Den fre 5 juni 2026 kl 17:35 skrev Jordan Peck via dev < [email protected]>:
> I've been working on this patch for about 3 weeks, I wrote the first > working prototype without any AI but I've been using Claude Opus mainly for > code review and to help flesh things like the tests. I've been a > professional C++ developer for 10+ years, so C isn't my primary language > but I'm happy to put my name on this code. Learning the SVN codebase has > been a bit of an experience though :D > Nice to know! I think it is great to see some productive AI usage! That said, I was about to voice some of the same concerns Branko already wrote (and he did it much better than I would do): To get a code dump of MANY lines like this makes it a bit difficult to digest. Some before-the-fact discussion would have increased the chance that this was merged in a timely manner. I'm also curious on how this compares with the already existing pristine-on-demand feature coming up in Subversion 1.15. Have you made any tests to compare the WC size? With PoD you should get more or less half the WC size. No idea about the speed difference. Of course, with PoD you would get the network overhead when for example doing a diff, something I presume would not happen here. Kind regards, Daniel > > On Fri, 5 Jun 2026, 14:49 Ivan Zhakov, <[email protected]> wrote: > >> On Thu, 4 Jun 2026 at 14:57, Jordan Peck via dev < >> [email protected]> wrote: >> >>> This patch saves disk space on supported filesystems for byte-identical >>> file installation by utilising filesystem clone APIs. >>> >>> The main benefit is reduced disk space usage during checkout, update, >>> and other working-copy file installs on filesystems that support >>> copy-on-write/block cloning. When the working file is byte-identical to the >>> pristine file, Subversion can install it by copying or reflinking from the >>> pristine instead of storing two independent physical copies of the same >>> bytes. >>> >>> [..] >> >>> >>> Let me know what you think, questions welcome! >>> >>> Just out of curiosity, was this patch by any chance created with AI/LLM? >> >> -- >> Ivan Zhakov >> >
