def add_mark(components, mark):
"""
Case Mark.NONE will be deal with separately by user
"""
comp = list(components)
if mark == Mark.BAR and comp[0] and comp[0][-1].lower() in FAMILY_D:
comp[0] = add_mark_at(comp[0], len(comp[0]) - 1, Mark.BAR)
else:
#remove all marks and accents in vowel part
raw_vowel = accent.add_accent(comp, Accent.NONE)[1].lower()
raw_vowel = utils.join(
[add_mark_char(c, Mark.NONE) for c in raw_vowel])
if mark == Mark.HAT:
pos = max(raw_vowel.find(u"a"), raw_vowel.find(u"o"),
raw_vowel.find(u"e"))
comp[1] = add_mark_at(comp[1], pos, Mark.HAT)
elif mark == Mark.BREVE:
if raw_vowel != u"ua":
comp[1] = add_mark_at(comp[1], raw_vowel.find(u"a"),
Mark.BREVE)
elif mark == Mark.HORN:
if raw_vowel in (u"uo", u"uoi", u"uou"):
comp[1] = utils.join(
[add_mark_char(c, Mark.HORN)
for c in comp[1][:2]]) + comp[1][2:]
elif raw_vowel == u"oa":
comp[1] = add_mark_at(comp[1], 1, Mark.HORN)
else:
pos = max(raw_vowel.find(u"u"), raw_vowel.find(u"o"))
comp[1] = add_mark_at(comp[1], pos, Mark.HORN)
return comp
def add_mark(components, mark):
"""
Case Mark.NONE will be deal with separately by user
"""
comp = list(components)
if mark == Mark.BAR and comp[0] and comp[0][-1].lower() in FAMILY_D:
comp[0] = add_mark_at(comp[0], len(comp[0])-1, Mark.BAR)
else:
#remove all marks and accents in vowel part
raw_vowel = accent.add_accent(comp, Accent.NONE)[1].lower()
raw_vowel = utils.join([add_mark_char(c, Mark.NONE) for c in raw_vowel])
if mark == Mark.HAT:
pos = max(raw_vowel.find(u"a"), raw_vowel.find(u"o"),
raw_vowel.find(u"e"))
comp[1] = add_mark_at(comp[1], pos, Mark.HAT)
elif mark == Mark.BREVE:
if raw_vowel != u"ua":
comp[1] = add_mark_at(comp[1], raw_vowel.find(u"a"), Mark.BREVE)
elif mark == Mark.HORN:
if raw_vowel in (u"uo", u"uoi", u"uou"):
comp[1] = utils.join([add_mark_char(c, Mark.HORN) for c in comp[1][:2]]) + comp[1][2:]
elif raw_vowel == u"oa":
comp[1] = add_mark_at(comp[1], 1, Mark.HORN)
else:
pos = max(raw_vowel.find(u"u"), raw_vowel.find(u"o"))
comp[1] = add_mark_at(comp[1], pos, Mark.HORN)
return comp
def reverse(components, trans):
"""
Reverse the effect of transformation 'trans' on 'components'
If the transformation does not effect the components, return the original string
Workflow:
- Find the part of components that is effected by the transformation
- Transform this part to the original state (remove accent if the trans
is ADD_ACCENT action, remove mark if the trans is ADD_MARK action)
"""
action, factor = get_action(trans)
comps = list(components)
string = utils.join(comps)
if action == Action.ADD_CHAR and string[-1] == trans[1]:
if comps[2]: i = 2
elif comps[1]: i = 1
else: i = 0
comps[i] = comps[i][:-1]
elif action == Action.ADD_ACCENT:
comps = accent.add_accent(comps, Accent.NONE)
elif action == Action.ADD_MARK:
if factor == Mark.BAR:
comps[0] = comps[0][:-1] + \
mark.add_mark_char(comps[0][-1:], Mark.NONE)
else:
if mark.is_valid_mark(comps, trans):
comps[1] = u"".join(
[mark.add_mark_char(c, Mark.NONE) for c in comps[1]])
return comps
def reverse(components, trans):
"""
Reverse the effect of transformation 'trans' on 'components'
If the transformation does not effect the components, return the original string
Workflow:
- Find the part of components that is effected by the transformation
- Transform this part to the original state (remove accent if the trans
is ADD_ACCENT action, remove mark if the trans is ADD_MARK action)
"""
action, factor = get_action (trans)
comps = list(components)
string = utils.join(comps)
if action == Action.ADD_CHAR and string[-1] == trans[1]:
if comps[2]: i = 2
elif comps[1] : i = 1
else: i = 0
comps[i] = comps[i][:-1]
elif action == Action.ADD_ACCENT:
comps = accent.add_accent(comps, Accent.NONE)
elif action == Action.ADD_MARK:
if factor == Mark.BAR:
comps[0] = comps[0][:-1] + \
mark.add_mark_char(comps[0][-1:], Mark.NONE)
else:
if mark.is_valid_mark(comps, trans):
comps[1] = u"".join([mark.add_mark_char(c, Mark.NONE)
for c in comps[1]])
return comps
def is_valid_mark(comps, mark_trans):
"""
Check whether the mark given by mark_trans is valid to add to the components
"""
components = list(comps)
if components[1] != u"":
raw_vowel = accent.add_accent(components, Accent.NONE)[1].lower()
raw_vowel = utils.join([add_mark_char(c, Mark.NONE) for c in raw_vowel])
if mark_trans[0] == 'd' and components[0] \
and components[0][-1].lower() in (u"d", u"đ"):
return True
elif components[1] != u"" and raw_vowel.find(mark_trans[0]) != -1:
