def _addSource(lp):
"""For a lighter 'main' function."""
DBSession.add(
common.Source(id=lp[0],
name=lp[0],
author=lp[2],
year=lp[3],
title=lp[4],
url=lp[5],
note=lp[6]))
DBSession.flush()
def bibtex2source(rec): # pragma: no cover
year = rec.get('year', 'nd')
fields = {}
jsondata = {}
for field in FIELDS:
if field in rec:
value = unescape(rec[field])
container = fields if hasattr(common.Source, field) else jsondata
container[field] = value
# remove \\ from url fields!
if field == 'url':
container[field] = container[field].replace('\\', '')
etal = ''
eds = ''
authors = rec.get('author')
if not authors:
authors = rec.get('editor', '')
if authors:
eds = ' (eds.)'
if authors:
authors = unescape(authors).split(' and ')
if len(authors) > 2:
authors = authors[:1]
etal = ' et al.'
authors = [HumanName(a) for a in authors]
authors = [n.last or n.first for n in authors]
authors = '%s%s%s' % (' and '.join(authors), etal, eds)
if rec.genre == 'thesis':
if rec['type'] == 'phdthesis':
rec.genre = 'phdthesis'
else:
rec.genre = 'mastersthesis'
try:
bibtex_type = EntryType.from_string(rec.genre)
except:
bibtex_type = EntryType.from_string('misc')
return common.Source(id=rec.id,
name=('%s %s' % (authors, year)).strip(),
description=unescape(
rec.get('title', rec.get('booktitle', ''))),
jsondata=jsondata,
bibtex_type=bibtex_type,
**fields)
def get_source(source, id_):
author, year, description = source
url = None
match = re.search('(?P<url>http(s)?://[^\s]+)(\s+|$)', description)
if match:
url = match.group('url')
res = common.Source(id='%s' % id_,
name='%s %s' % (author or 'n.a.', year or 'n.d.'),
description=description,
author=author,
year=year,
title=description,
url=url,
bibtex_type=EntryType.misc)
DBSession.add(res)
return res
def bibtex2source(rec):
year = bibtex.unescape(rec.get('year', 'nd'))
fields = {}
jsondata = {}
for field in bibtex.FIELDS:
if field in rec:
value = bibtex.unescape(rec[field])
container = fields if hasattr(common.Source, field) else jsondata
container[field] = value
return common.Source(
id=slug(rec.id),
name=('%s %s' % (bibtex.unescape(
rec.get('author', rec.get('editor', ''))), year)).strip(),
description=bibtex.unescape(rec.get('title', rec.get('booktitle',
''))),
jsondata=jsondata,
bibtex_type=rec.genre,
**fields)
def setUp(self):
TestWithDb.setUp(self)
DBSession.add(
common.Dataset(id='dataset',
name='dataset',
description='desc',
domain='clld'))
source = common.Source(id='source')
contributors = {
'contributor': 'A Name',
'b': 'b Name',
'c': 'c Name',
'd': 'd Name'
}
for id_, name in contributors.items():
contributors[id_] = common.Contributor(id=id_, name=name)
contribution = common.Contribution(id='contribution',
name='Contribution')
cr = common.ContributionReference(contribution=contribution,
source=source)
assert common.ContributionContributor(
contribution=contribution,
primary=True,
contributor=contributors['contributor'])
assert common.ContributionContributor(contribution=contribution,
primary=False,
contributor=contributors['b'])
assert common.ContributionContributor(contribution=contribution,
primary=True,
contributor=contributors['c'])
assert common.ContributionContributor(contribution=contribution,
primary=False,
contributor=contributors['d'])
DBSession.add(contribution)
language = common.Language(id='language',
name='Language 1',
latitude=10.5,
longitude=0.3)
language.sources.append(source)
identifier = common.Identifier(type='iso639-3', id='iso')
li = common.LanguageIdentifier(language=language,
identifier=identifier)
for i in range(2, 102):
_l = common.Language(id='l%s' % i, name='Language %s' % i)
_i = common.Identifier(type='iso639-3',
id='%.3i' % i,
name='%.3i' % i)
_li = common.LanguageIdentifier(language=_l, identifier=_i)
DBSession.add(_l)
param = common.Parameter(id='parameter', name='Parameter')
de = common.DomainElement(id='de',
name='DomainElement',
parameter=param)
de2 = common.DomainElement(id='de2',
name='DomainElement2',
parameter=param)
valueset = common.ValueSet(id='valueset',
language=language,
parameter=param,
contribution=contribution)
