Hi Armin, I'm thinking that if you could review and noticed so many "potential issues" in 15 minutes, I'm entirely sure that many people here would appreciate you could invert 1 day of your life in a deeper and exhaustive review!! I'm sure that would make stronger framework. :-D
cheers, alex f El 03/08/2009 10:22, Armin Ronacher escribió: > Hi, > > >> True. but I would not call it a race condition. We timestamp >> everything with the time when a request arrives, not when it is >> processed, unless specified otherwise (datetime.now() instead of >> request.now) >> > True. But that does not make it a better idea. Also, datetime.now() > should be consistently replaced with datetime.utcnow() because using > anythign else than UTC data internally is problematic for various > reasons. See the discussion on that topic in various i18n/l10n > libraries such as babel / pytz. > > >> True but I believe we never do that in web2py. It is also true that >> nothing prevent the user from doing it but the same would be true with >> other frameworks. >> > You're not doing it, a user might be doing that by accident or because > he things it should work. This problem does not exist in other > frameworks because besides web2py I don't know a single one that does > this sort of execfile() + namespace thing or uses any other kind of > throwaway modules. As soon as a single reference leaks from the > execfile()'d namespace you're in big troubles and due to the open > nature of Python this could happen very, very fast. > > >> Yes but because all relevant application code is executed within a >> context and there are no references outside the context to stuff >> inside the context, when a request is completed, the context is >> deleted and everything should be garbage collected. >> > That depends on two things. First not having a reference leaked, > which could happen with abstract base classes and other stuff that > uses registries or steals non-weak references. Also and more > importantly, the file descriptor limit is very low and the majority of > Python implementations will only collect that on the GC run (always, > no matter where references are). Say you're opening three files per > request and you have more than 100 requests/sec you could lose all > file descriptors that are available before the GC even thought about > running. > > > Regards, > Armin > > > > > -- Alejandro Fanjul Fdez. [email protected] www.mhproject.org --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "web2py-users" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/web2py?hl=en -~----------~----~----~----~------~----~------~--~---
