-anglický slovník'),
('title', 'Anglicko-český a česko-anglický slovník'),
],
pk=pk)
#wals3=# select author, name, pages, title, description from source where id = 'De-Lancey-2003';
# author | name | pages | title | description
#------------------+----------------+---------+-------------------+-------------------
# De Lancey, Scott | De Lancey 2003 | 255-269 | Classical Tibetan | Classical Tibetan
#(1 row)
#-> DeLancey 270-288 Lhasa Tibetan
pk = conn.pk(Source, 'De-Lancey-2003')
conn.update(
Source,
[
('name', 'DeLancey 2003'),
('author', 'DeLancey, Scott'),
('description', 'Lhasa Tibetan'),
('title', 'Lhasa Tibetan'),
],
pk=pk)
pk = conn.pk(Source, 'Frajzyngier-and-Shay-2002')
conn.update(Source, [('author', 'Frajzyngier, Zygmunt with Erin Shay')], pk=pk)
# wals-data #9
conn.execute("update domainelement set name = 'No voicing contrast' where name = 'No voicing constrast'")
# wals-data #7
pk = conn.pk(Source, 'Brown-1990')
conn.update(
Source,
[
('description', 'Mai Brat Nominal Phrases'),
('title', 'Mai Brat Nominal Phrases'),
('booktitle', 'Miscellaneous Studies of Indonesian and Other Languages of Indonesia, Part X'),
],
pk=pk)
pk = conn.pk(Source, 'Saltarelli-et-al-1988')
conn.update(
Source,
[('author',
'Saltarelli, Mario with Miren Azkarate, and Farwell, David and de Urbina, Jon Ortiz and Oñederra, Lourdes')],
pk=pk)
#wals3=# select xhtml from sentence where id = '478';
#replace:
#'anywhere' -> 'whenever'
#'where' -> 'when'
res = conn.first(Sentence, id='478')
xhtml = res['xhtml']
for s, r in [("'anywhere'", "'whenever'"), ("'where'", "'when'")]:
assert s in xhtml
xhtml = xhtml.replace(s, r)
conn.update(Sentence, [('xhtml', xhtml)], id='478')
def upgrade():
conn = Connection(op.get_bind())
conn.update(Source, [('pages', '424-438')], id='Frantz-1973')
def upgrade():
conn = Connection(op.get_bind())
# We need to change the name of Kaliai-Kove (kkv) to Lusi (which is the Glottolog name).
pk = conn.pk(Language, 'kkv')
conn.update(Language, [('name', 'Lusi')], pk=pk)
conn.update(WalsLanguage, [('ascii_name', 'lusi'), ('iso_codes', 'khl')],
pk=pk)
#wals3=# select li.pk, i.name, i.type, i.description from languageidentifier as li, identifier as i where li.language_pk = 2163 and li.identifier_pk = i.pk;
# pk | name | type | description
#-------+--------------+-----------+-------------
# 1302 | Kove | name | ruhlen
# 2749 | Kove-Kaliai | name | routledge
# 4202 | Kaliai | name | other
# 4406 | Kandoka-Lusi | name | other
# 4901 | Kaliai Kove | name | other
# 7797 | khl | iso639-3 | Lusi
# 7798 | kvc | iso639-3 | Kove
# 10488 | Lusi | name | ethnologue
# 10489 | Kove | name | ethnologue
# 13658 | kali1298 | glottolog |
conn.update(Identifier, [('name', 'lusi1240')], name='kali1298')
remove = []
for row in conn.execute("""
select li.pk, i.name from languageidentifier as li, identifier as i
where li.language_pk = %s and li.identifier_pk = i.pk""" % pk):
if 'Kove' in row['name'] or row['name'] == 'kvc':
remove.append(row['pk'])
for pk in remove:
conn.delete(LanguageIdentifier, pk=pk)
# The language Dizi (wals code diz) needs to be removed from North Omotic and placed in a
# separate (new) genus called Dizoid, within the Omotic subfamily of Afro-Asiatic.
# (Glottolog actually treats Dizoid as a separate family altogether, which is good reason
# to conclude that it is at least a separate genus.)
pk = conn.pk(Family, 'afroasiatic')
dizoid = conn.insert(Genus,
id='dizoid',
name='Dizoid',
family_pk=pk,
subfamily='Omotic',
icon='tdd0000')
pk = conn.pk(Language, 'diz')
conn.update(WalsLanguage, [('genus_pk', dizoid)], pk=pk)
# The name for what is now Siraiya (sry) should be changed to Siraya. (Siraya is also the Glottolog name)
pk = conn.pk(Language, 'sry')
conn.update(Language, [('name', 'Siraya')], pk=pk)
conn.update(WalsLanguage, [('ascii_name', 'siraya')], pk=pk)
