On Sat, Nov 29, 2014 at 9:47 AM, Ivan Delalande <col...@ycc.fr> wrote: > On Sat, Nov 29, 2014 at 09:28:24AM +0100, FRIGN wrote: >> On Sat, 29 Nov 2014 08:42:17 +0100 >> Ivan Delalande <col...@ycc.fr> wrote: >> > This is implemented by moving the reqbuf buffer in the middle of a >> > bigger buffer, reqpath. That buffer contains the value of docroot at its >> > beginning and reqbuf simply points to the first byte after this value. >> >> Hey Ivan, >> >> thanks for your patch! I'll take a close look at it as soon as I have >> the time (probably this evening).
Can you attach the patch so I can review it aswell? >> You may want to note that current development has been happening in a >> branch[0] by Hiltjo Posthuma converting quark away from an accept-based >> infrastructure to a stack-based one. I'm not sure on this. I will keep the accept()-based one for now. >> Hiltjo told me he was almost done with the changes, so as soon as he >> finishes this masterpiece, I'll merge it into mainline. It needs alot more polish and is not ready for the consumers yet :). I can't give a time when it will be done. I'm also working on other things. I will need to rewrite some parts because they are ugly. >> If you like you can check out if the problem still exists in the heavily >> modified branch. Yeah it is reproducable in my version. > Oh, that’s interesting, he also has other features I also need and was > planning to write patches for. :) > Great, I'm curious which features do you need / use ? > From the current state of his repository, it seems that the specific > problem fixed by my patch still exists. However I think it would be way > easier to first merge his changes, and I can then re-do my patch. > Please just apply any fixes to the quark version on git.suckless.org, don't wait for my version :) Kind regards, Hiltjo
