12:01:23 <werdahias> #startmeeting 12:01:23 <MeetBot> Meeting started Sat Aug 16 12:01:23 2025 UTC. The chair is werdahias. Information about MeetBot at http://wiki.debian.org/MeetBot. 12:01:23 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic. 12:01:37 <werdahias> #topic rollcall 12:01:49 <werdahias> please indicate you're here 12:01:58 <bastif[mds]> . 12:02:56 <AlexisOpolka[mds]> Hi, I'll also be observing since my planning for today changed quite a bit. 12:03:58 <federico3> hello there 12:05:48 <werdahias> ok ;) 12:06:13 <werdahias> #topic toolchain updates 12:06:40 <werdahias> nothing much to report here; we are on rustc 1.86 thanks to f_g 12:07:00 <werdahias> also we have a new debcargo allowing architecture override 12:07:13 <werdahias> for packages 12:07:25 <werdahias> #topic transitions 12:08:16 <werdahias> I have planned /staged the following: which 6->8, zerocopy / hashbrown, gtk-rs and notify 12:09:02 <werdahias> zerocopy needs a semver package in NEW; gtk-rs will also need to wait, same as notify 12:09:02 <bastif[mds]> started i18n-embed-fl. 12:09:38 <bastif[mds]> now in experimental for trial by jonas for sdml 12:09:53 <bastif[mds]> (includes update of fluent) 12:10:34 <werdahias> noisycoil[mds]: which one do you have planned ? 12:11:53 <werdahias> please add anything you are planning to ##120 12:11:55 <noisycoil[mds]> I'm planning bindgen first 12:12:00 <werdahias> eh ##130 12:12:19 <noisycoil[mds]> And rustic after 12:12:32 <werdahias> time /9 12:12:37 <noisycoil[mds]> Rustix 12:12:44 <werdahias> right 12:13:14 <werdahias> if the time allows it for me I'd like to go ahead with which 12:13:39 <noisycoil[mds]> Yep, looking forward to it 12:14:11 <werdahias> zerocopy will have to wait until hashbrown 0.14 is ready 12:14:20 <werdahias> s/ready/accepted 12:14:30 <werdahias> I'd like to point out rand 12:14:42 <noisycoil[mds]> Have we already started it though? 12:14:57 <werdahias> it's used pretty much anywhere, and has a hug rdep tree 12:14:57 <noisycoil[mds]> Zerocopy I mean 12:15:04 <werdahias> noisycoil[mds]: no 12:15:22 <noisycoil[mds]> OK I thought I saw some packages already uploaded 12:15:33 <werdahias> well, staged in exp, but not uploaded 12:15:35 <noisycoil[mds]> Nevermind 12:15:45 <noisycoil[mds]> Ack 12:16:09 <noisycoil[mds]> Maybe we can ping ftpmasters for hashbrown 12:16:20 <werdahias> already did 12:16:25 <werdahias> #121 12:16:30 <werdahias> ##121 12:16:36 <noisycoil[mds]> Nice, thanks! 12:16:58 <werdahias> I would like some help there; as this is really huge 12:17:14 <noisycoil[mds]> How many packages? 12:17:21 <werdahias> the base rand crates are in exp already 12:17:28 <werdahias> noisycoil[mds]: 100+ 12:17:35 <noisycoil[mds]> Oof 12:17:40 <noisycoil[mds]> Right 12:18:02 <noisycoil[mds]> Can we leave this for after rustix? 12:18:11 <werdahias> sure :) 12:18:12 <noisycoil[mds]> And maybe gtk? 12:18:23 <noisycoil[mds]> I think that's a large one too? 12:18:37 <werdahias> makes sense, yeah 12:18:55 <werdahias> though gtk will have to wait until after 13.1 12:18:59 <noisycoil[mds]> Also I'm pretty sure rand requires actual patching 12:19:14 <werdahias> it's mostly selfcontained though 12:19:16 <noisycoil[mds]> Not just semver bumping 12:19:35 <werdahias> yeah, they changed a bit of API 12:19:44 <noisycoil[mds]> Yep 12:20:21 <werdahias> I would propose a wip-rand brach where updates can be committed 12:21:18 <noisycoil[mds]> Agreed, so we can collaborate 12:21:46 <noisycoil[mds]> I don't think I can commit to (another) such large transition, but I can certainly help 12:21:59 <werdahias> yeah me neither 12:22:08 <werdahias> that's why I asked for help 12:22:35 <noisycoil[mds]> Nice let's do that then 12:22:43 <werdahias> anything else to add for the transitions topic? 