value = common.Value(id='value',
domainelement=de,
valueset=valueset,
frequency=50,
confidence='high')
DBSession.add(value)
paramnd = common.Parameter(id='no-domain',
name='Parameter without domain')
valueset = common.ValueSet(id='vs2',
language=language,
parameter=paramnd,
contribution=contribution)
vr = common.ValueSetReference(valueset=valueset, source=source)
value = common.Value(id='v2',
valueset=valueset,
frequency=50,
confidence='high')
DBSession.add(value)
unit = common.Unit(id='unit', name='Unit', language=language)
up = common.UnitParameter(id='unitparameter', name='UnitParameter')
DBSession.add(unit)
DBSession.add(
common.UnitValue(id='unitvalue',
name='UnitValue',
unit=unit,
unitparameter=up))
up2 = common.UnitParameter(id='up2', name='UnitParameter with domain')
de = common.UnitDomainElement(id='de', name='de', parameter=up2)
DBSession.add(
common.UnitValue(id='uv2',
name='UnitValue2',
unit=unit,
unitparameter=up2,
unitdomainelement=de))
DBSession.add(common.Source(id='s'))
sentence = common.Sentence(id='sentence',
name='sentence name',
description='sentence description',
analyzed='a\tmorpheme\tdoes\tdo',
gloss='a\tmorpheme\t1SG\tdo.SG2',
source='own',
comment='comment',
original_script='a morpheme',
language=language)
sr = common.SentenceReference(sentence=sentence, source=source)
DBSession.add(common.Config(key='key', value='value'))
DBSession.flush()
def load():
wals = create_engine('postgresql://robert@/wals3')
contributor = common.Contributor(id='gastvolker', name='Volker Gast')
contribution = common.Contribution(
id='tdir', name='Typological Database of Intensifiers and Reflexives')
cc = common.ContributionContributor(
contribution=contribution, contributor=contributor)
DBSession.add(cc)
for row in read('glosses'):
DBSession.add(common.GlossAbbreviation(id=row['gloss'], name=row['explanation']))
params = {}
for id_, name in PARAMS.items():
params[id_] = common.Parameter(id='tdir-' + id_, name=name)
DBSession.add(params[id_])
#
# TODO: domain for sortal restrictions!
#
values = {}
languages = {}
for row in read('languages'):
if row['adn'] and '<br>' in row['adn']:
row['adn'], other = row['adn'].split('<br>', 1)
if not row['otherint']:
row['otherint'] = ''
row['otherint'] = '\n'.join(filter(None, row['otherint'].split('<br>') + other.split('<br>')))
row['sil'] = row['sil'].lower()
row['sil'] = {
'arm': 'hye',
'vmn': 'mig',
'gli': 'gle',
'grk': 'ell',
'hbr': 'heb',
'ltn': 'lat',
'chn': 'cmn',
'ota': 'ote',
'pnj': 'pan',
'pba': 'rap',
'esg': 'kal',
'vla': 'zea',
'lat': 'lav',
}.get(row['sil'], row['sil'])
l = common.Language(id=row['sil'].lower(), name=row['language'])
languages[row['language']] = l
res = wals.execute("select l.latitude, l.longitude from language as l, languageidentifier as li, identifier as i where l.pk = li.language_pk and li.identifier_pk = i.pk and i.id = '%s' and i.type = 'iso639-3';" \
% row['sil']).fetchone()
if not res:
res = wals.execute("select latitude, longitude from language where name = '%s';" % row['language']).fetchone()
if res:
l.latitude, l.longitude = res
else:
print(row['language'], row['sil'])
#(u'Classical Nahuatl', u'nci') ???
#(u'Ancient Greek', u'gko')
for pid in params.keys():
value = row[pid]
if value:
value = common.Value(
id='tdir-%s-%s' % (pid, l.id),
name=unicode(bs(value)),
contribution=contribution,
parameter=params[pid],
language=l)
values['%s-%s' % (pid, row['language'])] = value
DBSession.add(value)
def normalize_ref(ref):
ref = re.sub('\s+', ' ', ref).strip()
return unicode(bs(ref)).replace('<i>', '"').replace('</i>', '"')
"""
Ogawa, A. (1998)
Wali, K. et al. (2000)
Lyutikova. -> Lyutikova,
se-Bertit -> se-Berit
missing refs:
Sengupta, G. (2000). Lexical anaphors and pronouns in Bangla. In Lust et al. (eds.), <i>Lexical Anaphors and Pronouns in Selected South Asian Languages</i>. Berlin: Mouton de Gruyter.
Davison, A. Mistry (2000). Lexical anaphors and pronouns in Hindi/Urdu. In Lust et al. (eds.), <i>Lexical Anaphors and Pronouns in Selected South Asian Languages</i>. Berlin: Mouton de Gruyter.
