| arraybolt3 | Possibly silly question, but does there already exist a tool for determining what packages could use merging from Debian? I know we have merges.ubuntu.com to help with merges, but AFAICT it just shows when packages have a difference, not when a package version in Ubuntu is falling behind and needs the Debian version pulled in. | 00:48 |
|---|---|---|
| arraybolt3 | Since Ubuntu has managed to work for years, I assume something like this already exists or there's a reason why it's not necessary, but if it would be helpful, perhaps something that took all the packages in Ubuntu and all the packages in Debian and used dpkg --compare-versions on them to see if Ubuntu has an equal or newer version - does that seem like a useful tool? That way anything that | 00:49 |
| arraybolt3 | needed a merge would pop out and could be dealt with. | 00:49 |
| jbicha | arraybolt3: that's what merges.ubuntu.com does 😉 | 01:17 |
| arraybolt3 | Maybe you can show me what page? I kept seeing awfully old stuff in there, places where Ubuntu and Debian had diverged in packaging but were still probably keeping on par as far as version number. | 01:17 |
| arraybolt3 | I don't want to try to do a merge on some ancient relic that isn't in sync for a good reason. | 01:18 |
| jbicha | https://merges.ubuntu.com/main and https://merges.ubuntu.com/universe | 01:18 |
| jbicha | you would need to compare the debian/changelog files and the packaging diff to see why Ubuntu has diverged | 01:19 |
| arraybolt3 | Yeah, but that takes a while, and if I'm just wanting to help clean up merges for the sake of keeping Ubuntu maintained, I don't want to pour a few hours into seeing why something diverged only to learn "Yeah ok this is *supposed* to have diverged, I can't do anything with it". | 01:20 |
| jbicha | here's one example: https://tracker.debian.org/pkg/wpa . There is an infobox labeled Ubuntu wit ha link to "patches for 2:2.10-10ubuntu1" | 01:20 |
| arraybolt3 | Or is nothing supposed to have diverged and stayed that way and so any sync can be cleaned up and is helpful? | 01:20 |
| jbicha | I can look at the packaging diff and then look at the new debian/changelog entries and realize that Ubuntu's changes made it into Debian so I'll go ahead and manually sync the wpa package now | 01:21 |
| jbicha | yes, it takes manual work to review new Debian changes when Ubuntu's packaging is different. That is why it's helpful for Ubuntu developers to upstream our changes to Debian | 01:23 |
| arraybolt3 | I'm guessing I must just not understand the purpose of merges.ubuntu.com entirely - I think what you're talking about helped. I'll possibly take a swing at it and see what I can make happen. | 01:24 |
| arraybolt3 | (I thought there were some packages that diverged and then stayed that way, it sounds more like everything should eventually be merged at some point so just going in, finding a merge that needs done, and doing it is helpful.) | 01:24 |
| jbicha | sometimes, our changes aren't appropriate for Debian so it's not possible. Sometimes our changes might be accepted into Debian but it takes extra work for it to be acceptable to Debian | 01:24 |
| jbicha | sometimes Ubuntu developers are too busy and just fix the Ubuntu issue. And then someone can come along later and either manually merge or try to push the work into Debian | 01:25 |
| arraybolt3 | That makes sense. | 01:26 |
| tintou | Hi there! I've opened a SRU https://bugs.launchpad.net/ubuntu/+source/appstream/+bug/2006110 if anyone has time to sponsor it that would be great 😇 | 09:12 |
| -ubottu:#ubuntu-devel- Launchpad bug 2006110 in appstream (Ubuntu) "[SRU] Update to latest bugfix release for jammy" [Undecided, New] | 09:12 | |
| ricotz | tintou, hi, you will need to subscribe ubuntu-sponsors and attach a debdiff? while even kinetic has a lower version it would require this SRU update first | 09:35 |
