On Mon Apr 21 11:02:42 2025 +0000, Jinoh Kang wrote:
Actually there's something I've missed: do you know any application that relies on renaming hardlink to itself? Otherwise, since this MR stemmed from !246 (which I assume is from Proton), could we proceed with the original MR without the rename-hardlink fix? Sorry for not noticing this earlier.
I think the intention of this MR is that, rather than fixing an app directly, it simplifies the work for !246 when dealing with casefold filesystems.
The next step would be to explore how to simplify this MR so that only the ncessary groundwork for !246 renamins. The extra complexity could only be justified if there was an app with a possibility of renaming hardlink to the link itself. But maybe, fixing !246 the proper way might end up needing this MR, in which case that alone could be a strong argument for this patch.
Again, my sincere apologies for backtracking from my original position. As someone who took up the reviewing work, I take responsibility for not thinking through what the Wine project's ultimate goal is when judging neccessity of a contribution. I nonetheless appreciate your effort so far, including the multiple iterations that led to the latest revision of !246. It has been a great time working with you!