[Date Prev][Date Next][Thread Prev][Thread Next][Thread Index]
Re: [XaraXtreme-dev] Have implemented a Linux port of GetMemoryStatus ("Function to find available RAM")
- From: Alex Bligh <alex@xxxxxxxxxxx>
- Date: Sat, 27 May 2006 11:40:16 +0100
- Subject: Re: [XaraXtreme-dev] Have implemented a Linux port of GetMemoryStatus ("Function to find available RAM")
--On 27 May 2006 02:11 +0100 "Israel G. Lugo" <ilugo@xxxxxxxxxxxxxxxxxx>
I'm not entirely sure as to how to proceed to submit the patch...
Should I just apply for read+write SVN access to the repository (as
per "Subversion Access Instructions")? That would seem to be the
logical course of action.
This is pretty much like any other project. First we take patches,
and if you end up contributing a lot, we give you svn write access.
From a technical point of view, you want to submit a unified diff.
If you got your source using svn, you can produce this simply by
typing "svn diff" (if you have modified other stuff and don't want
to submit that, then put the filenames you /do/ want to submit
on the line concerned).
If you pulled down the source tarball, you want "diff -u". It's
not as neat as the above because diff needs to be passed the names
of the files you want to diff AND the original:
diff -u wxOil/foobar.cpp.old wxOil/foobar.cpp
If you are careful, you can maintain an original code directory, and
a modified code directory (possibly actually using the same files on
disk using cp -l), then diff -ur between them to do a recursive diff.
As you will gather, the "svn diff" route is far easier and lets you
keep in sync with the rest of the repository.
As Charles mentioned, those with CVS write access will need to see you
have signed the CA before committing patches.