module alias for regex_version1

[GoogleCodeExporter]

module alias for regex_version1

[GoogleCodeExporter]

The only solution I can think of is to do this copy and alter regex.py and 
_regex_code.py:

1. Copy regex.py to regex_v1.py

2. Copy _regex_core.py to_regex_core_v1.py

3. In regex_v1.py, rename all "_regex_core" to "_regex_core_v1"

4. In regex_v1.py, change "DEFAULT_VERSION = VERSION0" to "DEFAULT_VERSION = 
VERSION1"

Original comment by [email protected] on 18 Sep 2011 at 12:04

[GoogleCodeExporter]

Thanks, this is simple enough to be handled individually.
Just the following quick and dirty code seems to work:

########### make_regex_v1.py #### python Lib folder ###########
#! Python
# -*- coding: utf-8 -*-

# Duplicate the py files of the regex module changing the default behaviour to 
VERSION1.
with open("_regex_core.py") as reg_c_orig, open("_regex_core_v1.py", "w") as 
reg_c_v1:
    reg_c_v1.write(reg_c_orig.read())

with open("regex.py") as reg_orig, open("regex_v1.py", "w") as reg_v1:
    reg_src = unicode(reg_orig.read(), "utf-8")
    reg_src = reg_src.replace("_regex_core", "_regex_core_v1")
    reg_src = reg_src.replace("DEFAULT_VERSION = VERSION0", "DEFAULT_VERSION = VERSION1")
    reg_v1.write(reg_src.encode("utf-8"))

################################################

>>> import regex
>>> import regex_v1
>>> regex.findall(r"[\w--b]", "abc")
['a', 'b', 'c']
>>> regex_v1.findall(r"[\w--b]", "abc")
['a', 'c']
>>> regex.findall(r"(?V1)[\w--b]", "abc")
['a', 'c']
>>> regex_v1.findall(r"(?V0)[\w--b]", "abc")
['a', 'b', 'c']
>>> 
######################

Just as an aside, it turns out, that the V0 behaviour isn't the same as re, as 
the latter throws errors on some patterns utilising the set operators:

>>> re.findall(r"[\w--b]", "abc")
Traceback (most recent call last):
  File "<input>", line 1, in <module>
  File "C:\Python27\lib\re.py", line 177, in findall
    return _compile(pattern, flags).findall(string)
  File "C:\Python27\lib\re.py", line 244, in _compile
    raise error, v # invalid expression
error: bad character range
>>> 

I just mention it in case it would matter for inclusion of the regex to the 
standard library, but I believe, it shouldn't be relevant (There is hardly some 
working code relying on some pattern being invalid - only such code would be 
"disturbed" if that pattern would work and could match something in regex.) 
Also, this one isn't a matter of nested flags but the handling of \w in the set 
anyway.

Regards,
    vbr

Original comment by [email protected] on 18 Sep 2011 at 8:50

[GoogleCodeExporter]

Original comment by [email protected] on 10 Jul 2012 at 4:52

• Changed state: Done