def find_second_marks(self):
"""Split clusters if a mark is a second or later stacking diacritic."""
for exemplar in list(self.clusters.keys()):
count = self.clusters[exemplar]
for trailer_index in range(len(exemplar.trailers)):
trailer = exemplar.trailers[trailer_index]
# If the mark has already been found to be a always separate mark,
# split the exemplar.
if trailer in self.always_separate_marks:
self.split_exemplar(exemplar, trailer_index, count)
# Only graphemes with more than one mark need to be looked at
# for finding stacking diacritics that are separate.
if trailer_index > 0:
current_mark_ccc = Char.getCombiningClass(trailer)
previous_mark_ccc = Char.getCombiningClass(previous_trailer)
# If a mark has the same combining class (ccc) as the previous mark,
# then the mark is a second or later stacking diacritic and is a separate mark.
# Also, if the mark has already been found to be a always separate mark,
# split the exemplar.
if current_mark_ccc == previous_mark_ccc:
self.always_separate_marks.add(trailer)
self.split_exemplar(exemplar, trailer_index, count)
previous_trailer = trailer
def find_second_marks(self):
"""Split clusters if a mark is a second or later stacking diacritic."""
for exemplar in list(self.clusters.keys()):
count = self.clusters[exemplar]
for trailer_index in range(len(exemplar.trailers)):
trailer = exemplar.trailers[trailer_index]
# If the mark has already been found to be a always separate mark,
# split the exemplar.
if trailer in self.always_separate_marks:
self.split_exemplar(exemplar, trailer_index, count)
# Only graphemes with more than one mark need to be looked at
# for finding stacking diacritics that are separate.
if trailer_index > 0:
current_mark_ccc = Char.getCombiningClass(trailer)
previous_mark_ccc = Char.getCombiningClass(previous_trailer)
# If a mark has the same combining class (ccc) as the previous mark,
# then the mark is a second or later stacking diacritic and is a separate mark.
# Also, if the mark has already been found to be a always separate mark,
# split the exemplar.
if current_mark_ccc == previous_mark_ccc:
self.always_separate_marks.add(trailer)
self.split_exemplar(exemplar, trailer_index, count)
previous_trailer = trailer
def __init__(self, uids, basename, logger):
self.logger = logger
self.uids = uids
self.basename = basename
if Char.isdefined(uids[0]):
self.general = Char.charType(uids[0])
self.cc = Char.getCombiningClass(uids[0])
else:
self.logger.log(
'USV %04X not in ICU; no properties known' % uids[0], 'W')
self.feats = set() # feat tags that affect this char
self.langs = set() # lang tags that affect this char
def __init__(self, uid, basename, logger):
self.logger = logger
self.uid = uid
self.basename = basename
if Char.isdefined(uid):
self.general = Char.charType(uid)
self.cc = Char.getCombiningClass(uid)
self.icuGC = Char.charType(uid)
self.icuJT = Char.getIntPropertyValue(uid, UProperty.JOINING_TYPE)
else:
self.logger.log('USV %04X not in ICU; no properties known' % uid,
'W')
self.feats = set() # feat tags that affect this char
self.langs = set() # lang tags that affect this char
# Additional info from UFO:
self.takesMarks = self.isMark = self.isBase = False
def isnukta(char):
"""True if the character is a nukta."""
if Char.getCombiningClass(char) == 7:
return True
return False
def isnukta(char):
"""True if the character is a nukta."""
if Char.getCombiningClass(char) == 7:
return True
return False
