def cut_greater_or_equal(self, End):
"""Cuts out any range that is above or equal 'End'."""
for i, x in r_enumerate(self.__intervals):
if x.begin < End: break
else:
del self.__intervals[:]
return
# HERE: '__intervals[i]' is the first interval that
# intersects with 'End-oo'
if x.end > End: x.end = End
if i == len(self.__intervals): return
# Delete all intervals that where skipped.
del self.__intervals[i+1:]
def cut_greater_or_equal(self, End):
"""Cuts out any range that is above or equal 'End'."""
for i, x in r_enumerate(self.__intervals):
if x.begin < End: break
else:
del self.__intervals[:]
return
# HERE: '__intervals[i]' is the first interval that
# intersects with 'End-oo'
if x.end > End: x.end = End
if i == len(self.__intervals): return
# Delete all intervals that where skipped.
del self.__intervals[i+1:]
def prune(TriggerMap, Begin, End):
"""Cut out any element in the trigger map which lies beyong [Begin:End)
"""
L = len(TriggerMap)
# Iterate from 'low' to 'high'
begin_i = None
for i, info in enumerate(TriggerMap):
interval, target = info
if interval.end <= Begin: continue
# Found an interval that intersects with 'Begin' line
if interval.begin < Begin:
interval.begin = Begin
begin_i = i
break
if begin_i is None: del TriggerMap[:] # No element is in range
elif begin_i != 0: del TriggerMap[:begin_i]
# Iterate from 'high' to 'low'
end_i = None
for i, info in r_enumerate(TriggerMap):
interval, target = info
if interval.begin >= End: continue
# Found an interval that intersects with 'End' line
if interval.end > End:
interval.end = End
end_i = i
break
if end_i is None: del TriggerMap[:] # No element in range
elif end_i != L - 1: del TriggerMap[end_i + 1:]
# Delete all intervals which have become empty
i = 0
size = len(TriggerMap)
while i < size:
if TriggerMap[i][0].begin == TriggerMap[i][0].end:
del TriggerMap[i]
size -= 1
else:
i += 1
return
def prune(TriggerMap, Begin, End):
"""Cut out any element in the trigger map which lies beyong [Begin:End)
"""
L = len(TriggerMap)
# Iterate from 'low' to 'high'
begin_i = None
for i, info in enumerate(TriggerMap):
interval, target = info
if interval.end <= Begin: continue
# Found an interval that intersects with 'Begin' line
if interval.begin < Begin:
interval.begin = Begin
begin_i = i
break
if begin_i is None: del TriggerMap[:] # No element is in range
elif begin_i != 0: del TriggerMap[:begin_i]
# Iterate from 'high' to 'low'
end_i = None
for i, info in r_enumerate(TriggerMap):
interval, target = info
if interval.begin >= End: continue
# Found an interval that intersects with 'End' line
if interval.end > End:
interval.end = End
end_i = i
break
if end_i is None: del TriggerMap[:] # No element in range
elif end_i != L-1: del TriggerMap[end_i+1:]
# Delete all intervals which have become empty
i = 0
size = len(TriggerMap)
while i < size:
if TriggerMap[i][0].begin == TriggerMap[i][0].end:
del TriggerMap[i]
size -= 1
else:
i += 1
return
def integrate_path_in_result(self):
#print "#integrate_path_in_result:"
#for i, x in enumerate(self.path):
# try: #print "# [%i] %s, %s, %s" % (i, x[0], x[1], x[2].get_string(Option="utf8"))
# except: #print "# [%i] %s" % (i, x)
for k, info in r_enumerate(self.path):
dummy, bi, dummy = info
if bi != self.operation_index and self.admissible.states[
bi].is_acceptance():
first_remainder_k = k + 1 # (ai, bi) is cut; next state is good
break
else:
first_remainder_k = 1
if first_remainder_k == len(self.path):
# The last element of the path is an acceptance in SM_B, thus it is cut too.
return # Nothing left.
