[00:25] @Eickmeyer is this a gitbuildpackage driven repository? [00:25] because you're missing a LOT of things to recreate without the orig tarball [00:25] i.e. it's not included in the repo or package [00:25] also [00:26] remind me if this is an SRU or not? Because your 0.8.x tarball isn't available [00:26] and i may have to manually import it [00:26] which is pain [00:27] Not an SRU. [00:27] And... gitbuildpackage? Shouldn't be, I don't use gitbuildpackage. [00:27] It has a git submodule. [00:28] The original tarball should be in pristine-tar. [00:28] @teward001 ^ [00:29] Actually.... [00:29] It doesn't have a git submodule. I'm thinking of a different package. [00:30] well [00:30] you need to tell me where the original tarballs are then [00:30] because this has a uscan that points at upstream git [00:30] and says there's a 1.0.0-alpha1 or 0.8.2 [00:30] also [00:30] your patch doesn't seem to have anything in it 😐 [00:31] somehow [00:31] maybe it's a bad clone [00:31] Let me see... [00:31] @Eickmeyer also the branch layout suggests its a gbp or similar - https://git.launchpad.net/raysession?h=master [00:33] teward: To get the original tarball: "pristine-tar checkout raysession_0.8.1.orig.tar.gz" That's standard procedure. [00:34] that's a gbp call [00:34] @teward001: That's the way I was taught how to do this. [00:34] and gbp buildpackage would do that, and it's complaining about 'bad checksum' [00:34] @Eickmeyer then the package is a gbp package [00:34] just saying [00:34] Launchpad didn't complain. [00:34] Launchpad is stupid 😜 [00:36] Did you pull the upstream branch too? [00:37] yep [00:37] xdelta3: target window checksum mismatch: XD3_INVALID_INPUT [00:37] That's completely bizarre. Launchpad built the package with zero issues. [00:38] And, launchpad is the standard we're going by. [00:39] Also, my patch isn't empty, so something must be wrong on your end. [00:39] where did you get the original tarball from? [00:40] the one included in the package here [00:40] because i need to do a diff here [00:40] https://github.com/Houston4444/raysession [00:40] tar.gz there you mean [00:40] for 0.8.1 [00:40] ? [00:40] Yes. [00:41] Though, now that 0.8.2 is out, I should update to that. I didn't expect him to release a bugfix so quickly. [00:41] what version of tar do you have on your system? `tar --version` you're using with your imports. [00:42] I'd have to check, I'm not on that system right now, nor can I easily check. It's whatever is included in eoan. [00:42] ah this explains the discrepancy [00:42] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=897653 is what i'm running into [00:42] Debian bug 897653 in tar "tar 1.30 breaks pristine-tar" [Grave,Fixed] [00:42] 1.30+ breaks older tar imports :P [00:42] Oh, fml. [00:43] Launchpad's probably patched [00:43] but Ubuntu generally might not be for older envs. [00:43] Likely. [00:43] give me 1 minute [00:43] *does evil and backports things* [00:43] LOL [00:44] thank you backportpackage tool [00:45] ... and PPAs. [01:28] @teward001 Any luck? [01:37] gotta install the backport lol [01:37] patience [01:46] No worries. [01:50] um.... [01:50] small problem [01:50] the 'file' being altered? [01:50] doesn't exist [01:50] https://paste.ubuntu.com/p/jYS8QC7Z5R/ [01:51] so i'm not sure HOW this passed on LP [01:52] hmmmm [01:55] @Eickmeyer I'm going to need a second set of eyes, I'll ask infinity, rbasak, or someone else on the release team to take a look [01:55] it's possible your patch is not compliant or doable [01:55] at least, based on debuild's output [01:55] Ok, I'll take a look. [01:55] fails in sbuild and in pdebuild [01:56] and there's no version release with the link patch i think [01:56] so the hard link is 'stuck' I think [01:58] Which patch? [01:58] The link patch? [01:58] Nm, that's what you said. [01:58] That is completely weird. It shouldn't have even built for me if that's the case. [01:59] Yet, I got a build from debuild -S and it uploaded and built just fine in my PPA. [02:00] I think I see the problem. [02:03] well i git cloned and then used it locally. I'll do more tests [02:04] Well, it might be a path issue. [02:04] possibly [02:04] but i'mma test a Focal env just in case [02:04] see if debuild gives me the same warns [02:07] Ok. I just pushed a change to my patch, give that a shot if all else fails. [02:08] I noticed a discrepency between the paths in the other patch and this one. [02:09] nope still fails [02:09] from a git clone [02:09] it fails because the symlink is bad and isn't cloned down it seems [02:09] so i'll have to do this old school :? [02:10] nope *still* fails [02:10] https://paste.ubuntu.com/p/ZydhwhC5d4/ [02:11] *sigh* [02:11] Might be that upstream has to fix the issue first. [02:13] So, not just me. https://paste.ubuntu.com/p/87c7YDmN4T/ [02:13] yeah I think upstream has to [02:13] i'll talk to infinity and others tomorrow in #ubuntu-devel on IRC to ask them if they can provide a solution to fix the unsafe symlink issue [02:14] OK [02:15] it might be possible to just hand-wave 0.8.1 without the symlink patch in because we can't patch it [02:15] but that's a decision above me [02:23] Ok, keep me posted.