OORE (Object oriented regular expressions) is a python library that attempts to provide an object oriented layer above the re standard library.
The idea behind this project is to enable users to create and combine regular expressions programmatically. The re module which is a more or less simple wrapper over an underlying C library doesn't provide any interface that truly makes this possible, which means that especially complex regular expressions need to be generated by concatenating strings.
This means that dealing with such code is annoying, difficult, and error-prone.
To give you an example, this is how you would create a regular expression that matches a Unicode Language Identifier:
from oore import r
digit = r(u'[0-9]')
alpha = r(u'[A-Za-z')
alphanum = r(u'[0-9A-Za-z]')
unicode_variant_subtag = alphanum[5, 8] | (digit + alphanum[3])
unicode_region_subtag = alpha[2] | digit[3]
unicode_script_subtag = alpha[4]
unicode_language_subtag = alpha[2, 8]
sep = r(u'[-_]')
unicode_language_id = r(u'root') | (
unicode_language_subtag +
(sep + unicode_script_subtag)[0, 1] +
(sep + unicode_region_subtag)[0, 1] +
(sep + unicode_variant_subtag)[0, ...]
)
This can now be used like a re.RegexObject, to get a re.MatchObject you simply use the .match() method:
match = unicode_language_id.match('de-DE')