return True
else:
return False
def is_valid_mark(comps, mark_trans):
"""
Check whether the mark given by mark_trans is valid to add to the components
"""
components = list(comps)
if components[1] != u"":
raw_vowel = accent.add_accent(components, Accent.NONE)[1].lower()
raw_vowel = utils.join(
[add_mark_char(c, Mark.NONE) for c in raw_vowel])
if mark_trans[0] == 'd' and components[0] \
and components[0][-1].lower() in (u"d", u"đ"):
return True
elif components[1] != u"" and raw_vowel.find(mark_trans[0]) != -1:
return True
else:
return False
def transform(comps, trans):
"""
Transform the given string with transform type trans
"""
components = list(comps)
# Special case for 'ư, ơ'
#if trans[0] == '<' and not trans[1] in (u'ư', u'ơ', u'Ư', u'Ơ'):
# trans = '+' + trans[1]
# (Not our job)
if trans[0] == u'<':
if not components[2]:
# Undo operation
if components[1][-1:] == trans[1]:
return components
# Only allow ư, ơ or ươ sitting alone in the middle part
elif not components[1] or \
(components[1].lower() == u'ư' and trans[1].lower() == u'ơ'):
components[1] += trans[1]
# Quite a hack. If you want to type gi[f = 'giờ', separate()
# will create ['g', 'i', '']. Therefore we have to allow
# components[1] == 'i'.
elif components[1].lower() == 'i' and components[0].lower() == 'g':
components[1] += trans[1]
components = separate(utils.join(components))
if trans[0] == u'+':
# See this and you'll understand:
# transform([u'nn', '', ''],'+n') = [u'nnn', '', '']
# transform([u'c', '', ''],'+o') = [u'c', 'o', '']
# transform([u'c', 'o', ''],'+o') = [u'c', 'oo', '']
# transform([u'c', 'o', ''],'+n') = [u'c', 'o', 'n']
if components[1] == u'':
if utils.is_vowel(trans[1]):
components[1] += trans[1]
else:
components[0] += trans[1]
else:
if components[2] == u'' and utils.is_vowel(trans[1]):
components[1] += trans[1]
else:
components[2] += trans[1]
# If there is any accent, remove and reapply it
# because it is likely to be misplaced in previous transformations
ac = accent.Accent.NONE
for c in components[1]:
ac = accent.get_accent_char(c)
if ac:
break
if ac != accent.Accent.NONE:
# Remove accent
components = accent.add_accent(components, Accent.NONE)
components = accent.add_accent(components, ac)
return components
action, factor = get_action(trans)
if action == Action.ADD_ACCENT:
components = accent.add_accent(components, factor)
elif action == Action.ADD_MARK:
if (mark.is_valid_mark(components, trans)):
components = mark.add_mark(components, factor)
return components
def transform(comps, trans):
"""
Transform the given string with transform type trans
"""
components = list(comps)
# Special case for 'ư, ơ'
#if trans[0] == '<' and not trans[1] in (u'ư', u'ơ', u'Ư', u'Ơ'):
# trans = '+' + trans[1]
# (Not our job)
if trans[0] == u'<':
if not components[2]:
# Undo operation
if components[1][-1:] == trans[1]:
return components
# Only allow ư, ơ or ươ sitting alone in the middle part
elif not components[1] or \
(components[1].lower() == u'ư' and trans[1].lower() == u'ơ'):
components[1] += trans[1]
# Quite a hack. If you want to type gi[f = 'giờ', separate()
# will create ['g', 'i', '']. Therefore we have to allow
# components[1] == 'i'.
elif components[1].lower() == 'i' and components[0].lower() == 'g':
components[1] += trans[1]
components = separate(utils.join(components))
if trans[0] == u'+':
# See this and you'll understand:
# transform([u'nn', '', ''],'+n') = [u'nnn', '', '']
# transform([u'c', '', ''],'+o') = [u'c', 'o', '']
# transform([u'c', 'o', ''],'+o') = [u'c', 'oo', '']
# transform([u'c', 'o', ''],'+n') = [u'c', 'o', 'n']
if components[1] == u'':
if utils.is_vowel(trans[1]):
components[1] += trans[1]
else:
components[0] += trans[1]
else:
if components[2] == u'' and utils.is_vowel(trans[1]):
components[1] += trans[1]
else:
components[2] += trans[1]
# If there is any accent, remove and reapply it
# because it is likely to be misplaced in previous transformations
ac = accent.Accent.NONE
for c in components[1]:
ac = accent.get_accent_char(c)
if ac:
break
if ac != accent.Accent.NONE:
# Remove accent
components = accent.add_accent(components, Accent.NONE)
components = accent.add_accent(components, ac)
return components
action, factor = get_action (trans)
if action == Action.ADD_ACCENT:
components = accent.add_accent(components, factor)
elif action == Action.ADD_MARK:
if (mark.is_valid_mark(components, trans)):
components = mark.add_mark(components, factor)
return components