def test_Example(self):
from dictionaria.lib.ingest import Example
e = Example([('tx', 'a'), ('ft', 'b')])
id_ = e.id
e.set('ref', 'x')
self.assertNotEquals(id_, e.id)
def __call__(self, entry):
example = None
lx = None
rf = None
items = []
for marker, content in entry:
if marker == 'lx':
lx = content
if marker in self.example_props:
if marker == 'rf':
rf = content
elif marker == 'xv':
# new example starts
if example:
# but last one is unfinished
self.log.write(
'# incomplete example in lx: %s - missing xe:\n%s\n\n'
% (lx, example))
example = Example([('tx', content)])
elif marker == 'xe':
# example ends
if example:
if rf:
example.insert(0, ('rf', rf))
example.append(('ft', content))
items.append(('xref', self.xref(example)))
rf = None
example = None
else:
self.log.write(
'# incomplete example in lx: %s - missing xv\n' % lx)
else:
if not example:
self.log.write('incomplete example in lx: %s - missing xv\n' % lx)
else:
example.append((self.example_props[marker], content))
else:
items.append((marker, content))
return entry.__class__(items)
def get_words(self):
"""
:return: generator for the words contained within the entry.
"""
word = None
# if an entry has only one \ps marker but multiple words, the value of \ps is used
# as part-of-speech for all words.
pos = None
example = None
meaning = None
# flag signaling whether we are dealing with the first meaning of a word or
# subsequent ones.
first_meaning = True
# now we loop over the (marker, value) pairs of the entry:
for k, v in self:
# individual words are identified by \lx or \se (sub-entry) markers.
if k in ['lx', 'se']:
if word:
yield self.checked_word(word, meaning)
word = Word(v)
if pos:
word.ps = pos
meaning = Meaning()
elif k == 'sn': # a new sense number: initialize a new Meaning.
if not first_meaning:
self.checked_word(word, meaning)
meaning = Meaning()
first_meaning = False
# meaning-specific markers:
elif k in ['de', 'ge']:
# FIXME: we must support multiple meanings expressed by
# semicolon-separated \ge values, e.g. "jump ; jump at"
setattr(meaning, k, v)
elif k == 'sd':
meaning.sd.append(v)
elif k == 'xv':
if example:
example.xv += ' %s' % v
else:
example = Example(v)
elif k in ['xvm', 'xeg']:
if getattr(example, k):
v = getattr(example, k) + ' ' + v
setattr(example, k, v)
elif k == 'xe':
if example:
example.xe = v
try:
assert meaning
meaning.x.append(example)
except AssertionError:
print(
'no meanings for (sense or subentry of) word %s' % word.form)
example = None
else:
print('xe without xv for word %s' % word.form)
elif k == 'xref':
meaning.xref.append(v)
# word-specific markers:
elif k in ['hm', 'ph']:
if getattr(word, k) is None:
# only record first occurrence of the marker!
setattr(word, k, v)
elif k == 'ps':
pos = word.ps = v
elif k in ['cf', 'mn']:
for vv in v.split(','):
if vv.strip():
word.rel.append((k, vv.strip()))
elif k == 'gxx':
word.non_english_meanings[k].extend(sfm.FIELD_SPLITTER_PATTERN.split(v))
else:
word.data[k].append(v)
if word:
yield self.checked_word(word, meaning)