| bdmurray | bryceh: 2.1.2-1 | 15:31 |
| bdmurray | bryceh: I'm seeing output now but upgraded to lunar over the weekend so ? | 15:37 |
| bdmurray | !dmb-ping | 16:04 |
| ubottu | bdmurray, kanashiro, rbasak, seb128, sil2100, teward, utkarsh2102: DMB ping | 16:04 |
| teward | *burps loudly* | 16:04 |
| utkarsh21021 | :P | 16:04 |
| rbasak | ahasenack, dannf: FYI, https://code.launchpad.net/~racb/git-ubuntu/+git/usd-importer/+merge/436915 should fix the openssh imports | 19:42 |
| vpa1977 | Hi, in debian there are new versions of openjdk (openjdk-8, openjdk-11, openjdk-17). Since I've made builds for the security release early they have 0ubuntu version. Would it be possible to run sync from debian for those packages to pull in debian versions? | 20:50 |
| vpa1977 | Thank you! =) | 20:51 |
| mwhudson | vpa1977: are the debian and ubuntu versions the same? | 20:52 |
| mwhudson | were we in sync before the security release? | 20:52 |
| vpa1977 | yes, the openjdk-8 contains fixes for sparc64 from the maintainer | 20:52 |
| vpa1977 | yes | 20:52 |
| mwhudson | can't sync openjdk-11 / openjdk-lts surely? | 20:54 |
| vpa1977 | Not sure about sync process, looks like -lts is always Xubuntu, so those are manually uploaded -> the package should not be touched then | 20:56 |
| mwhudson | i mean a sync will preserve the source package name :-) | 20:56 |
| mwhudson | but the other two look doable, let me just do some pedantic due diligence | 20:57 |
| vpa1977 | -8 is maintained in debian and -17 was always in sync | 20:57 |
| === ivoks_ is now known as ivoks | ||
| === coreycb_ is now known as coreycb | ||
| === dupondje1 is now known as dupondje | ||
| bryceh | bdmurray, interesting, well that's good. I'm not spotting anything in terminator's 2.1.2 changelog suggesting a fix but who knows. If you were running 2.1.1 previously, that should have displayed the log urls as clickable (works for me). But for <2.0 terminator they didn't support clickable urls, which is why the -L flag is there. | 22:01 |
| bryceh | it's possible screen is involved in some fashion, like filtering out the clickable links or something, but I'm not as familiar with its use so don't know. screen 4.9.0-4 has some unicode fixes in it, that may have improved behavior. | 22:03 |
| bdmurray | bryceh: oh actually it doesn't work I had '--show-urls' in the comamnd and it is screen that is the problem | 22:04 |
| bryceh | bdmurray, anyway, glad it's working. ideas for improving UX with this are welcome. It's proven hard to programmatically detect clickable-url support in terminals, so hard to work around seamlessly | 22:04 |
| bryceh | bdmurray, ah screen filters the links out? | 22:05 |
| bryceh | https://unix.stackexchange.com/questions/575337/using-terminal-escape-sequences-within-gnu-screen?noredirect=1&lq=1 | 22:07 |
| bryceh | bdmurray, ^^ that may be the issue | 22:07 |
| bdmurray | bryceh: and `--show-urls` doesn't do anything for the results log | 22:08 |
| bryceh | bdmurray, if you file a bug report against ppa-dev-tools I can try to tinker with it and see if it can be improved | 22:08 |
| bdmurray | I'll report a bug about the `--show-urls` issue and the results - the screen issue IDK how important that is | 22:09 |
| bryceh | yeah I don't know how fixable the screen issue is, but if nothing else it'll flag the issue if others run into it. People do use screen. The unicode for creating the links isn't well specified or standardized, so there's some variability in how they're constructed. Possibly a different formatting would make screen work. | 22:12 |
| bryceh | the missing log urls is going to be an easy fix | 22:12 |
| bdmurray | bryceh: I've reported a bug or two | 23:13 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!