Hi, Thank you so much for reviewing this change.
First I want to mention here that I did deprecate the svn_client_patch() function, but I simply forgot to add SVN_DEPRECATED attribute/modificator. I've just reverted this revision. Your suggestion about svn_client_patch_stream() sounds great to me. I guess I can add a new function, but continue using the old one in the command line. Later on, when we will be implementing xpatch feature, which will also rely on `svn patch` subcommand, we can migrate it (without deprecation). Also, I think svn_client_patch_stream() will be implemented to consume the patch by an APR file, but later on we can rely on stream API. The thing is the whole patch implementation has to be redesigned, but it is going to be an improvement on the side of code. On Fri, May 9, 2025 at 4:02 AM Branko Čibej <br...@apache.org> wrote: > On 9. 5. 25 03:57, Branko Čibej wrote: > > The more I look at this change, the more I wonder if svn_client_patch() > should be deprecated at all. It's clearly a useful bit of code, since we > can use it in our command-line implementation. It's also far easier to use > by third parties if they're not doing something complicated. There's no > reason to not have both svn_client_patch() and svn_client_patch2() in the > non-deprecated API. > > > Although in that case it could be svn_client_patch_stream() or something > like that. And could use our stream API, if that makes any kind of sense. > > -- Brane > -- Timofei Zhakov
