def parseMorph():
print('Parsing morphological data')
global unknowns
for (sura, suraData) in morphDb.items():
for (aya, ayaData) in suraData.items():
for (group, groupData) in ayaData.items():
for (word, (form, tag, featureStr)) in groupData.items():
wordFeatures.setdefault((sura, aya, group, word),
{})['ascii'] = form
wordFeatures.setdefault((sura, aya, group, word),
{})['unicode'] = tr.to_arabic(form)
(theseFeatures,
theseUnknowns) = parseMorphItem(tag, featureStr)
for (k, v) in theseFeatures.items():
wordFeatures.setdefault((sura, aya, group, word),
{})[k] = v
unknowns |= theseUnknowns
if unknownFeatures:
feats = ' '.join(unknownFeatures)
print(f'\tUnknown features: {feats}')
else:
print(f'\tAll features known')
if (unknownPerFeat):
for feat in sorted(unknownPerFeat):
vals = ' '.join(sorted(unknownPerFeat[feat]))
print(f'\tUnknown: {feat}: {vals}')
if unknowns:
vals = ' '.join(sorted(unknowns))
print(f'\tUnknown labels: {vals}')
if not unknownPerFeat and not unknowns:
print(f'\tAll feature values known')
print(f'Done')
def director(cv):
print('Parsing morphological data')
global unknowns
lemmaIndex = {}
sectionIndex = {}
for (sura, suraData) in morphDb.items():
curSura = cv.node('sura')
cv.feature(curSura, number=sura)
theseSuraFeatures = suraFeatures.get(sura, None)
if theseSuraFeatures:
cv.feature(curSura, **theseSuraFeatures)
for (aya, ayaData) in suraData.items():
curAya = cv.node('aya')
cv.feature(curAya, number=aya)
transFeatures = {
f'translation@{lang}': trans[(sura, aya)]
for (lang, trans) in translations.items()
}
cv.feature(curAya, **transFeatures)
for s in sectionEnd.get((sura, aya), []):
curSection = sectionIndex[s]
cv.terminate(curSection)
del sectionIndex[s]
for (sName, sI, sFeatures) in sectionStart.get((sura, aya), []):
curSection = cv.node(sName)
cv.feature(curSection, number=sI, **sFeatures)
sectionIndex[(sName, sI)] = curSection
nAya = len(ayaData)
for (ig, (group, groupData)) in enumerate(ayaData.items()):
curGroup = cv.node('group')
cv.feature(curGroup, number=group)
nGroup = len(groupData)
for (iw, (word, (form, tag,
featureStr))) in enumerate(groupData.items()):
(theseFeatures,
theseUnknowns) = parseMorphItem(tag, featureStr)
lemma = theseFeatures.get('lemma', None)
if lemma:
thisLemma = lemmaIndex.get(lemma, None)
if thisLemma:
cv.resume(thisLemma)
else:
thisLemma = cv.node('lex')
lemmaIndex[lemma] = thisLemma
cv.feature(thisLemma, lemma=lemma)
curWord = cv.slot()
if lemma:
cv.terminate(thisLemma)
cv.feature(
curWord,
ascii=form,
unicode=tr.to_arabic(form),
space=' '
if iw == nGroup - 1 and ig != nAya - 1 else '',
number=word,
)
cv.feature(curWord, **theseFeatures)
unknowns |= theseUnknowns
cv.terminate(curGroup)
cv.terminate(curAya)
cv.terminate(curSura)
for curSection in sectionIndex.values():
cv.terminate(curSection)
if unknownFeatures:
feats = ' '.join(unknownFeatures)
print(f'\tUnknown features: {feats}')
else:
print(f'\tAll features known')
if (unknownPerFeat):
for feat in sorted(unknownPerFeat):
vals = ' '.join(sorted(unknownPerFeat[feat]))
print(f'\tUnknown: {feat}: {vals}')
if unknowns:
vals = ' '.join(sorted(unknowns))
print(f'\tUnknown labels: {vals}')
if not unknownPerFeat and not unknowns:
print(f'\tAll feature values known')
print(f'Done')