[02:31] <gamer191> https://ubuntu-archive-team.ubuntu.com/phased-updates.html hasn't been updated for 3 days
[02:31] <gamer191> Is the bot down?
[02:37] <gamer191> (I'm gonna disconnect now because I'll check replies on the logs)
=== john-cabaj1 is now known as john-cabaj
[04:37] <liushuyu> Eickmeyer: https://launchpad.net/~liushuyu-011/+archive/ubuntu/misc/+sourcepub/16242445/+listing-archive-extra However I have no idea how to test the thing properly
[04:39] <Eickmeyer> liushuyu: That looks about right, I guess we'll see how it plays out in the build. Large diff (obviously), but I looked at it and as long as debhelper-compat =13 then it should do the right thing.
[04:44] <liushuyu> Eickmeyer: Also do you know the reason why a lot of the activities in the software got blocklisted in the Debian version? I did not see the original maintainer note down the reason
[04:44] <Eickmeyer> liushuyu: I can only speculate it has something to do with dfsg, but again, only speculation.
[04:45] <Eickmeyer> I've seen the software in action in actual classrooms in use under Windows, but not under any Linux distro.
[04:46] <liushuyu> Eickmeyer: https://launchpad.net/~liushuyu-011/+archive/ubuntu/misc/+build/28612043 amd64 version is now built. Unfortunately, due to the newer version moving to Qt WebEngine (Chromium-based engine), the software will no longer build on a lot of architectures
[04:47] <Eickmeyer> liushuyu: I imagine we'll lose namely s390x and armhf? I hope we don't lose arm64; need that for Raspberry Pi.
[04:48]  * Eickmeyer isn't in a position to check right now
[04:48] <liushuyu> Eickmeyer: It will only build on amd64, armhf and arm64. Nothing else
[04:48] <Eickmeyer> Ok, I can accept those terms.
[04:54] <liushuyu> Looks like Qt WebEngine is a hard requirement for the newer version of openboard too. It's kind of unfortunate that browser engines usually do not (want to) support many CPU architectures due to the complexity of what they do
[08:27] <philroche> ahasenack>: RE Mantic livecd-rootfs . Thank you
=== gurupras- is now known as guruprasad
=== LocutusO- is now known as LocutusOfBorg
[12:46] <ahasenack> jbicha: wrt poppler, NBS is now just gambas and inkscape
[12:47] <ahasenack> inkscape I made progress, but it still ftbfs elsewhere
[12:47] <ahasenack> gambas I can do more troubleshooting, I was chatting with LocutusOfBorg yesterday here for some ideas
[12:47] <ahasenack> both have bugs with details
[12:49] <jbicha> ahasenack: I almost replied to the gambas bug yesterday. The gambas dh_install bug is the reason that Locutus started building it with gcc-11 (the actual error is earlier)
[12:49] <ahasenack> so we need to find a way to use gcc-11 with std=c++20
[12:49] <ahasenack> or figure out why gcc-11 is needed
[12:50] <jbicha> ahasenack: or we could probably just drop gambas3 on armhf since it has no rdepends and armhf isn't a fully complete architecture since noble
[12:50] <ahasenack> and in armhf, which is just barely faster than riscv64 :(
[12:50] <ahasenack> I think it takes 40min on my pi4 to reach the failure point each time
[12:51] <ahasenack> well, I still have a few hours of +1 today
[12:51] <tarzeau> is kai chuan hsieh here?
[12:52] <jbicha> I vote for just removing it on armhf :)
[12:52] <ahasenack> jbicha: inkscape is https://bugs.launchpad.net/ubuntu/+source/inkscape/+bug/2070367
[12:52] -ubottu:#ubuntu-devel- Launchpad bug 2070367 in inkscape (Ubuntu) "FTBFS with new poppler" [High, Confirmed]
[12:52] <ahasenack> I just can't grok those c++ error messages
[12:52] <ahasenack> In instantiation of ‘Inkscape::UI::Widget::ComboBoxEnum<E>* Inkscape::UI::Dialog::FilterEffectsDialog::Settings::add_combo(T, SPAttr, const Glib::ustring&, const Inkscape::Util::EnumDataConverter<E>&, char*) [with T = SPBlendMode]’:
[12:52] <ahasenack> it's like a poem
[12:53] <ahasenack> written in klingon
[12:53] <jbicha> 🤣
[13:05] <LocutusOfBorg> kapla!
