def handle_backspace(converted_string, raw_sequence):
"""
Returns a new converted_string and a new raw_sequence
after a backspace.
"""
# I can't find a simple explanation for this, so
# I hope this example can help clarify it:
#
# handle_backspace(thương, thuwongw) -> (thươn, thuwonw)
# handle_backspace(thươn, thuwonw) -> (thươ, thuwow)
# handle_backspace(thươ, thuwow) -> (thư, thuw)
# handle_backspace(thươ, thuw) -> (th, th)
#
# The algorithm for handle_backspace was contributed by @hainp.
deleted_char = converted_string[-1]
converted_string = converted_string[:-1]
_accent = accent.get_accent_char(deleted_char)
_mark = mark.get_mark_char(deleted_char)
if _mark and _accent:
raw_sequence = raw_sequence[:-3]
elif _mark or _accent:
raw_sequence = raw_sequence[:-2]
else:
index = raw_sequence.rfind(deleted_char)
raw_sequence = raw_sequence[:index] + raw_sequence[(index + 1):]
return converted_string, raw_sequence
def make_im_list(word):
"""
Generate a list of possible input method characters that can be used to
type the given word. Each IM character is presented as a list of string with
the IM character in the equivalent position as it would be in the word, all
other position are empty strings.
>>> make_im_list("bến")
[['', 's', ''], ['', 'e', '']]
"""
im_keys = []
l = len(word)
for index, char in enumerate(word):
ac = accent.get_accent_char(char)
mk = mark.get_mark_char(char)
stripped_char = strip(char)
if ac != accent.Accent.NONE:
im_keys.append([""] * index + [reversed_accents[ac]] + [""] * (l - index - 1))
if mk != mark.Mark.NONE:
if mk in [mark.Mark.HORN, mark.Mark.BREVE]:
mk = "w"
else:
mk = stripped_char
im_keys.append([""] * index + [mk] + [""] * (l - index - 1))
return im_keys
def handle_backspace(converted_string, raw_sequence, im_rules=None):
"""
Returns a new raw_sequence after a backspace. This raw_sequence should
be pushed back to process_sequence().
"""
# I can't find a simple explanation for this, so
# I hope this example can help clarify it:
#
# handle_backspace(thương, thuwongw) -> thuwonw
# handle_backspace(thươn, thuwonw) -> thuwow
# handle_backspace(thươ, thuwow) -> thuw
# handle_backspace(thươ, thuw) -> th
#
# The algorithm for handle_backspace was contributed by @hainp.
if im_rules == None:
im_rules = get_telex_definition()
deleted_char = converted_string[-1]
_accent = accent.get_accent_char(deleted_char)
_mark = mark.get_mark_char(deleted_char)
if _mark or _accent:
# Find a sequence of IM keys at the end of
# raw_sequence
ime_keys_at_end = ""
len_raw_sequence = len(raw_sequence)
i = len_raw_sequence - 1
while i >= 0:
if raw_sequence[i] not in im_rules and \
raw_sequence[i] not in "aeiouyd":
i += 1
break
else:
ime_keys_at_end = raw_sequence[i] + ime_keys_at_end
i -= 1
# Try to find a subsequence from that sequence
# that can be converted to the deleted_char
k = 0
while k < len_raw_sequence:
if process_sequence(raw_sequence[i + k:],
im_rules) == deleted_char:
# Delete that subsequence
raw_sequence = raw_sequence[:i + k]
break
k += 1
else:
index = raw_sequence.rfind(deleted_char)
raw_sequence = raw_sequence[:index] + raw_sequence[(index + 1):]
return raw_sequence
def handle_backspace(converted_string, raw_sequence, im_rules=None):
"""
Returns a new raw_sequence after a backspace. This raw_sequence should
be pushed back to process_sequence().
"""
# I can't find a simple explanation for this, so
# I hope this example can help clarify it:
#
# handle_backspace(thương, thuwongw) -> thuwonw
# handle_backspace(thươn, thuwonw) -> thuwow
# handle_backspace(thươ, thuwow) -> thuw
# handle_backspace(thươ, thuw) -> th
#
# The algorithm for handle_backspace was contributed by @hainp.
if im_rules == None:
im_rules = get_telex_definition()
deleted_char = converted_string[-1]
_accent = accent.get_accent_char(deleted_char)
_mark = mark.get_mark_char(deleted_char)
if _mark or _accent:
# Find a sequence of IM keys at the end of
# raw_sequence
ime_keys_at_end = ""
len_raw_sequence = len(raw_sequence)
i = len_raw_sequence - 1
while i >= 0:
if raw_sequence[i] not in im_rules and \
raw_sequence[i] not in "aeiouyd":
i += 1
break
else:
ime_keys_at_end = raw_sequence[i] + ime_keys_at_end
i -= 1
# Try to find a subsequence from that sequence
# that can be converted to the deleted_char
k = 0
while k < len_raw_sequence:
if process_sequence(raw_sequence[i + k:], im_rules) == deleted_char:
# Delete that subsequence
raw_sequence = raw_sequence[:i + k]
break
k += 1
else:
index = raw_sequence.rfind(deleted_char)
raw_sequence = raw_sequence[:index] + raw_sequence[(index + 1):]
return raw_sequence
def add_mark_char(char, mark):
"""
Add mark to a single char.
"""
if char == "":
return ""
case = char.isupper()
ac = accent.get_accent_char(char)
char = accent.add_accent_char(char.lower(), Accent.NONE)
new_char = char
if mark == Mark.HAT:
if char in FAMILY_A:
new_char = "â"
elif char in FAMILY_O:
new_char = "ô"
elif char in FAMILY_E:
new_char = "ê"
elif mark == Mark.HORN:
if char in FAMILY_O:
new_char = "ơ"
elif char in FAMILY_U:
new_char = "ư"
elif mark == Mark.BREVE:
if char in FAMILY_A:
new_char = "ă"
elif mark == Mark.BAR:
if char in FAMILY_D:
new_char = "đ"
elif mark == Mark.NONE:
if char in FAMILY_A:
new_char = "a"
elif char in FAMILY_E:
new_char = "e"
elif char in FAMILY_O:
new_char = "o"
elif char in FAMILY_U:
new_char = "u"
elif char in FAMILY_D:
new_char = "d"
new_char = accent.add_accent_char(new_char, ac)
return utils.change_case(new_char, case)