# I want to modify the assignment of languages to genera for the Austronesian languages of Taiwan.
# (This change brings WALS in line with Glottolog.) Namely, I want to split up the current Paiwanic genus as follows:
#
# - Amis and Siraya should be moved into a new genus called East Formosan.
# - Thao should be moved into a new genus called Western Plains Austronesian.
#
paiwanic = conn.first(Genus, id='paiwanic')
eastformosan = conn.insert(Genus,
id='eastformosan',
name='East Formosan',
family_pk=paiwanic['family_pk'],
icon='fdd0000')
westernplainsaustronesian = conn.insert(Genus,
id='westernplainsaustronesian',
name='Western Plains Austronesian',
family_pk=paiwanic['family_pk'],
icon='f0000dd')
for lid in ['ami', 'sry']:
pk = conn.pk(Language, lid)
conn.update(WalsLanguage, [('genus_pk', eastformosan)], pk=pk)
pk = conn.pk(Language, 'thw')
conn.update(WalsLanguage, [('genus_pk', westernplainsaustronesian)], pk=pk)
# The sole remaining language in Paiwanic is Paiwan, but since this genus now contains only one language,
# we should rename the genus Paiwan.
conn.update(Genus, [('name', 'Paiwan')], id='paiwanic')
# For similar reasons, the genus Tsouic should just be called Tsou.
conn.update(Genus, [('name', 'Tsou')], id='tsouic')
# The genus currently called Southern Atlantic should be renamed Mel. (This brings it in line with Glottolog.)
conn.update(Genus, [('name', 'Mel')], id='southernatlantic')
for id in ['Ashton-1947', 'Ashton-1969']:
pk = conn.pk(Source, id)
conn.update(Source, [('author', 'Ethel O. Ashton')], pk=pk)
pk = conn.pk(Source, 'Poldaufem-1971')
conn.update(Source, [
('name', 'Poldauf 1971'),
('author', 'Poldauf, Ivan'),
('description', 'Anglicko-český a česko-anglický slovník'),
('title', 'Anglicko-český a česko-anglický slovník'),
],
pk=pk)
#wals3=# select author, name, pages, title, description from source where id = 'De-Lancey-2003';
# author | name | pages | title | description
#------------------+----------------+---------+-------------------+-------------------
# De Lancey, Scott | De Lancey 2003 | 255-269 | Classical Tibetan | Classical Tibetan
#(1 row)
#-> DeLancey 270-288 Lhasa Tibetan
pk = conn.pk(Source, 'De-Lancey-2003')
conn.update(Source, [
('name', 'DeLancey 2003'),
('author', 'DeLancey, Scott'),
('description', 'Lhasa Tibetan'),
('title', 'Lhasa Tibetan'),
],
pk=pk)
pk = conn.pk(Source, 'Frajzyngier-and-Shay-2002')
conn.update(Source, [('author', 'Frajzyngier, Zygmunt with Erin Shay')],
pk=pk)
# wals-data #9
conn.execute(
"update domainelement set name = 'No voicing contrast' where name = 'No voicing constrast'"
)
# wals-data #7
pk = conn.pk(Source, 'Brown-1990')
conn.update(Source, [
('description', 'Mai Brat Nominal Phrases'),
('title', 'Mai Brat Nominal Phrases'),
('booktitle',
'Miscellaneous Studies of Indonesian and Other Languages of Indonesia, Part X'
),
],
pk=pk)
pk = conn.pk(Source, 'Saltarelli-et-al-1988')
conn.update(Source, [(
'author',
'Saltarelli, Mario with Miren Azkarate, and Farwell, David and de Urbina, Jon Ortiz and Oñederra, Lourdes'
)],
pk=pk)
#wals3=# select xhtml from sentence where id = '478';
#replace:
#'anywhere' -> 'whenever'
#'where' -> 'when'
res = conn.first(Sentence, id='478')
xhtml = res['xhtml']
for s, r in [("'anywhere'", "'whenever'"), ("'where'", "'when'")]:
assert s in xhtml
xhtml = xhtml.replace(s, r)
conn.update(Sentence, [('xhtml', xhtml)], id='478')
def upgrade():
conn = Connection(op.get_bind())
for lid, glottocode in GLOTTOCODES.items():
conn.set_glottocode(lid, glottocode)
def upgrade():
conn = Connection(op.get_bind())