12:23:55 <noisycoil[mds]> I think plugwash had a couple others but he's not there, so 12:24:03 <AlexisOpolka[mds]> If I happen to have time to learn how to package for debian, I could help. 12:24:10 <noisycoil[mds]> s/there/here/ 12:24:12 <AlexisOpolka[mds]> But I can't promise anything rn 12:24:15 <noisycoil[mds]> At least afaics 12:24:23 <werdahias> noisycoil[mds]: they opened some issues in dcc 12:24:49 <noisycoil[mds]> Yep I saw that 12:25:01 <noisycoil[mds]> We'll coordinate later 12:25:42 <werdahias> I uploaded mostly small crate updates or selfcontained transitions like logos 12:26:03 <werdahias> ok, I'd move to AOB then unless someone objects ? 12:26:20 <bastif[mds]> what is AOB? 12:26:31 <noisycoil[mds]> Yep 12:26:40 <AlexisOpolka[mds]> I don't know either 12:27:07 <LeChatP[mds]> aob -> any other business If i'm not wrong 12:27:12 <werdahias> any other business 12:27:15 <werdahias> yeah 12:27:20 <AlexisOpolka[mds]> Ohh 12:27:26 <AlexisOpolka[mds]> It makes sense thx 12:27:37 <werdahias> #topic AOB 12:28:20 <federico3> are the team meetings going to be scheduled at the same time every 2 weeks? 12:28:33 <werdahias> federico3: no. 12:28:50 <werdahias> well, if someone organizes them, sure 12:29:01 <noisycoil[mds]> I see no benefit in having teem meeting every 2 weeks 12:29:02 <ncts[m]> I have a draft calling for the first editor meeting for quite a few months, so 12:29:23 <werdahias> We agreed on bimonthly iirc and that's what I strived to do 12:29:38 <bastif[mds]> What about modifying the version string for uploads to experimental so that in unstable we have "-1" instead of "-2" when a crate went through exp? 12:29:54 <werdahias> ncts[m]: annouce it on the ML ;) 12:30:00 <AlexisOpolka[mds]> I thought it was monthly but I'm okay with bimonthly too 12:30:27 <federico3> ah bimonthly as in every 2 months, ok 12:32:12 <werdahias> https://lists.debian.org/debian-rust/2025/01/msg00003.html 12:32:20 <noisycoil[mds]> irc_bastif[mds]: I saw your MR. I'm not entirely sure I agree. While I kinda like the exp suffix, many don't use it, and since you introduced a new environment variable to manage that I think it may ultimately slow down the workflow for little gain 12:32:36 <werdahias> 2 months back when we discussed it 12:32:44 <noisycoil[mds]> But if there's a clear preference from the team to add it I'm ok with that and I'll use it 12:33:00 <noisycoil[mds]> So mine is not in any way a nack 12:33:03 <werdahias> I don't use ~exp either tbh 12:33:26 <werdahias> If the majority agrees, sure, why not 12:34:15 <bastif[mds]> How would that slow down the workflow? You mean the script? 12:34:32 <noisycoil[mds]> Because you need to remember to add an extra environment variable 12:34:51 <noisycoil[mds]> Not a big deal, I recognize 12:34:57 <noisycoil[mds]> But still something 12:35:00 <bastif[mds]> If there was no need for a variable would it be better? 12:35:16 <noisycoil[mds]> Absolutely, but it would still require consensus 12:35:46 <noisycoil[mds]> You may perhaps open a thread on the mailing list and see what feedback you receive 12:35:53 <werdahias> agreed 12:35:55 <bastif[mds]> I was also wondering, how do you just bump the deb version, without updatin the latest upstream version? 12:36:12 <werdahias> REALVER=somever ./update.sh 12:36:29 <bastif[mds]> but that will add version suffix to package name, no? 12:36:48 <noisycoil[mds]> Nono, it won't 12:37:14 <werdahias> no, if somever is the current version (e.g. 0.9) then it will just increase the revision 12:37:43 <bastif[mds]> Ok. I'll have to remember that one 12:38:11 <LeChatP[mds]> ./repackage.sh does that too if i'm not wrong ? 