Re: rfc2616 14.9 polipo compliance (Cache-Control)
Caeies <[email protected]>
| Newsgroups | gmane.comp.web.polipo.user |
|---|---|
| Message-ID | <CAKXuj21+f940tNJqpExnAt50GQLM7BnnmcRVpgpVscwA8TeNBw@mail.gmail.com> |
Hi List, Juliusz, 2012/3/24, Juliusz Chroboczek <[email protected]>: >> Do you agree with me, or do I'm broking something ? > > You're right, as usual. > > IMHO, the right way to fix it would be similar to what is done about > unknown headers -- see hbuf at the end of the same function. That was my first commit (you can look at my repo on github), but I finally revert it since it not the right way to do it. In fact the bug is not in the "full" Cache-Control Directive but only with max_age and s_maxage ones. So I do a new commit that I hope won't break too much things : - max-age and s-maxage are now saved in the disk entry using the object->{max_age,s_maxage} members when available. - They are restored when read back from disk entry in the validateEntry function. People may look at https://github.com/caeies/polipo-caeies/commit/b652512a8640f2f896d71de4c1f92edcdb3bd9dd for review. Comments welcome. Caeies, PS: I'm working on Vary Header too atm, If you want to comment, I'll be happy. ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure