def test_Source_from_entry(mocker):
src = Source.from_entry(
'xyz', mocker.Mock(type='misc', fields={'title': 'abc'}, persons=None))
assert src.id == 'xyz'
assert src.genre == 'misc'
assert 'author' not in src
assert src['title'] == 'abc'
assert '{0}'.format(src) == 'n.d. abc.'
assert repr(src) == '<Source xyz>'
src = Source.from_entry(
'xyz',
mocker.Mock(type='misc',
fields={'title': 'abc'},
persons={'author': ['Alfred E. Neumann', 'T. M.']}))
assert src['author'] == 'Alfred E. Neumann and T. M.'
def bibliography(self) -> typing.Dict[str, Source]:
"""
:returns: `dict` mapping BibTeX IDs to `Reference` instances.
"""
return to_dict(
Source.from_entry(key, entry)
for key, entry in pybtex.database.parse_string(
self.bibfile.read_text(
encoding='utf8'), bib_format='bibtex').entries.items())
def bibliography(self):
"""
:returns: `dict` mapping BibTeX IDs to `Reference` instances.
"""
refs = []
with self.bibfile.open(encoding='utf8') as fp:
for key, entry in pybtex.database.parse_string(
fp.read(), bib_format='bibtex').entries.items():
refs.append(Source.from_entry(key, entry))
return to_dict(refs)
def __init__(self, repos=None, datasets=None, concepticon=None):
API.__init__(self, repos)
self.datasets = datasets or collections.OrderedDict()
concepticon = concepticon
if not concepticon: # pragma: no cover
try:
concepticon = Concepticon(
Config.from_file().get_clone('concepticon'))
except KeyError:
pass
datasets = set()
self.annotations = collections.defaultdict(
lambda: collections.OrderedDict())
for row in reader(self.repos / 'norare.tsv',
delimiter='\t',
dicts=True):
self.annotations[row['DATASET']][row['NAME'].lower()] = {
k.lower(): row[k]
for k in [
'DATASET', 'NAME', 'LANGUAGE', 'STRUCTURE', 'TYPE',
'NORARE', 'RATING', 'SOURCE', 'OTHER', 'NOTE'
]
}
datasets.add(row['DATASET'])
# get bibliography
self.refs = collections.OrderedDict()
with self.repos.joinpath(
'references', 'references.bib').open(encoding='utf-8') as fp:
for key, entry in pybtex.database.parse_string(
fp.read(), bib_format='bibtex').entries.items():
self.refs[key] = Source.from_entry(key, entry)
all_refs = set(self.refs)
if concepticon:
all_refs = all_refs.union(concepticon.bibliography)
for row in reader(self.repos / 'concept_set_meta.tsv',
delimiter='\t',
dicts=True):
row['norare'] = self
row['path'] = self.repos.joinpath('concept_set_meta', row['ID'],
row['ID'] + '.tsv-metadata.json')
self.datasets[row['ID']] = ConceptSetMeta(
**{k.lower(): v
for k, v in row.items()})
self.datasets[row['ID']].source_language = [
lg.lower().strip()
for lg in self.datasets[row['ID']].source_language.split(',')
]
# remaining datasets come from concepticon, we identify them from datasets
concepticon_datasets = [d for d in datasets if d not in self.datasets]
for dataset in concepticon_datasets:
ds = concepticon.conceptlists[dataset]
self.datasets[ds.id] = ConceptSetMeta(
id=ds.id,
author=ds.author,
year=ds.year,
tags=', '.join(ds.tags),
source_language=ds.source_language,
target_language=ds.target_language,
url=ds.url,
refs=ds.refs,
note=ds.note,
alias=ds.alias,
norare=self,
path=concepticon.repos.joinpath('concepticondata',
'conceptlists',
ds.id + '.tsv-metadata.json'))
for dataset in self.datasets.values():
if dataset.refs:
refs = [dataset.refs] if isinstance(dataset.refs,
str) else dataset.refs
for ref in refs:
if ref not in all_refs: # pragma: no cover
raise ValueError(
'missing references.bib: {}'.format(ref))