# The language Nishi, with WALS code nis should be renamed Nyishi.
pk = conn.pk(Language, 'nis')
conn.update(Language, [('name', 'Nyishi')], pk=pk)
conn.update(WalsLanguage, [('ascii_name', 'nyishi')], pk=pk)
# Munzumbo [mnz] should be Munzombo
pk = conn.pk(Language, 'mnz')
conn.update(Language, [('name', 'Munzombo')], pk=pk)
conn.update(WalsLanguage, [('ascii_name', 'munzombo')], pk=pk)
def upgrade():
conn = Connection(op.get_bind())
# Khoisan needs to be replaced by four families:
# Sandawe (no name change)
# Kxa: two genera: =|Hoan, Ju-Kung (a name change from 'Northern Khoisan')
# Tu (a name change from Southern Khoisan)
# Khoe-Kwadi (a name change from Central Khoisan)
genera = {
'hoan': ('Kxa', None),
'sandawe': ('Sandawe', None),
'centralkhoisan': ('Khoe-Kwadi', 'Khoe-Kwadi'),
'southernkhoisan': ('Tu', 'Tu'),
'northernkhoisan': ('Kxa', 'Ju-Kung'),
}
families = {}
for gid, (fname, gname) in genera.items():
fid = slug(fname)
fpk = families.get(fid)
if fpk is None:
families[fid] = fpk = conn.insert(Family, id=fid, name=fname)
values = dict(family_pk=fpk)
if gname:
values['name'] = gname
conn.update(Genus, values, id=gid)
conn.insert(Config,
key=Config.replacement_key(Family, 'khoisan'),
value=Config.gone)
conn.delete(Family, id='khoisan')
# Nilo-Saharan needs to be removed and replaced by the following families, with
# constituent genera.
# The following genera will now be families containing a single genus with the
# same name:
# Berta Kunama Songhay Gumuz Koman Shabo Kuliak Maban Fur
for gname in 'Berta Kunama Songhay Gumuz Koman Shabo Kuliak Maban Fur'.split(
):
gid = gname.lower()
fpk = conn.insert(Family, id=gid, name=gname)
gpk = conn.pk(Genus, gid)
conn.update(Genus, dict(family_pk=fpk), pk=gpk)
# The following is one family with two genera:
# Saharan: Eastern Saharan, Western Saharan
fpk = conn.insert(Family, id='saharan', name='Saharan')
for gname in ['Eastern Saharan', 'Western Saharan']:
conn.update(Genus,
dict(family_pk=fpk),
pk=conn.pk(Genus, gname, attr='name'))
# What is currently the subfamily Central Sudanic becomes a family with the same name
# and same constituent genera
# and what is currently the subfamily Eastern Sudanic becomes a family with the same
# name and the same constituent genera.
for sfn in ['Central Sudanic', 'Eastern Sudanic']:
fpk = conn.insert(Family, id=slug(sfn), name=sfn)
conn.update(Genus, dict(family_pk=fpk, subfamily=None), subfamily=sfn)
conn.insert(Config,
key=Config.replacement_key(Family, 'nilosaharan'),
value=Config.gone)
conn.delete(Family, id='nilosaharan')
# The following groups need to be removed from Niger-Congo and set up as separate
# families. The last two (Dogon and Ijoid) are families containing one genus.
# - Mande (currently a subfamily) with the following two genera:
# Eastern Mande, Western Mande
# - Kordofanian (currently a subfamily) with the following five genera:
# Katla-Tima, Rashad, Heiban, Talodi (a name change from Talodi Proper),
# and Lafofa (renamed from Tegem)
# - Dogon
# - Ijoid
for sfn in ['Mande', 'Kordofanian']:
fpk = conn.insert(Family, id=slug(sfn), name=sfn)
conn.update(Genus, dict(family_pk=fpk, subfamily=None), subfamily=sfn)
for old, new in {'Talodi Proper': 'Talodi', 'Tegem': 'Lafofa'}.items():
conn.update(Genus, dict(name=new), name=old)
for sfn in ['Dogon', 'Ijoid']:
fpk = conn.insert(Family, id=slug(sfn), name=sfn)
conn.update(Genus, dict(family_pk=fpk), name=sfn)
# A minor change in the name of the family Kadugli to just Kadu.
conn.update(Family, dict(name='Kadu'), name='Kadugli')
def upgrade():
conn = Connection(op.get_bind())
# Khoisan needs to be replaced by four families:
# Sandawe (no name change)
# Kxa: two genera: =|Hoan, Ju-Kung (a name change from 'Northern Khoisan')
# Tu (a name change from Southern Khoisan)
# Khoe-Kwadi (a name change from Central Khoisan)
genera = {
'hoan': ('Kxa', None),
'sandawe': ('Sandawe', None),
'centralkhoisan': ('Khoe-Kwadi', 'Khoe-Kwadi'),
'southernkhoisan': ('Tu', 'Tu'),
'northernkhoisan': ('Kxa', 'Ju-Kung'),
}
families = {}
for gid, (fname, gname) in genera.items():
fid = slug(fname)
fpk = families.get(fid)
if fpk is None:
families[fid] = fpk = conn.insert(Family, id=fid, name=fname)
values = dict(family_pk=fpk)
if gname:
values['name'] = gname
conn.update(Genus, values, id=gid)
conn.insert(Config, key=Config.replacement_key(Family, 'khoisan'), value=Config.gone)
conn.delete(Family, id='khoisan')
