On Feb 28, 2007, at 06:22, Michael Williams wrote:
One thing: I wasn't sure whether to create the diff in the subdirectory of the source tree in which the problem file resides, or the root of the source tree (plumbed for the former in the end).
Create the diff from the directory where the Portfile exists. The old documentation talks about this: http://darwinports.opendarwin.org/docs/ch15.html#id1033934 Hmm.. it doesn't explicitly state to do it from the Portfile directory. There must be another page where I read that before.
Also, if the fix is accepted, I assume the patch-file will reside somewhere on the MacPorts website, which is how a "port install a2ps" will find it. Before then, while I'm testing my own fixes, how do I persuade MacPorts to look for the patch on the local machine? Is there a convention for giving such patch-files a more descriptive name than just patch-file?
If you have a file foo.c that you're submitting a patch for, the recommendation is to name the patchfile "patch-foo.c.diff" as per the old documentation: http://darwinports.opendarwin.org/docs/ch04s07.html "The standard convention is to name the patch file 'patch- <filename>.diff, with one diff file per file altered in the source. If the filename is ambiguous because there are multiple files with that name in the distribution, supply the path components to uniquely identify the file being patched. Diffs should be created from the top level of worksrcdir, the working source directory where the port was extracted [...]" Ok, oh, so that's where it talked about how to create a diff from the root directory. I was misremembering it.