"""
refs = {}
for row in read('references'):
name = re.sub('\s+', ' ', row['entry'].split(').')[0].strip()) + ')'
src = common.Source(
id=row['ref'].strip(), name=name, description=normalize_ref(row['entry']))
refs[name] = src
DBSession.add(src)
for row in read('examples'):
if row['language'] not in languages:
print('example for unknown language "%s"' % row['language'])
continue
s = common.Sentence(
id=row['Nr'].strip(),
name=fix_example(row['original'], repl=' '),
language=languages[row['language']],
analyzed=fix_example(row['original']),
gloss=fix_example(row['gloss']),
description=row['translation'],
source=row['source'],
comment=row['comments'])
has_refs = False
for ref in refs:
if ref in row['source']:
if normalize_ref(row['source']) != refs[ref].description:
print('-->')
print(row['source'])
has_refs = True
common.SentenceReference(sentence=s, source=refs[ref])
if not has_refs:
print('+++++')
print(row['source'])
pid = EXAMPLE_MAP[row['pov']]
if pid:
# associate with value!
o = common.ValueSentence(value=values['%s-%s' % (pid, row['language'])], sentence=s)
DBSession.add(s)
def populate_test_db(engine):
set_alembic_version(engine, '58559d4eea0d')
data = TestData()
data.add_default(common.Dataset,
domain='clld',
jsondata={
'license_icon': 'cc-by',
'license_url': 'http://example.org'
})
data.add_default(common.Contributor, name='A Name', email='*****@*****.**')
for id_, name in {
'b': 'b Name',
'c': 'c Name',
'd': 'd Name',
}.items():
data.add(common.Contributor,
id_,
id=id_,
name=name,
url='http://example.org')
DBSession.add(
common.Editor(dataset=data[common.Dataset],
contributor=data[common.Contributor]))
data.add_default(common.Source)
data.add(common.Source,
'replaced',
id='replaced',
active=False,
jsondata={'__replacement_id__': 'source'})
data.add_default(common.Contribution)
common.ContributionReference(contribution=data[common.Contribution],
source=data[common.Source])
for primary, c in [(True, 'contributor'), (False, 'b'), (True, 'c'),
(False, 'd')]:
common.ContributionContributor(contribution=data[common.Contribution],
primary=primary,
contributor=data['Contributor'][c])
data.add_default(common.Language, latitude=10.5, longitude=0.3)
data[common.Language].sources.append(data[common.Source])
for i, type_ in enumerate(common.IdentifierType):
common.LanguageIdentifier(
language=data[common.Language],
identifier=common.Identifier(
type=type_.value,
id=type_.value + str(i),
name='abc' if type_.name == 'iso' else 'glot1234'))
common.LanguageIdentifier(language=data[common.Language],
identifier=common.Identifier(type='name',
id='name',
name='a'))
for i in range(2, 102):
_l = common.Language(id='l%s' % i, name='Language %s' % i)
_i = common.Identifier(type='iso639-3', id='%.3i' % i, name='abc')
common.LanguageIdentifier(language=_l, identifier=_i)
DBSession.add(_l)
param = data.add_default(common.Parameter)
de = common.DomainElement(id='de', name='DomainElement', parameter=param)
de2 = common.DomainElement(id='de2',
name='DomainElement2',
parameter=param)
valueset = data.add_default(common.ValueSet,
language=data[common.Language],
parameter=param,
contribution=data[common.Contribution])
common.ValueSetReference(valueset=valueset,
source=data[common.Source],
description='10-20')
data.add_default(common.Value,
domainelement=de,
valueset=valueset,
frequency=50,
confidence='high')
data.add(common.Value,
'value2',
id='value2',
domainelement=de2,
valueset=valueset,
frequency=50,
confidence='high')
paramnd = data.add(common.Parameter,
'no-domain',
id='no-domain',
name='Parameter without domain')
valueset = common.ValueSet(id='vs2',
language=data[common.Language],
parameter=paramnd,
contribution=data[common.Contribution])
common.ValueSetReference(valueset=valueset,
source=data[common.Source],
description='10-20')
common.Value(id='v2', valueset=valueset, frequency=50, confidence='high')
unit = data.add_default(common.Unit, language=data[common.Language])
up = data.add_default(common.UnitParameter)
common.UnitValue(id='unitvalue',
name='UnitValue',
unit=unit,
unitparameter=up)
up2 = common.UnitParameter(id='up2', name='UnitParameter with domain')
de = common.UnitDomainElement(id='de', name='de', parameter=up2)
DBSession.add(
common.UnitValue(id='uv2',
name='UnitValue2',
unit=unit,
unitparameter=up2,
unitdomainelement=de))