12:38:41 <werdahias> LeChatP[mds]: that just creates the files in build/ 12:38:47 <bastif[mds]> I think it doesn't update any version, or I'm misusing it 12:38:52 <noisycoil[mds]> And does not bump the debian revision 12:38:54 <werdahias> it does not up the revision 12:39:06 <werdahias> ok ;) 12:39:25 <werdahias> anything else to add for AOB ? 12:39:55 <bastif[mds]> Yes. How do we deal with security issues reported on trixie? 12:40:05 <bastif[mds]> or updates for 13.x 12:40:20 <noisycoil[mds]> We talked this a bit with f_g a few days ago 12:40:31 <bastif[mds]> I already mentioned that point here, but it seems not fully decided 12:40:31 <werdahias> for the former f_g wrote a mail to the release team 12:41:14 <werdahias> for 13.x updates: small, targeted fixes might work, but we'd also need to ask the release team 12:41:18 <noisycoil[mds]> I think his stance is sensible in that regards: both security and stable updates assume fixes are small and directed. As such they can be dealt with by simply re-uploading the source package 12:41:30 <bastif[mds]> We'd have to wait for their reply then, IIUC 12:41:37 <werdahias> yeah 12:41:50 <noisycoil[mds]> Still, I would like to have them committed in git 12:41:58 <werdahias> agreed 12:42:17 <bastif[mds]> In a specific branch I guess? 12:42:21 <noisycoil[mds]> And for that we need the tooling, which we don't have 12:43:09 <noisycoil[mds]> There's an open MR of mine dealing with the initial bits in that direction (namely, teaching the tooling how to deal with different branches), but it was not reviewed yet AFAIK and there's an open thread on the mailing list 12:43:23 <federico3> (speaking of which, I'm restarting some work on https://salsa.debian.org/rust-team/debcargo-conf/-/commits/dabs/?ref_type=heads ) 12:43:40 <bastif[mds]> Oh, by tooling you mean the scripts? I though you meant debcargo 12:44:10 <noisycoil[mds]> No debcargo should mostly be fine 12:44:15 <noisycoil[mds]> Yes I meant the scripts 12:44:23 <bastif[mds]> ok! 12:44:28 <noisycoil[mds]> federico3 (IRC): Very nice! Looking forward to it 12:44:54 <federico3> <3 12:45:15 <noisycoil[mds]> Ah and f_g mentioned debcargo needs compat levels to deal with old format 12:45:31 <noisycoil[mds]> So in the end debcargo too will need some tweaking probably 12:46:11 <LeChatP[mds]> I'm packaging RootAsRole. It’s roughly my 26th to 28th new package. It's a never ending loop with always some new weird dev things. I'm contacting them for each exception I do when packaging. For now I don't update any crate, just repackage and relax or upgrade packages. 12:46:22 <LeChatP[mds]> It was a bit tricky to start the process 12:46:59 <noisycoil[mds]> Eh I know =S 12:47:19 <werdahias> personally, I tend to ignore -dev deps 12:47:44 <werdahias> ok, anything for AOB that should be discussed ? 12:47:46 <bastif[mds]> But that would lead to failing autopkgtests, no? 12:47:57 <werdahias> bastif[mds]: not necessarily 12:48:30 <bastif[mds]> do you have an example? 12:48:32 <werdahias> if you grep where the dependency is used it's sometimes just in benches/examples 12:48:38 <bastif[mds]> ok 12:50:15 <werdahias> https://salsa.debian.org/rust-team/debcargo-conf/-/commit/e84aa9e5bfdb5d305012f22cd08ba89bb5fda225 12:50:19 <werdahias> basically that 12:50:34 <werdahias> ok, going to next meet then 12:50:41 <werdahias> #topic next-meet 12:51:06 <bastif[mds]> ok, thanks 12:51:22 <werdahias> I would suggest 16th october then; maybe a bit earlier so people from other TZ can attend 12:52:39 <werdahias> will ask on the ML again, so we can discuss 12:53:25 <werdahias> ok, would like to end the meet then unless someone has to add anything ? 12:54:34 <werdahias> #endmeeting