Arch Linux User Repository

@dopsi Removing IgnorePkg = qelectrotech in my /etc/pacman.conf... Updating to 0.4-3 and compiling (+2 hours on my Celeron 1GB RAM)... It's working now, thanks very much! @awpe and other QET users Please, vote for this packaage in AUR!

@dopsi It's working now, thanks!

@DanielH I reverted the PKGBUILD to the original version.

@dopsi At https://github.com/aur-archive/qelectrotech , you will find the initial commit of qelectrotech 0.4-1: PKGBUILD, .SRCINFO and qelectrotech.install files. I've used them to build and install qelectrotech and... by Neptune's beards! It works! Qelectrotech has resurrected without issues: QET and user collections are availables in Elements Panel, GUI translations works as expected... IMHO, reverting AUR's QET to this initial commit will kill all the bugs at once. Could you take this task on your hands? Will be appreciated. BTW: I have no knowledge nor time to collaborate now, perhaps I will when next QET version comes... I'm trying to learn...

@DanielH if you had qelectrotech from the AUR3, I do not have the PKGBUILD. I took it over as it seemed unmaintained since the switch to the AUR4. The diff from 0.4-1 to 0.4-2 on the AUR4 can be found here (https://aur.archlinux.org/cgit/aur.git/commit/?h=qelectrotech&id=ba06d6452c81de41477b252bcd6fd2d4cbab4a7c). If reverting the changes fixes the bug, let me know.

Yesterday, using yaourt I've updated qelectrotech to 0.4-2 in my home machine (Core2 Duo 2GB RAM, Arch + Openbox). During compilation stage, I've received a lot of warnings about replacements on QET_COMMON_COLLECTION_PATH; I thought that these warnings were due to qmake-qt4 DEFINES+=QET_COMMON_COLLECTION_PATH=/usr/share/qelectrotech/elements line in PKGBUILD. Now, when Qelectrotech runs the QET collection (official elements) isn't available in Elements Panel, despite those folders, subfolders and .eml elements are normally placed at /usr/share/qelectrotech/elements . Translations (english, spanish) doesn't works, program language is always french, as awpe commented. Pityfully, I've not stated EXPORT=2 in /etc/yaourtrc, so my pacman cache doesn't store my previous qelectrotech 0.4-1 package :-( I freezed Qelectrotech at 0.4-1 at my work machine (Arch + dwm, Celeron 1GB RAM), stating IgnorePkg = qelectrotech in /etc/pacman.conf, until these issues are solved. By the way: if you have the previous 0.4-1 PKGBUILD, could you publish it to diff against the new 0.4-2? I think source code has not changes, perhaps these issues are originated in the new PKGBUILD...

@awpe: I have not yet found the fix for the socket error, but the language change should be fixed now.

And sometimes it gives this error: QETSingleApplication::sendMessage() : QLocalSocket::connectToServer: ÐÑказано в Ñоединении ( qelectrotech-UNKNOWN ) with unprinted chars...

Bad bad bad... $ valgrind qelectrotech --lang-dir=qelectrotech /usr/share/qelectrotech/lang/qet_en.qm ==2510== Memcheck, a memory error detector ==2510== Copyright (C) 2002-2013, and GNU GPL'd, by Julian Seward et al. ==2510== Using Valgrind-3.10.1 and LibVEX; rerun with -h for copyright info ==2510== Command: qelectrotech --lang-dir=qelectrotech /usr/share/qelectrotech/lang/qet_en.qm ==2510== ==2515== Warning: invalid file descriptor 1024 in syscall close() ==2515== Warning: invalid file descriptor 1025 in syscall close() ==2515== Warning: invalid file descriptor 1026 in syscall close() ==2515== Warning: invalid file descriptor 1027 in syscall close() ==2515== Use --log-fd=<number> to select an alternative log fd. ==2515== Warning: invalid file descriptor 1028 in syscall close() ==2515== Warning: invalid file descriptor 1029 in syscall close() ==2510== Conditional jump or move depends on uninitialised value(s) ==2510== at 0x501F87: ??? (in /usr/bin/qelectrotech) ==2510== by 0x502198: ??? (in /usr/bin/qelectrotech) ==2510== by 0x4EBC0D: ??? (in /usr/bin/qelectrotech) ==2510== by 0x4EDFD9: ??? (in /usr/bin/qelectrotech) ==2510== by 0x4DA0C8: ??? (in /usr/bin/qelectrotech) ==2510== by 0x4DCAD4: ??? (in /usr/bin/qelectrotech) ==2510== by 0x4DE840: ??? (in /usr/bin/qelectrotech) ==2510== by 0x435E70: ??? (in /usr/bin/qelectrotech) ==2510== by 0x736B78F: (below main) (in /usr/lib/libc-2.21.so) ==2510== transaction began for "custom" transaction commited for "custom" ==2510== ==2510== HEAP SUMMARY: ==2510== in use at exit: 386,941 bytes in 4,515 blocks ==2510== total heap usage: 118,779 allocs, 114,264 frees, 29,393,780 bytes allocated ==2510== ==2510== LEAK SUMMARY: ==2510== definitely lost: 2,144 bytes in 18 blocks ==2510== indirectly lost: 66,535 bytes in 722 blocks ==2510== possibly lost: 9,084 bytes in 87 blocks ==2510== still reachable: 309,178 bytes in 3,688 blocks ==2510== suppressed: 0 bytes in 0 blocks ==2510== Rerun with --leak-check=full to see details of leaked memory ==2510== ==2510== For counts of detected and suppressed errors, rerun with: -v ==2510== Use --track-origins=yes to see where uninitialised values come from ==2510== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 1 from 1)

Yes, I've also tried `LC_ALL="en_US" qelectrotech` nothing changed - program language is always french, that is strange I have nothing to do with french in my system. There are and were only two messages in terminal: $ qelectrotech or $ LC_ALL=... qel... or LANG=... qel... transaction began for "custom" transaction commited for "custom" $ locale -a C cs_CZ.utf8 en_US.utf8 POSIX ru_RU.utf8