def test_Reference(self):
from pyglottolog.objects import Reference
ref = Reference('bib:key', '12-34', 'German')
self.assertEqual('{0}'.format(ref), '**bib:key**:12-34<trigger "German">')
Reference.from_list(['{0}'.format(ref)])
with self.assertRaises(ValueError):
Reference.from_list(['abc'])
def show(args):
"""Display details of a Glottolog object.
glottolog show <GLOTTOCODE>|<ISO-CODE>|<BIBTEXKEY>
"""
if args.args and ':' in args.args[0]:
if args.args[0].startswith('**'):
ref = Reference.from_string(args.args[0])
else:
ref = Reference(key=args.args[0])
sprint('Glottolog reference {0}'.format(ref),
attrs=['bold', 'underline'])
print()
src = ref.get_source(args.repos)
sprint(src.text())
print()
sprint(src)
return
lang = existing_lang(args)
print()
sprint('Glottolog languoid {0}'.format(lang.id),
attrs=['bold', 'underline'])
print()
sprint('Classification:', attrs=['bold', 'underline'])
args.repos.ascii_tree(lang, maxlevel=1)
print()
sprint('Info:', attrs=['bold', 'underline'])
sprint('Path: {0}'.format(lang.fname), 'green', attrs=['bold'])
sources = lang.sources
if sources:
del lang.cfg['sources']['glottolog']
del lang.cfg['sources']
for line in lang.cfg.write_string().split('\n'):
if not line.startswith('#'):
sprint(line, None, attrs=['bold'] if line.startswith('[') else [])
sprint('Sources:', attrs=['bold', 'underline'])
for src in sources:
src = src.get_source(args.repos)
sprint(src.id, color='green')
sprint(src.text())
print()
def sources(self):
if self.cfg.has_option('sources', 'glottolog'):
return Reference.from_list(self.cfg.getlist(
'sources', 'glottolog'))
return []
def check(args):
"""Check the glottolog data for consistency.
glottolog check [tree|refs]
"""
def error(obj, msg):
args.log.error(message(obj, msg))
def warn(obj, msg):
args.log.warn(message(obj, msg))
def info(obj, msg):
args.log.info(message(obj, msg))
what = args.args[0] if args.args else 'all'
if what in ['all', 'refs']:
for bibfile in args.repos.bibfiles:
bibfile.check(args.log)
if what not in ['all', 'tree']:
return
hhkeys = args.repos.bibfiles['hh.bib'].keys()
iso = args.repos.iso
args.log.info('checking ISO codes against %s' % iso)
args.log.info('checking tree at %s' % args.repos)
by_level = Counter()
by_category = Counter()
iso_in_gl, languoids, iso_splits = {}, {}, []
names = defaultdict(set)
for lang in args.repos.languoids():
# duplicate glottocodes:
if lang.id in languoids:
error(
lang.id, 'duplicate glottocode\n{0}\n{1}'.format(
languoids[lang.id].dir, lang.dir))
languoids[lang.id] = lang
for lang in languoids.values():
ancestors = lang.ancestors_from_nodemap(languoids)
children = lang.children_from_nodemap(languoids)
if lang.latitude and not (-90 <= lang.latitude <= 90):
error(lang, 'invalid latitude: {0}'.format(lang.latitude))
if lang.longitude and not (-180 <= lang.longitude <= 180):
error(lang, 'invalid longitude: {0}'.format(lang.longitude))
assert isinstance(lang.countries, list)
assert isinstance(lang.macroareas, list)
if 'sources' in lang.cfg:
for ref in Reference.from_list(
lang.cfg.getlist('sources', 'glottolog')):
if ref.provider == 'hh' and ref.key not in hhkeys:
error(lang, 'missing source: {0}'.format(ref))
for attr in ['classification_comment', 'ethnologue_comment']:
obj = getattr(lang, attr)
if obj:
obj.check(lang, hhkeys, args.log)
names[lang.name].add(lang)
by_level.update([lang.level.name])
if lang.level == Level.language:
by_category.update([lang.category])
if iso and lang.iso:
if lang.iso not in iso:
warn(lang, 'invalid ISO-639-3 code [%s]' % lang.iso)
else:
isocode = iso[lang.iso]
if lang.iso in iso_in_gl:
error(
isocode,
'duplicate: {0}, {1}'.format(iso_in_gl[lang.iso].id,
lang.id))
iso_in_gl[lang.iso] = lang
if isocode.is_retired and lang.category != 'Bookkeeping':
if isocode.type == 'Retirement/split':
iso_splits.append(lang)
else:
msg = repr(isocode)
level = info
if len(isocode.change_to) == 1:
level = warn
msg += ' changed to [%s]' % isocode.change_to[
0].code
level(lang, msg)
if not lang.id.startswith(
'unun9') and lang.id not in args.repos.glottocodes:
error(lang, 'unregistered glottocode')
for attr in ['level', 'name']:
if not getattr(lang, attr):
error(lang, 'missing %s' % attr)
if lang.level == Level.language:
parent = ancestors[-1] if ancestors else None
if parent and parent.level != Level.family:
error(
lang, 'invalid nesting of language under {0}'.format(
parent.level))
for child in children:
if child.level != Level.dialect:
error(
child, 'invalid nesting of {0} under language'.format(
child.level))
elif lang.level == Level.family:
for d in lang.dir.iterdir():
if d.is_dir():
break
else:
error(lang, 'family without children')
if iso:
changed_to = set(chain(*[code.change_to for code in iso.retirements]))
for code in sorted(iso.languages):
if code.type == 'Individual/Living':
if code not in changed_to:
if code.code not in iso_in_gl:
info(repr(code), 'missing')
for lang in iso_splits:
isocode = iso[lang.iso]
missing = [
s.code for s in isocode.change_to if s.code not in iso_in_gl
]
if missing:
warn(
lang, '{0} missing new codes: {1}'.format(
repr(isocode), ', '.join(missing)))
for name, gcs in sorted(names.items()):
if len(gcs) > 1:
# duplicate names:
method = error
if len([1 for n in gcs if n.level != Level.dialect]) <= 1:
# at most one of the languoids is not a dialect, just warn
method = warn
if len([
1 for n in gcs
if (not n.lineage) or (n.lineage[0][1] != 'book1242')
]) <= 1:
# at most one of the languoids is not in bookkeping, just warn
method = warn
method(
name, 'duplicate name: {0}'.format(', '.join(
sorted([
'{0} <{1}>'.format(n.id, n.level.name[0]) for n in gcs
]))))
def log_counter(counter, name):
msg = [name + ':']
maxl = max([len(k) for k in counter.keys()]) + 1
for k, l in counter.most_common():
msg.append(('{0:<%s} {1:>8,}' % maxl).format(k + ':', l))
msg.append(
('{0:<%s} {1:>8,}' % maxl).format('', sum(list(counter.values()))))
print('\n'.join(msg))
log_counter(by_level, 'Languoids by level')
log_counter(by_category, 'Languages by category')
return by_level