DBSession.add(common.Source(id='s'))
sentence = data.add_default(common.Sentence,
description='sentence description',
analyzed='a\tmorpheme\tdoes\tdo',
gloss='a\tmorpheme\t1SG\tdo.SG2',
source='own',
comment='comment',
original_script='a morpheme',
language=data[common.Language],
jsondata={'alt_translation': 'Spanish: ...'})
common.SentenceReference(sentence=sentence, source=data[common.Source])
DBSession.add(common.Config(key='key', value='value'))
common.Config.add_replacement('replaced',
'language',
model=common.Language)
common.Config.add_replacement('gone', None, model=common.Language)
DBSession.flush()
def setUp(self):
TestWithDb.setUp(self)
DBSession.add(
common.Dataset(id='dataset',
name='dataset',
description='desc',
domain='clld',
jsondata={'license_icon': 'cc-by'}))
DBSession.add(
common.Source(id='replaced',
active=False,
jsondata={'__replacement_id__': 'source'}))
source = common.Source(id='source')
contributors = {
'contributor': 'A Name',
'b': 'b Name',
'c': 'c Name',
'd': 'd Name'
}
for id_, name in contributors.items():
contributors[id_] = common.Contributor(id=id_,
name=name,
url='http://example.org')
contribution = common.Contribution(id='contribution',
name='Contribution')
common.ContributionReference(contribution=contribution, source=source)
assert common.ContributionContributor(
contribution=contribution,
primary=True,
contributor=contributors['contributor'])
assert common.ContributionContributor(contribution=contribution,
primary=False,
contributor=contributors['b'])
assert common.ContributionContributor(contribution=contribution,
primary=True,
contributor=contributors['c'])
assert common.ContributionContributor(contribution=contribution,
primary=False,
contributor=contributors['d'])
DBSession.add(contribution)
language = common.Language(id='language',
name='Language 1',
latitude=10.5,
longitude=0.3)
language.sources.append(source)
for i, type_ in enumerate(common.IdentifierType):
id_ = common.Identifier(type=type_.value,
id=type_.value + str(i),
name='abc')
common.LanguageIdentifier(language=language, identifier=id_)
for i in range(2, 102):
_l = common.Language(id='l%s' % i, name='Language %s' % i)
_i = common.Identifier(type='iso639-3',
id='%.3i' % i,
name='%.3i' % i)
common.LanguageIdentifier(language=_l, identifier=_i)
DBSession.add(_l)
param = common.Parameter(id='parameter', name='Parameter')
de = common.DomainElement(id='de',
name='DomainElement',
parameter=param)
de2 = common.DomainElement(id='de2',
name='DomainElement2',
parameter=param)
valueset = common.ValueSet(id='valueset',
language=language,
parameter=param,
contribution=contribution)
value = common.Value(id='value',
domainelement=de,
valueset=valueset,
frequency=50,
confidence='high')
DBSession.add(value)
value2 = common.Value(id='value2',
domainelement=de2,
valueset=valueset,
frequency=50,
confidence='high')
DBSession.add(value2)
paramnd = common.Parameter(id='no-domain',
name='Parameter without domain')
valueset = common.ValueSet(id='vs2',
language=language,
parameter=paramnd,
contribution=contribution)
common.ValueSetReference(valueset=valueset, source=source)
value = common.Value(id='v2',
valueset=valueset,
frequency=50,
confidence='high')
DBSession.add(value)
unit = common.Unit(id='unit', name='Unit', language=language)
up = common.UnitParameter(id='unitparameter', name='UnitParameter')
DBSession.add(unit)
DBSession.add(
common.UnitValue(id='unitvalue',
name='UnitValue',
unit=unit,
unitparameter=up))
up2 = common.UnitParameter(id='up2', name='UnitParameter with domain')
de = common.UnitDomainElement(id='de', name='de', parameter=up2)
DBSession.add(
common.UnitValue(id='uv2',
name='UnitValue2',
unit=unit,
unitparameter=up2,
unitdomainelement=de))
DBSession.add(common.Source(id='s'))
sentence = common.Sentence(
id='sentence',
name='sentence name',
description='sentence description',
analyzed='a\tmorpheme\tdoes\tdo',
gloss='a\tmorpheme\t1SG\tdo.SG2',
source='own',
comment='comment',
original_script='a morpheme',
language=language,
jsondata={'alt_translation': 'Spanish: ...'})
common.SentenceReference(sentence=sentence, source=source)
DBSession.add(common.Config(key='key', value='value'))
common.Config.add_replacement('replaced',
'language',
model=common.Language)
common.Config.add_replacement('gone', None, model=common.Language)
DBSession.flush()