# Nilo-Saharan needs to be removed and replaced by the following families, with
# constituent genera.
# The following genera will now be families containing a single genus with the
# same name:
# Berta Kunama Songhay Gumuz Koman Shabo Kuliak Maban Fur
for gname in 'Berta Kunama Songhay Gumuz Koman Shabo Kuliak Maban Fur'.split():
gid = gname.lower()
fpk = conn.insert(Family, id=gid, name=gname)
gpk = conn.pk(Genus, gid)
conn.update(Genus, dict(family_pk=fpk), pk=gpk)
# The following is one family with two genera:
# Saharan: Eastern Saharan, Western Saharan
fpk = conn.insert(Family, id='saharan', name='Saharan')
for gname in ['Eastern Saharan', 'Western Saharan']:
conn.update(Genus, dict(family_pk=fpk), pk=conn.pk(Genus, gname, attr='name'))
# What is currently the subfamily Central Sudanic becomes a family with the same name
# and same constituent genera
# and what is currently the subfamily Eastern Sudanic becomes a family with the same
# name and the same constituent genera.
for sfn in ['Central Sudanic', 'Eastern Sudanic']:
fpk = conn.insert(Family, id=slug(sfn), name=sfn)
conn.update(Genus, dict(family_pk=fpk, subfamily=None), subfamily=sfn)
conn.insert(Config, key=Config.replacement_key(Family, 'nilosaharan'), value=Config.gone)
conn.delete(Family, id='nilosaharan')
# The following groups need to be removed from Niger-Congo and set up as separate
# families. The last two (Dogon and Ijoid) are families containing one genus.
# - Mande (currently a subfamily) with the following two genera:
# Eastern Mande, Western Mande
# - Kordofanian (currently a subfamily) with the following five genera:
# Katla-Tima, Rashad, Heiban, Talodi (a name change from Talodi Proper),
# and Lafofa (renamed from Tegem)
# - Dogon
# - Ijoid
for sfn in ['Mande', 'Kordofanian']:
fpk = conn.insert(Family, id=slug(sfn), name=sfn)
conn.update(Genus, dict(family_pk=fpk, subfamily=None), subfamily=sfn)
for old, new in {'Talodi Proper': 'Talodi', 'Tegem': 'Lafofa'}.items():
conn.update(Genus, dict(name=new), name=old)
for sfn in ['Dogon', 'Ijoid']:
fpk = conn.insert(Family, id=slug(sfn), name=sfn)
conn.update(Genus, dict(family_pk=fpk), name=sfn)
# A minor change in the name of the family Kadugli to just Kadu.
conn.update(Family, dict(name='Kadu'), name='Kadugli')
def upgrade():
conn = Connection(op.get_bind())
conn.set_glottocode('19', 'yaku1245') # Sakha
conn.set_glottocode('11', 'oldh1241') # Old High German
conn.set_glottocode('9', 'west2376') # Selice Romani
conn.set_glottocode('17', 'mana1288') # Manange
def upgrade():
conn = Connection(op.get_bind())
example_map = {}
sid = 204
for example in jsonload(data_file('lingala_examples.json')):
sid += 1
kw = {
'id': '60-%s' % sid,
'language_pk': conn.pk(Language, '60'),
'name': example['Text'],
'description': example['Translation'],
'gloss': '\t'.join(example['Gloss'].split()),
'analyzed': '\t'.join(example['Text'].split()),
'type': example['Type'].strip().lower(),
'jsondata': {
'sort': int(example['Order_number']),
'alt_translation': None
}
}
example_map[example['Example_number']] = conn.insert(Sentence, **kw)
for ve in jsonload(data_file('lingala_value_examples.json')):
vspk = conn.pk(ValueSet, '60-%s' % ve['Features::Feature_number'])
vpk = conn.pk(Value, vspk, attr='valueset_pk')
conn.insert(ValueSentence,
value_pk=vpk,
sentence_pk=example_map[ve['Example_number']])
for i, comment in enumerate(
reader(data_file('lingala_valueset_comments.tab'),
delimiter='\t',
dicts=True)):
vspk = conn.pk(ValueSet, '60-%s' % comment['Features::Feature_number'])
comment['Comments_on_value_assignment'] = comment[
'Comments_on_value_assignment'].replace('\x0b', '\n')
conn.update(ValueSet, {
'description': comment['Comments_on_value_assignment'],
'markup_description': None,
},
pk=vspk)
def upgrade():
conn = Connection(op.get_bind())
for lid, glottocode in GLOTTOCODES.items():
conn.set_glottocode(lid, glottocode)
