elopio: on the first one, l.41, wondering if you can't know the url prefix at that point, instead of allowing anything as a url_path prefix (I guess url_path should be the path from /, but there may be a url suffix that matches a suffix of another url? or a login redirect, with a next via query string?) matiasb: I didn't understand the first part, but I think I understood the second. so, do you think it would be better to use urlparse and match just the path ? elopio: yeah, maybe; wondering if it is worth it, but when I read that assert, I read it as check the url ends with url_path, so checking if that's enough nessita: they do. I'm wondering if it will work setting the info level as default. Would that be ok for you? elopio: or maybe change the regex to force the match to match starting from the root, if that's not too complicated matiasb: I think it would be easier to get the path from urlparse, and then match the regex there. which makes me wonder if we should add helpers to check query strings and other stuff. but for now, I just want to check the url_path. yeap, makes sense elopio: +1 to the add_open_page one \o/ matiasb: please take a look at the url_path assertion. -*- matiasb looks elopio: I think that would work better; but I think that would still work if current_url_path='/123/testing/test' and url_path='/321/testing/test', right? but maybe that's not a problem, I guess sorry, url_path='/testing/test' humph, that would be a problem. but easy fixable putting $ at the end of the url_path. matiasb: but the url_path is documented as a regexp. Should I hardcode the $ on the assert_url_path method, or should I define the url_path on the pages with '$'? elopio: hmm... that makes sense; putting a ^ at the beginning and a $ at the end? (similar to django urls, indeed) I would add that to the url_path definition, to make it more flexible and in that case you have a re.match instead of a re.search, heh matiasb: ok, but what to you mean with the url_path definition? in the actual page object instances and not in the base class? (ie, using the regex with ^ and $ for re.search would be really a re.match without the ^ and $) matiasb: ah, yes, you are right. I think, if it is not too restrictive, having re.match(current_url_path, self.url_path) sounds good but still I don't get what's the best solution for you. no, sounds fine for me. matiasb I'll add a test and change it. ack, thanks (it would be the other way round, indeed: re.match(self.url_path, current_url_path), the pattern first) matiasb: pushed. It seems that match takes care only of the ^, so I had to add the $ really? ok, looking matiasb: yes, the suffix test I added was failing. elopio: ok, looks good :) you just need to escape the + in l.173 and l.227 matiasb: good catch. I'll be struggling with that for some time :) pushed. elopio: you missed one +? l.173 :) agh, I hate this. Sorry matiasb. Fixed and pushed. elopio, cake! -*- nessita runs away -*- elopio runs too