#print "#first_remainder_k:", first_remainder_k
ai, bi, trigger_set = self.path[first_remainder_k]
#print "#ai, bi:", ai, bi
state_index, state = self.get_state(ai, bi)
if state_index != self.result.init_state_index:
##print "#(%s, %s) %s -- epsilon --> %s" % (ai, bi, self.result.init_state_index, state_index)
self.result.get_init_state().target_map.add_transition(
trigger_set, state_index)
#print "#state.target_map:", state.target_map.get_map()
#old_ti = state_index
for ai, bi, trigger_set in islice(self.path, first_remainder_k + 1,
None):
target_index, target_state = self.get_state(ai, bi)
state.add_transition(trigger_set, target_index)
#print "# %i -- %s --> %s" % (old_ti, trigger_set.get_utf8_string(), target_index)
state = target_state
#old_ti = target_index
return
def integrate_path_in_result(self):
#print "#integrate_path_in_result:"
#for i, x in enumerate(self.path):
# try: #print "# [%i] %s, %s, %s" % (i, x[0], x[1], x[2].get_string(Option="utf8"))
# except: #print "# [%i] %s" % (i, x)
for k, info in r_enumerate(self.path):
dummy, bi, dummy = info
if bi != self.operation_index and self.admissible.states[bi].is_acceptance():
first_remainder_k = k + 1 # (ai, bi) is cut; next state is good
break
else:
first_remainder_k = 1
if first_remainder_k == len(self.path):
# The last element of the path is an acceptance in SM_B, thus it is cut too.
return # Nothing left.
#print "#first_remainder_k:", first_remainder_k
ai, bi, trigger_set = self.path[first_remainder_k]
#print "#ai, bi:", ai, bi
state_index, state = self.get_state(ai, bi)
if state_index != self.result.init_state_index:
##print "#(%s, %s) %s -- epsilon --> %s" % (ai, bi, self.result.init_state_index, state_index)
self.result.get_init_state().target_map.add_transition(trigger_set, state_index)
#print "#state.target_map:", state.target_map.get_map()
#old_ti = state_index
for ai, bi, trigger_set in islice(self.path, first_remainder_k+1, None):
target_index, target_state = self.get_state(ai, bi)
state.add_transition(trigger_set, target_index)
#print "# %i -- %s --> %s" % (old_ti, trigger_set.get_utf8_string(), target_index)
state = target_state
#old_ti = target_index
return
def TRY_terminal_delimiter_sequence(Mode, UnicodeSequence,
UnicodeEndSequencePattern,
UponReloadDoneAdr):
UnicodeEndSequencePattern.prepare_count_info(Mode.counter_db,
Setup.buffer_codec)
# Trasform letter by letter.
sequence = flatten_list_of_lists(
transformation.do_character(x, Setup.buffer_codec)
for x in UnicodeSequence)
EndSequenceChunkN = len(sequence)
# Column and line number count for closing delimiter
run_time_counting_required_f, counter_txt = \
counter_for_pattern.get(UnicodeEndSequencePattern, ShiftF=False)
