On Thu, 15 Feb 2007 15:31:32 +0100, you wrote: >Hi, >we all know that dwm code base is frozen since several months ago ;) > >But the reality is that every other week I am patching the vanilla sources >with a non-fixed combination of patches (bstack, grid, warp...etc.) >When we apply more than one patch is very easy to find a rejection that >should be fixed manually. The more the patches applied, the more the >rejections, the more complex get to fix it.
Been there. Done that. Know the phenomenon quite well. >Then, what about sharing bundles (mercurial bundles) with applied and >working patches? Everyone can then get the bundle and run the corresponding >unbundle against a clean copy of dwm. With a lot of time saving. I do indeed integrate multiple patches. I have developed a set of scripts to provide some structure to the process. Nothing as fancy as Andrew Morton's quilt scripts, but enough to save me from making too many cockpit errors. >Anyone has an opinion? arg, it's the wiki the place for these files? For starter I would like to see a page with a more structured set of links to available patches and the dwm releases supported. Ideally this would not be a set of links to individual contributor's web spaces but a repository at suckless.org. Similarly I would like to see a collected repository of interesting config files. >As a side efect people will find less resistance trying feature cocktails >and then reporting their feedback to the list. What I understand you to be suggesting is if I integrate patches X, Y, and Z that I would post a bundle providing the results of my integration efforts. Is this correct? If it existed I would be both a client and a contributor. /john
