revbump(main/jackett): 0.24.275-1

[robertkirkman]

• Test 5 related to revbump(main/llama-cpp): 0.0.0-b6996-1 #27168

[robertkirkman]

@TomJo2000 after thinking about the problem for a while, I came up with this potential idea, if you check the log this generated and look at the base_commit: log message I printed below this, it looks like this is correct, what do you think?

[TomJo2000]

I think you're applying the wrong formula and getting the right result, and I don't know how to explain that.

You mentioned that you could go into more detail about how it's breaking, and I think I don't have a full picture of what is causing it to break.

If this workaround is consistent I'm fine with it, though I'd prefer you write it as either HEAD^ or HEAD~1, unless those are actually different, which I don't see how they could be.

[robertkirkman]

I think you're applying the wrong formula and getting the right result, and I don't know how to explain that.

I agree that this is confusing, and I don't know exactly how to explain fully why my idea works to prevent the error, but you can see that the printed base_commit: log message in this PR's CI run appears to be correct, right?

I basically reached this current patch by inspecting what various git commands showed in CI versus locally, and how they are different, and noticed that this change appears to produce the correct result in CI,

while the code that produces the correct result locally outside of CI (the current lint-packages.sh in master branch) doesn't work reliably in CI - this edge case I keep reproducing causes the current lint-packages.sh in master branch to fail every time I set up a certain crafted input (these manipulated test draft PRs I keep opening)

You mentioned that you could go into more detail about how it's breaking, and I think I don't have a full picture of what is causing it to break.

I agree it would be best to continue discussing this until the picture of what is happening is clearer, I can try adding log messages of any other commands you can think of that might show additional information, do you have ideas for what other logs you would want to see to inspect this?

I could try to explain in more detail exactly how I set up the crafted input test draft PRs so you can make one too and step through the process of what is happening, would you like me to try that?

If this workaround is consistent I'm fine with it, though I'd prefer you write it as either HEAD^ or HEAD~1,

sure, I'll test that as well.

[robertkirkman]

HEAD~1 instead of HEAD~ is also working.