I have a db field that holds a regular expression (one unique regex for
each row in the table). Another row in the table
(db.steps.readable_response) lists sample strings that should satisfy the
regex. I'm trying to create a custom validator that tests the regular
expression against the sample strings at form submission, to highlight
problems in the regex before it's submitted. But at the moment the
re.match() is failing even when I'm positive it should pass (I've tested
them several times using Kiki). Here's the validator class.
class IS_VALID_REGEX(object):
"""
custom validator to check regex in step definitions against the given
readable responses.
"""
def __init__(self):
self.error_message='Given answers do not satisfy regular
expression.'
def __call__(self, value):
request = current.request
answers = request.vars.readable_response
alist = answers.split('|')
value = value.encode('UTF-8')
regex = value.encode('string-escape')
for a in alist:
if re.match(a.strip(), regex, re.I):
print a.strip()
print 'it matched!'
else:
print 'answer ', a, ' did not match the regular expression
provided.'
print regex
return (value, self.error_message)
return (value, None)
The print output show that the correct regex and sample strings are being
evaluated. So why is it not passing?
To complicate things a bit, the regex and sample string both include
unicode characters (polytonic Greek). But elsewhere in the app this doesn't
seem to have created problems when performing a similar re.match().
Thanks,
Ian
