def get_word_before_cursor_with_kind(self, document: Document) -> WordToken:
"""
Get the word before the cursor and clasify it into one of the kinds
of tokens: NamedCharacter, AsciiOperator, Symbol, etc.
If we have whitespace before the cursor this returns an empty string.
"""
text_before_cursor = document.text_before_cursor
if self._is_space_before_cursor(
document=document, text_before_cursor=text_before_cursor
):
return WordToken("", TokenKind.Null)
start = (
document.find_start_of_previous_word(
WORD=False, pattern=FIND_MATHICS_WORD_RE
)
or 0
)
word_before_cursor = text_before_cursor[len(text_before_cursor) + start :]
text_before_word = text_before_cursor[: len(word_before_cursor)]
if word_before_cursor.startswith(r"\["):
return WordToken(word_before_cursor[2:], TokenKind.NamedCharacter)
elif text_before_word.endswith(r"\["):
return WordToken(word_before_cursor, TokenKind.NamedCharacter)
elif text_before_word[-1:] in CHARGROUP_START:
word_before_cursor = word_before_cursor
elif word_before_cursor[1:] in CHARGROUP_START:
word_before_cursor = word_before_cursor[1:]
if word_before_cursor.isnumeric():
return WordToken(word_before_cursor, TokenKind.Null)
elif word_before_cursor in self.ascii_operators:
return WordToken(word_before_cursor, TokenKind.ASCII_Operator)
elif word_before_cursor.startswith("\1xb"):
return WordToken(word_before_cursor, TokenKind.EscapeSequence)
return word_before_cursor, TokenKind.Symbol
