1
0
mirror of https://github.com/janeczku/calibre-web synced 2024-12-11 02:30:30 +00:00
calibre-web/vendor/babel/messages/jslexer.py

167 lines
5.2 KiB
Python
Raw Normal View History

# -*- coding: utf-8 -*-
"""
babel.messages.jslexer
~~~~~~~~~~~~~~~~~~~~~~
A simple JavaScript 1.5 lexer which is used for the JavaScript
extractor.
:copyright: (c) 2013 by the Babel Team.
:license: BSD, see LICENSE for more details.
"""
from operator import itemgetter
import re
from babel._compat import unichr
operators = [
'+', '-', '*', '%', '!=', '==', '<', '>', '<=', '>=', '=',
'+=', '-=', '*=', '%=', '<<', '>>', '>>>', '<<=', '>>=',
'>>>=', '&', '&=', '|', '|=', '&&', '||', '^', '^=', '(', ')',
'[', ']', '{', '}', '!', '--', '++', '~', ',', ';', '.', ':'
]
operators.sort(key=lambda a: -len(a))
escapes = {'b': '\b', 'f': '\f', 'n': '\n', 'r': '\r', 't': '\t'}
rules = [
(None, re.compile(r'\s+(?u)')),
(None, re.compile(r'<!--.*')),
('linecomment', re.compile(r'//.*')),
('multilinecomment', re.compile(r'/\*.*?\*/(?us)')),
('name', re.compile(r'(\$+\w*|[^\W\d]\w*)(?u)')),
('number', re.compile(r'''(?x)(
(?:0|[1-9]\d*)
(\.\d+)?
([eE][-+]?\d+)? |
(0x[a-fA-F0-9]+)
)''')),
('operator', re.compile(r'(%s)' % '|'.join(map(re.escape, operators)))),
('string', re.compile(r'''(?xs)(
'(?:[^'\\]*(?:\\.[^'\\]*)*)' |
"(?:[^"\\]*(?:\\.[^"\\]*)*)"
)'''))
]
division_re = re.compile(r'/=?')
regex_re = re.compile(r'/(?:[^/\\]*(?:\\.[^/\\]*)*)/[a-zA-Z]*(?s)')
line_re = re.compile(r'(\r\n|\n|\r)')
line_join_re = re.compile(r'\\' + line_re.pattern)
uni_escape_re = re.compile(r'[a-fA-F0-9]{1,4}')
class Token(tuple):
"""Represents a token as returned by `tokenize`."""
__slots__ = ()
def __new__(cls, type, value, lineno):
return tuple.__new__(cls, (type, value, lineno))
type = property(itemgetter(0))
value = property(itemgetter(1))
lineno = property(itemgetter(2))
def indicates_division(token):
"""A helper function that helps the tokenizer to decide if the current
token may be followed by a division operator.
"""
if token.type == 'operator':
return token.value in (')', ']', '}', '++', '--')
return token.type in ('name', 'number', 'string', 'regexp')
def unquote_string(string):
"""Unquote a string with JavaScript rules. The string has to start with
string delimiters (``'`` or ``"``.)
"""
assert string and string[0] == string[-1] and string[0] in '"\'', \
'string provided is not properly delimited'
string = line_join_re.sub('\\1', string[1:-1])
result = []
add = result.append
pos = 0
while 1:
# scan for the next escape
escape_pos = string.find('\\', pos)
if escape_pos < 0:
break
add(string[pos:escape_pos])
# check which character is escaped
next_char = string[escape_pos + 1]
if next_char in escapes:
add(escapes[next_char])
# unicode escapes. trie to consume up to four characters of
# hexadecimal characters and try to interpret them as unicode
# character point. If there is no such character point, put
# all the consumed characters into the string.
elif next_char in 'uU':
escaped = uni_escape_re.match(string, escape_pos + 2)
if escaped is not None:
escaped_value = escaped.group()
if len(escaped_value) == 4:
try:
add(unichr(int(escaped_value, 16)))
except ValueError:
pass
else:
pos = escape_pos + 6
continue
add(next_char + escaped_value)
pos = escaped.end()
continue
else:
add(next_char)
# bogus escape. Just remove the backslash.
else:
add(next_char)
pos = escape_pos + 2
if pos < len(string):
add(string[pos:])
return u''.join(result)
def tokenize(source):
"""Tokenize a JavaScript source. Returns a generator of tokens.
"""
may_divide = False
pos = 0
lineno = 1
end = len(source)
while pos < end:
# handle regular rules first
for token_type, rule in rules:
match = rule.match(source, pos)
if match is not None:
break
# if we don't have a match we don't give up yet, but check for
# division operators or regular expression literals, based on
# the status of `may_divide` which is determined by the last
# processed non-whitespace token using `indicates_division`.
else:
if may_divide:
match = division_re.match(source, pos)
token_type = 'operator'
else:
match = regex_re.match(source, pos)
token_type = 'regexp'
if match is None:
# woops. invalid syntax. jump one char ahead and try again.
pos += 1
continue
token_value = match.group()
if token_type is not None:
token = Token(token_type, token_value, lineno)
may_divide = indicates_division(token)
yield token
lineno += len(line_re.findall(token_value))
pos = match.end()