[13:05] <dviererbe> Hi, can someone review & sponsor https://code.launchpad.net/~dviererbe/ubuntu/+source/dash/+git/dash/+merge/468432 ?
[13:07] <LocutusOfBorg> dviererbe, sponsored after fixing changelog version
[13:08] <LocutusOfBorg> btw dviererbe please merge 0.5.12-9 from Debian next time :)
[13:09] <LocutusOfBorg> ahasenack, https://launchpad.net/~costamagnagianfranco/+archive/ubuntu/locutusofborg-ppa/+sourcepub/16243012/+listing-archive-extra
[13:10] <LocutusOfBorg> that none needed flag is coming from an m4 internal of gambas3
[13:10] <LocutusOfBorg> I'm adding some troubleshooting there
[13:11] <dviererbe> LocutusOfBorg: Oh, sorry didn't notice that debian had a higher version number. I will merge dash then next pulse
[13:11] <ahasenack> I'd like to see the gcc command line invocation at that stage
[13:12] <LocutusOfBorg> ahasenack, sure!
[13:12] <LocutusOfBorg> I also printed config.log, hopefully will give us what we need
[13:12] <LocutusOfBorg> dviererbe, if you prepare it.... just tell me
[13:12] <LocutusOfBorg> I can do it
[13:12]  * LocutusOfBorg afk
[13:40] <jbicha> seb128: could you check into removing gambas3/armhf? it was discussed a bit above, relates to bug 2012420
[13:40] -ubottu:#ubuntu-devel- Bug 2012420 in gambas3 (Ubuntu) "armhf patch no longer necessary, causes FTBFS with new poppler" [Undecided, Confirmed] https://launchpad.net/bugs/2012420
[13:43] <LocutusOfBorg> jbicha, we can try to fix still, but removing works even better
[13:49] <jbicha> LocutusOfBorg: I don't think it's worth the time, but if it is fixed, it can come back :)
[14:23] <seb128> jbicha, I didn't realize it had that list of binaries, need to script that, I will try to squeeze it before eod but if some other archive admin want to deal with it please do
[15:59] <LocutusOfBorg> I don't think too, gambas is a programming ide, not sure if anybody runs a basic ide on armhf :D
[17:01] <liushuyu> Have you tried building gambas using clang on armhf as well?
[18:22] <jbicha> ahasenack: ^
[18:23] <ahasenack> I'm not sure I understand
[18:23] <ahasenack> all I tried was to drop the patch that was forcing armhf to use gcc-11, and see that it fails elsewhere
[18:24] <jbicha> ahasenack: there is code in debian/rules & debian/control to use clang for armel, you could try switching that to happen for armhf too
[18:24] <jbicha> I tried in a PPA but I probably did something wrong because it failed to build on all architectures
[18:24] <ahasenack> armel is armhf but without floating point by hardware?
[18:25] <liushuyu> ahasenack: Yes. armel requires fp emulation in the software
[18:27] <ahasenack> lemme finish a review and then I can try
[18:27] <ahasenack> but it takes about 1h to build (when it fails)
[18:28] <liushuyu> ahasenack: I have a local ARM device
[18:28] <liushuyu> Clang: gbx_exec_loop.c:4301:2: error: cannot jump from this indirect goto statement to one of its possible targets
[18:28] <ahasenack> me too, a pi4, I use an armhf lxd container on it
[18:28] <liushuyu> Uh-oh
[18:29] <ahasenack> and i/o is ok-ish, 150Mb/s on hdparm
[18:29] <ahasenack> I use external storage
[18:29] <ahasenack> but this code is just a beast
[18:37] <ahasenack> liushuyu: that error was on armhf using the armel trick?
[18:38] <liushuyu> ahasenack: Yes, I am using clang 18
[18:38] <liushuyu> Clang does not support some of the GCC extensions