# The Closer Delimiter must be a string. As such it has a pre-determined size.
assert not run_time_counting_required_f
# Column and line number count for 'normal' character.
tm, column_counter_per_chunk = \
counter.get_XXX_counter_map(Mode.counter_db, "me->buffer._input_p",
Trafo=Setup.buffer_codec)
dummy, character_count_txt, dummy = \
counter.get_core_step(tm, "me->buffer._input_p")
txt = []
for i, x in enumerate(sequence):
txt.append(i)
txt.append(Lng.IF_INPUT("==", "0x%X" % x,
FirstF=True)) # Opening the 'if'
txt.append(i + 1)
txt.append("%s\n" % Lng.INPUT_P_INCREMENT())
Lng.INDENT(counter_txt, i + 1)
if column_counter_per_chunk:
txt.append(i + 1)
if column_counter_per_chunk == UnicodeEndSequencePattern.count_info(
).column_n_increment_by_lexeme_length:
txt += Lng.REEFERENCE_P_COLUMN_ADD("me->buffer._input_p",
column_counter_per_chunk)
else:
txt += Lng.REEFERENCE_P_COLUMN_ADD(
"(me->buffer._input_p - %i)" % EndSequenceChunkN,
column_counter_per_chunk)
txt.append(i + 1)
txt.extend(counter_txt)
txt.append(i + 1)
txt.append("break;\n")
for i, x in r_enumerate(sequence):
txt.append(i)
txt.append("%s" % Lng.IF_INPUT(
"==", "0x%X" % Setup.buffer_limit_code, FirstF=False)) # Check BLC
txt.append(i + 1)
txt.append("%s\n" %
Lng.LEXEME_START_SET("me->buffer._input_p - %i" % i))
txt.append(i + 1)
txt.append(
"%s\n" %
Lng.GOTO_RELOAD(UponReloadDoneAdr, True, engine.FORWARD)) # Reload
if i == 0: break
txt.append(i)
txt.append("%s" % Lng.ELSE)
txt.append(i + 1)
txt.append("%s\n" % Lng.INPUT_P_ADD(-i))
txt.append(i)
txt.append("%s\n" % Lng.END_IF())
txt.append(i)
txt.append("%s\n" % Lng.END_IF())
txt.extend(character_count_txt)
# print "##DEBUG:\n%s" % "".join(Lng.GET_PLAIN_STRINGS(txt))
return txt
def TRY_terminal_delimiter_sequence(Mode, UnicodeSequence, UnicodeEndSequencePattern, UponReloadDoneAdr):
UnicodeEndSequencePattern.prepare_count_info(Mode.counter_db,
Setup.buffer_codec)
# Trasform letter by letter.
sequence = flatten_list_of_lists(
transformation.do_character(x, Setup.buffer_codec)
for x in UnicodeSequence
)
EndSequenceChunkN = len(sequence)
# Column and line number count for closing delimiter
run_time_counting_required_f, counter_txt = \
counter_for_pattern.get(UnicodeEndSequencePattern, ShiftF=False)
# The Closer Delimiter must be a string. As such it has a pre-determined size.
assert not run_time_counting_required_f
# Column and line number count for 'normal' character.
tm, column_counter_per_chunk = \
counter.get_XXX_counter_map(Mode.counter_db, "me->buffer._input_p",
Trafo=Setup.buffer_codec)
dummy, character_count_txt, dummy = \
counter.get_core_step(tm, "me->buffer._input_p")
txt = []
for i, x in enumerate(sequence):
txt.append(i)
txt.append(Lng.IF_INPUT("==", "0x%X" % x, FirstF=True)) # Opening the 'if'
txt.append(i+1)
txt.append("%s\n" % Lng.INPUT_P_INCREMENT())
Lng.INDENT(counter_txt, i+1)
if column_counter_per_chunk:
txt.append(i+1)
if column_counter_per_chunk == UnicodeEndSequencePattern.count_info().column_n_increment_by_lexeme_length:
txt += Lng.REEFERENCE_P_COLUMN_ADD("me->buffer._input_p",
column_counter_per_chunk)
else:
txt += Lng.REEFERENCE_P_COLUMN_ADD("(me->buffer._input_p - %i)" % EndSequenceChunkN,
column_counter_per_chunk)
txt.append(i+1)
txt.extend(counter_txt)
txt.append(i+1)
txt.append("break;\n")
for i, x in r_enumerate(sequence):
txt.append(i)
txt.append("%s" % Lng.IF_INPUT("==", "0x%X" % Setup.buffer_limit_code, FirstF=False)) # Check BLC
txt.append(i+1)
txt.append("%s\n" % Lng.LEXEME_START_SET("me->buffer._input_p - %i" % i))
txt.append(i+1)
txt.append("%s\n" % Lng.GOTO_RELOAD(UponReloadDoneAdr, True, engine.FORWARD)) # Reload
if i == 0: break
txt.append(i)
txt.append("%s" % Lng.ELSE)
txt.append(i+1)
txt.append("%s\n" % Lng.INPUT_P_ADD(- i))
txt.append(i)
txt.append("%s\n" % Lng.END_IF())
txt.append(i)
txt.append("%s\n" % Lng.END_IF())
txt.extend(character_count_txt)
# print "##DEBUG:\n%s" % "".join(Lng.GET_PLAIN_STRINGS(txt))
return txt