import zlib
from .compat import (
+ compat_basestring,
compat_chr,
compat_getenv,
compat_html_entities,
def find_xpath_attr(node, xpath, key, val):
# Here comes the crazy part: In 2.6, if the xpath is a unicode,
# .//node does not match if a node is a direct child of . !
- if isinstance(xpath, unicode):
+ if isinstance(xpath, compat_str):
xpath = xpath.encode('ascii')
for f in node.findall(xpath):
# %z (UTC offset) is only supported in python>=3.2
date_str = re.sub(r' ?(\+|-)[0-9]{2}:?[0-9]{2}$', '', date_str)
# Remove AM/PM + timezone
- date_str = re.sub(r'(?i)\s*(?:AM|PM)\s+[A-Z]+', '', date_str)
+ date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
format_expressions = [
'%d %B %Y',
def parse_duration(s):
- if not isinstance(s, basestring if sys.version_info < (3, 0) else compat_str):
+ if not isinstance(s, compat_basestring):
return None
s = s.strip()
(?P<only_hours>[0-9.]+)\s*(?:hours?)|
(?:
- (?:(?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*)?
+ (?:
+ (?:(?P<days>[0-9]+)\s*(?:[:d]|days?)\s*)?
+ (?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*
+ )?
(?P<mins>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*
)?
(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*(?:s|secs?|seconds?)?
res += int(m.group('mins')) * 60
if m.group('hours'):
res += int(m.group('hours')) * 60 * 60
+ if m.group('days'):
+ res += int(m.group('days')) * 24 * 60 * 60
if m.group('ms'):
res += float(m.group('ms'))
return res
def escape_rfc3986(s):
"""Escape non-ASCII characters as suggested by RFC 3986"""
- if sys.version_info < (3, 0) and isinstance(s, unicode):
+ if sys.version_info < (3, 0) and isinstance(s, compat_str):
s = s.encode('utf-8')
return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]")
res = re.sub(r'''(?x)
"(?:[^"\\]*(?:\\\\|\\")?)*"|
'(?:[^'\\]*(?:\\\\|\\')?)*'|
- [a-zA-Z_][a-zA-Z_0-9]*
+ [a-zA-Z_][.a-zA-Z_0-9]*
''', fix_kv, code)
res = re.sub(r',(\s*\])', lambda m: m.group(1), res)
return res