Re: How do i get news of git releases
- From
Jeff King <peff@peff.net>
- Date
- Sep 22, 2025, 21:35 UTC
- Message-ID
- <20250922213558.GA2269472@coredump.intra.peff.net>
- In-Reply-To
- <xmqqtt0urxva.fsf@gitster.g>
On Mon, Sep 22, 2025 at 02:16:57PM -0700, Junio C Hamano wrote:
Show 14 quoted lines
> Jeff King <peff@peff.net> writes: > > > Yes, they're already annotated tags. But they contain only the version > > number and signature. I suppose they could include the whole set of > > release notes (and it looks like we used to do that in some very old > > tags), > > Eh, which one? I do not recall ever doing so, but I may be > mistaken. > > "git show v0.99.1" gives both tag object contents *and* the output > from "git show v0.99.1^0" for the commit, so it is possible that I > never did so, but those who ask "git show" may get such an > impression?
I looked at:
git for-each-ref --format='%(objectsize) %(refname)' refs/tags | sort -n
which shows a few bigger ones. v0.99.5 is the biggest, with what looks like shortlog output plus some hand-written notes. Ditto v1.4.3.2. But yeah, it is not very many.
Show 21 quoted lines
> > 3. The resulting objects would be much larger (the v2.51.0 tag is 974 > > bytes, but Documentation/RelNotes/2.51.0 is 14K, and some are even > > larger). Git may open them frequently to peel the tags, which may > > make some operations slower. Though it might be OK; we try to cache > > peeled values in packed-refs, and possibly the peeling code could > > learn to parse more progressively (e.g., grab the first 1K to see > > if we hit the end-of-header there). > > > > Those aren't necessarily show-stoppers, but just some top-of-the-head > > thoughts. Junio (the maintainer, who actually makes the tags) might have > > more thoughts on why we used to do that sometimes and don't now. > > I think #3 is a show-stopper. > > We will keep the RelNotes file updated with every batch that updates > the 'master' front, so the contents of that imaginary tag that has > the copy of the release notes would become identical to the in-tree > blob at the point of a release. There has to be a very good reason > why it is beneficial to _duplicate_ the information, not the other > way around to ask why we do not duplicate the information in > different places, I think.
Yeah, I agree the duplication is kind of unseemly. The main reason, I think, would be: some third-party tools may mine information out of the tag automatically, but do not know how to find Documentation/RelNotes. I'm assuming GitHub would do that for the releases page (but actually, I do not know).
If we did care about populating their releases page with more info, I suspect using their API to pass along the content would be a better solution.
-Peff