Monday, 2022-11-28

ektor5[m]<flypig> "ektor5, I've kicked off an ESR78..." <- flypig: Thanks!09:44
flypigektor5[m], sorry for the delay in getting back to you. I can confirm this piece of code is being compiled into the build, at least on ESR78.10:01
flypigI'll see if I can find out what's including <QPainterPath>10:01
ektor5[m]flypig: no problems, thanks for checking10:01
ektor5[m]Unfortunately esr78 is really painful to compile and many libraries don't match with the system ones10:02
flypigI was wondering about your reason for choosing ESR60, but that make sense. I can also check on ESR60, but that'll take a little longer as I'll need to install an older SDK.10:03
ektor5[m]I'm wondering if there is an intree implementations of the Qt Surfaces and maybe a mozconfig switch to enable them instead of Cairo's.10:10
flypigI'm not sure, but it doesn't look like it. In ESR78 at least, QPainterPath is coming from the Qt5 headers. Here's the header "backtrace":10:18
flypig2:40.20 In file included from /usr/include/qt5/QtGui/qtransform.h:37,10:18
flypig2:40.20                  from /usr/include/qt5/QtGui/qimage.h:41,10:18
flypig2:40.20                  from /usr/include/qt5/QtGui/QImage:1,10:18
flypig2:40.20                  from gecko-dev/obj-build-mer-qt-xr/dist/system_wrappers/QtGui/QImage: ,10:18
flypig2:40.20                  from gecko-dev/gecko-dev/gfx/cairo/cairo/src/cairo-qt.h:44,10:19
flypig2:40.20                  from gecko-dev/gecko-dev/gfx/cairo/cairo/src/cairo-qt-surface.cpp:47:10:19
flypig2:40.21 /usr/include/qt5/QtGui/qpainterpath.h:57:20: note: previous definition of ‘class QPainterPath’10:19
flypig2:40.21  class Q_GUI_EXPORT QPainterPath10:19
flypigMaybe that helps?10:19
flypigLike I say, that's for ESR78. I can get you the same for ESR60 later, but it might not be too different.10:19
ektor5[m]So maybe I'm missing a package from Qt?10:20
flypigIt could be, yes. I have to drop off for a bit, but will return later.10:21
ektor5[m]flypig: I'll try to add some dependencies and I'll keep you updated, thanks10:25
flypigektor5[m], sounds good. Let me know how you get on please!11:29
*** ggabriel is now known as Guest23513:42
ektor5[m]<flypig> "ektor5, sounds good. Let me know..." <- Seems like I have a different header inclusion hierarchy, adding QPainterPath in the includes fixed the problem15:37
ektor5[m]but I'm experiencing a different one now, I have no clue:... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/almPNiCgjRPyjnTHIlgQZZDU>)15:38
ilpianistawhat's wrong here?... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/MSVzslTwilhyTQudLoYPbDTJ>)16:12
ilpianista * what's wrong here?... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/JdEYAYgUsowwMikdizapICJl>)16:13
flypigektor5[m], MOZ_FALLTHROUGH is only a hint, you may find it builds if you just comment out the line.16:20
ektor5[m]flypig: It seems it succeeded to compile without that one16:20
flypigOkay, good.16:21
flypigThat might be a misuse of clang::fallthrough, I'm not sure.16:21
ektor5[m]The next step is to compile qtmozembed16:21
flypigYou got gecko-dev built?16:22
flypigFor info, just in case it's useful: https://clang.llvm.org/docs/AttributeReference.html#fallthrough16:24
flypigThere it says "It can only be applied to a null statement placed at a point of execution between any statement and the next switch label.", so I think that MOZ_FALLTHROUGH is just misplaced.16:25
flypigilpianista, I don't have coderus's docker to test it with, but on the standard sfdk, your sb2 command works just fine.16:31
ilpianistaflypig: thanks! I should have tried that one first16:32
flypigMaybe try going inside the chroot to dig around manually.16:32
flypigAlso, maybe drop the `-q` flag so you get a better idea about what it's doing.16:33
ilpianistaI added the -q flag to avoid some zypper output before pasting, but didn't help too much ;-)16:34
flypigAh, that makes sense :)16:42
poetasterflypig, did you have a look at the qtmozembed signal 'orientation' issues?17:48
flypigpoetaster, is that the issue you created webview-screensize to highlight?18:17
poetasterflypig, yes18:23
poetasterflypig, it's difficult to disentangle a potential port related issue, but the fact that size is not communicated on orientation change is constant accross ports.18:24
poetasterpiggz[m], the GS5 is really smooth. but it also bounces between all processors 100% and down all the time (which has drains the battery).18:26
poetasterpiggz[m], Load average: 29.18 29.31 29.80 2/1249 19602 on a sleeping phone with no acive apps18:28
flypigpoetaster, I've not looked at it I'm afraid. Did reverting qtmozembed eed2b5c8fb58d9e07b0 fix it for you?18:30
flypighttps://github.com/sailfishos/qtmozembed/commit/eed2b5c8fb58d9e07b060b614abfc061e26faade18:31
flypig(if not, I'll give it a try tomorrow).18:31
poetasterflypig, haven't tried reverting.18:36
flypigOkay, let me give it a go tomorrow and I'll let you know. If it does fix it, it'll still needs some work (to avoid the crash that the change was introduced to fix), but hopefully it'll clear up the reason for the issue you're experiencing.18:37
poetasterflypig, but I did note that WITHOUT your patch I was getting crash & burn with QML live testin (.64 on a GS29018:37
flypigOkay, that's useful to know. I'm sure there's going to be a middle ground between the two :)18:38
flypig(they'd better be!)18:38
poetastersure thing.18:38
Nekron[m]so what's up with Xperia 10 Mark IV... to buy or not to buy that's the question... any clues here what the folks at Jolla are up to lately?19:08
ektor5[m]flypig: I got it to compile https://github.com/ektor5/meta-browser/tree/ektor5/xulrunner-qt-60.9esr19:41
flypigektor5[m], great work, very impressive!20:14

Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!