def __init__(self, repos=None):
"""
:param repos: Path to a clone or source dump of concepticon-data.
"""
if (repos is None) and cldfcatalog:
repos = cldfcatalog.Config.from_file().get_clone('concepticon')
API.__init__(self, repos)
self._to_mapping = {}
def __init__(self,
repos: typing.Optional[typing.Union[str,
pathlib.Path]] = None):
"""
:param repos: Path to a clone or source dump of concepticon-data.
"""
repos = repos or cldfcatalog.Config.from_file().get_clone(
'concepticon')
API.__init__(self, repos)
self._to_mapping = {}
def __init__(self, repos='.'):
API.__init__(self, repos=repos)
self.repos = self.repos.resolve()
self.tree = self.repos / 'languoids' / 'tree'
if not self.tree.exists():
raise ValueError('repos dir %s missing tree dir: %s' % (self.repos, self.tree))
if not self.repos.joinpath('references').exists():
raise ValueError('repos dir %s missing references subdir' % (self.repos,))
for name, cls in self.__config__.items():
fname = self.repos / 'config' / (name + '.ini')
setattr(self, name, config.Config.from_ini(fname, object_class=cls))
def __init__(self, repos='.', *, cache: bool = False):
"""
:param repos: Path to a copy of `<https://github.com/glottolog/glottolog>`_
:param cache: Indicate whether to cache `Languoid` objects or not. If `True`, the API must \
be used read-only.
"""
API.__init__(self, repos=repos)
#: Absolute path to the copy of the data repository:
self.repos: pathlib.Path = pathlib.Path.cwd() / self.repos
#: Absolute path to the `tree` directory in the repos.
self.tree: pathlib.Path = self.repos / 'languoids' / 'tree'
if not self.tree.exists():
raise ValueError('repos dir %s missing tree dir: %s' %
(self.repos, self.tree))
if not self.repos.joinpath('references').exists():
raise ValueError('repos dir %s missing references subdir' %
(self.repos, ))
self.cache = Cache() if cache else None
def __init__(self, repos=None):
API.__init__(self, repos=repos)
# Initialize component registry:
self.gsm = getGlobalSiteManager()
# Add methods to register utilities to the registry, so plugins don't have to deal with
# ZCA details at all:
self.gsm.register_clusterer = register_clusterer.__get__(self.gsm)
self.gsm.register_colexifier = register_colexifier.__get__(self.gsm)
self.gsm.register_clicsform = register_clicsfom.__get__(self.gsm)
# Load defaults for pluggable functionality first:
plugin.includeme(self.gsm)
# Now load third-party plugins:
for ep in pkg_resources.iter_entry_points('clics.plugin'):
ep.load()(self.gsm)
def __init__(self, repos=None):
"""
:param repos: Path to a clone or source dump of concepticon-data.
"""
API.__init__(self, repos or REPOS_PATH)
self._to_mapping = {}
def test_API():
api = API('.')
assert api.repos.exists()
assert 'repository' in '%s' % api
assert not api.path('unknown', 'path').exists()
def __init__(self, repos, wiki=None):
API.__init__(self, repos)
self.wiki = wiki or self.repos.resolve().parent / 'grambank.wiki'
self.gb20 = GB20(self.path('gb20.txt'))
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))