import hypothesis.strategies as st
from hypothesis import reject
from hypothesis.internal.compat import PY3, hrange, hunichr, text_type, \
int_to_byte
from hypothesis.internal.charmap import categories, as_general_categories
HAS_SUBPATTERN_FLAGS = sys.version_info[:2] >= (3, 6)
UNICODE_CATEGORIES = set(categories())
SPACE_CHARS = set(u' \t\n\r\f\v')
UNICODE_SPACE_CHARS = SPACE_CHARS | set(u'\x1c\x1d\x1e\x1f\x85')
UNICODE_DIGIT_CATEGORIES = set(['Nd'])
UNICODE_SPACE_CATEGORIES = set(as_general_categories('Z'))
UNICODE_LETTER_CATEGORIES = set(as_general_categories('L'))
UNICODE_WORD_CATEGORIES = set(as_general_categories(['L', 'N']))
# This is verbose, but correct on all versions of Python
BYTES_ALL = set(int_to_byte(i) for i in range(256))
BYTES_DIGIT = set(b for b in BYTES_ALL if re.match(b'\\d', b))
BYTES_SPACE = set(b for b in BYTES_ALL if re.match(b'\\s', b))
BYTES_WORD = set(b for b in BYTES_ALL if re.match(b'\\w', b))
BYTES_LOOKUP = {
sre.CATEGORY_DIGIT: BYTES_DIGIT,
sre.CATEGORY_SPACE: BYTES_SPACE,
sre.CATEGORY_WORD: BYTES_WORD,
sre.CATEGORY_NOT_DIGIT: BYTES_ALL - BYTES_DIGIT,
sre.CATEGORY_NOT_SPACE: BYTES_ALL - BYTES_SPACE,
sre.CATEGORY_NOT_WORD: BYTES_ALL - BYTES_WORD,
import sre_parse
import sys
import hypothesis.strategies as st
from hypothesis import reject
from hypothesis.internal.charmap import as_general_categories, categories
from hypothesis.internal.compat import int_to_byte
HAS_SUBPATTERN_FLAGS = sys.version_info[:2] >= (3, 6)
UNICODE_CATEGORIES = set(categories())
SPACE_CHARS = set(" \t\n\r\f\v")
UNICODE_SPACE_CHARS = SPACE_CHARS | set("\x1c\x1d\x1e\x1f\x85")
UNICODE_DIGIT_CATEGORIES = {"Nd"}
UNICODE_SPACE_CATEGORIES = set(as_general_categories("Z"))
UNICODE_LETTER_CATEGORIES = set(as_general_categories("L"))
UNICODE_WORD_CATEGORIES = set(as_general_categories(["L", "N"]))
# This is verbose, but correct on all versions of Python
BYTES_ALL = {int_to_byte(i) for i in range(256)}
BYTES_DIGIT = {b for b in BYTES_ALL if re.match(b"\\d", b)}
BYTES_SPACE = {b for b in BYTES_ALL if re.match(b"\\s", b)}
BYTES_WORD = {b for b in BYTES_ALL if re.match(b"\\w", b)}
BYTES_LOOKUP = {
sre.CATEGORY_DIGIT: BYTES_DIGIT,
sre.CATEGORY_SPACE: BYTES_SPACE,
sre.CATEGORY_WORD: BYTES_WORD,
sre.CATEGORY_NOT_DIGIT: BYTES_ALL - BYTES_DIGIT,
sre.CATEGORY_NOT_SPACE: BYTES_ALL - BYTES_SPACE,
sre.CATEGORY_NOT_WORD: BYTES_ALL - BYTES_WORD,
import hypothesis.strategies as st
from hypothesis import reject
from hypothesis.internal.charmap import as_general_categories, categories
from hypothesis.internal.compat import PY3, hrange, hunichr, int_to_byte, text_type
HAS_SUBPATTERN_FLAGS = sys.version_info[:2] >= (3, 6)
UNICODE_CATEGORIES = set(categories())
SPACE_CHARS = set(u" \t\n\r\f\v")
UNICODE_SPACE_CHARS = SPACE_CHARS | set(u"\x1c\x1d\x1e\x1f\x85")
UNICODE_DIGIT_CATEGORIES = {"Nd"}
UNICODE_SPACE_CATEGORIES = set(as_general_categories("Z"))
UNICODE_LETTER_CATEGORIES = set(as_general_categories("L"))
UNICODE_WORD_CATEGORIES = set(as_general_categories(["L", "N"]))
# This is verbose, but correct on all versions of Python
BYTES_ALL = {int_to_byte(i) for i in range(256)}
BYTES_DIGIT = {b for b in BYTES_ALL if re.match(b"\\d", b)}
BYTES_SPACE = {b for b in BYTES_ALL if re.match(b"\\s", b)}
BYTES_WORD = {b for b in BYTES_ALL if re.match(b"\\w", b)}
BYTES_LOOKUP = {
sre.CATEGORY_DIGIT: BYTES_DIGIT,
sre.CATEGORY_SPACE: BYTES_SPACE,
sre.CATEGORY_WORD: BYTES_WORD,
sre.CATEGORY_NOT_DIGIT: BYTES_ALL - BYTES_DIGIT,
sre.CATEGORY_NOT_SPACE: BYTES_ALL - BYTES_SPACE,
sre.CATEGORY_NOT_WORD: BYTES_ALL - BYTES_WORD,