def _apply_strategy(self, **kwargs):
if not self._text.startswith(FILE_INCLUSION_SYM, self._next_index):
self.error_msg = \
"Invalid token. Expected a file to be included there " + \
"(starting with '" + FILE_INCLUSION_SYM + "')."
return False
self._next_index += 1
self._update_furthest_matched_index()
self._tokens.append(
LexicalToken(TerminalType.file_inclusion_marker,
FILE_INCLUSION_SYM))
if self._text[self._next_index].isspace():
self.error_msg = \
"Invalid token. Expected a file path here, got a whitespace."
return False
comment_start = find_next_comment(self._text, self._next_index)
if comment_start is not None:
file_path = self._text[self._next_index:comment_start].rstrip()
else:
file_path = self._text[self._next_index:].rstrip()
self._next_index += len(file_path)
self._update_furthest_matched_index()
self._tokens.append(LexicalToken(TerminalType.file_path, file_path))
return True
def _apply_strategy(self, **kwargs):
"""
`kwargs` can contain a boolean with key `parsing_slot_def` that is
`True` if the current text is part of a slot definition.
If this boolean is not in `kwargs`, defaults to `False`.
"""
parsing_slot_def = kwargs.get("parsing_slot_def", False)
if parsing_slot_def:
while self._text[self._next_index].isspace():
self._next_index += 1
self._update_furthest_matched_index()
if self._text.startswith(SLOT_VAL_SYM, self._next_index):
self._tokens.append(
LexicalToken(TerminalType.slot_val_marker, SLOT_VAL_SYM))
self._next_index += 1
self._update_furthest_matched_index()
while self._text[self._next_index].isspace():
self._next_index += 1
self._update_furthest_matched_index()
comment_sym = find_next_comment(self._text, self._next_index)
if comment_sym is not None:
slot_value = \
self._text[self._next_index:comment_sym].rstrip()
else:
slot_value = self._text[self._next_index:].rstrip()
self._tokens.append(
LexicalToken(TerminalType.slot_val, slot_value))
self._next_index += len(slot_value)
self._update_furthest_matched_index()
return True
return False
else:
raise ValueError(
"Tried to extract a slot value within a rule that is not " + \
"part of a slot definition."
)
def _apply_strategy(self, **kwargs):
"""
`kwargs` can contain a boolean with key `inside_choice` that is
`True` when the current word is inside a choice and `False` otherwise.
If this boolean is not in `kwargs`, defaults to `False`.
´kwargs´ can also contain a boolean with key `parsing_slot_def`
which is `True` iff the current is in a rule inside a slot definition.
If this boolean is not in `kwargs`, defaults to `False`.
"""
inside_choice = kwargs.get("inside_choice", False)
parsing_slot_def = kwargs.get("parsing_slot_def", False)
# TODO this might be better using regexes
if self._text[self._start_index].isspace():
self.error_msg = \
"Invalid token. Expected a word instead of a whitespace there."
return False
# Find whitespace after the word
next_word_index = self._start_index + 1 # NOTE exclusive
while True:
if (next_word_index == len(self._text)
or self._text[next_word_index].isspace()):
break
next_word_index += 1
next_word_index = \
min_if_exist(
next_word_index,
find_next_comment(self._text, self._start_index)
)
if next_word_index == self._start_index:
self.error_msg = "Invalid token. Expected a word to start here."
return False
for current_char in RuleWord._should_be_escaped_chars:
if next_word_index == self._start_index:
break
next_word_index = \
min_if_exist(
next_word_index,
find_unescaped(self._text, current_char, self._start_index)
)
if inside_choice and next_word_index > self._start_index:
for char_to_escape in RuleWord._should_be_escaped_in_choices_chars:
next_word_index = \
min_if_exist(
next_word_index,
find_unescaped(
self._text, char_to_escape, self._start_index
)
)
if parsing_slot_def and next_word_index > self._start_index:
for char_to_escape in RuleWord._should_be_escaped_in_slot_def_chars:
next_word_index = \
min_if_exist(
next_word_index,
find_unescaped(
self._text, char_to_escape, self._start_index
)
)
if next_word_index == self._start_index:
self.error_msg = "Invalid token. Expected a word to start here."
return False
word = self._text[self._start_index:next_word_index]
self._next_index = next_word_index
self._update_furthest_matched_index()
self._tokens.append(LexicalToken(TerminalType.word, word))
return True