Arch Linux User Repository

The next person who discusses keygens will have their account suspended indefinitely.

This company sucks. I have the paid license but I can't use it because it's only for 1 computer and even though I haven't changed computers it doesn't let me activate it in the master-pdf-editor-5.4.38-qt5-all.amd64.tar.gz version. I have already complained to this company several times. But it's no use if I'm the only one complaining.

@scimmia, sorry, didn't intend to make crazy accusations, and did take a look at the PKGBUILD, so stand corrected -- my bad. I read through the AUR comments here but failed to notice that the vendor supplies 2 .tgz files:

master-pdf-editor-5.4.38-qt5.amd64.tar.gz and master-pdf-editor-5.4.38-qt5-all.amd64.tar.gz

The "-all" version includes the Qt binaries, and you must run masterpdfeditor5.sh in order to correctly set the LD_LIBRARY_PATH.

So, updating my question, why not base the AUR package on master-pdf-editor-5.4.38-qt5-all.amd64.tar.gz to insure better stability?

@j1simon there is no problem at the moment. Just use latest tarball.

pgoetz, if you're going to make crazy accusations, you should at least be able to read the PKGBUILD and see what it actually does.

Hmmm, well yes. Since I paid for the software, I did complain and they just directed me to the .deb and .tgz files (the response from CI support was impressively speedy, too). The issue is that Code Industry bundles their own version of Qt which one gets with either the .deb or the .tgz. I believe the Qt components are removed in the AUR package, which is probably a bad idea; in particular because the newest version of Qt isn't compatible with the newest version of masterpdfeditor. Everything is moving to containers because of increasing software complexity, and the Code Industry solution is kind of a container-lite solution. They probably will update the version of Qt they're using, but it's pretty clearly not a priority; meanwhile I've got pdf's which need to be looked at. I've been using the masterpdfeditor-free AUR package, which is based on masterpdfeditor v.4, but am getting tired of waiting and will probably just uninstall both AUR packages and just install the .tgz myself.

Note to the package maintainer: the Arch philosophy is to mess with upstream as little as possible. Doesn't removing included libraries constitute messing with upstream? Just curious what the reasoning is here in the era of 10TB hard drives.

The thing to do is to complain to the developers of this program: support@code-industry.net Talking here is not the solution to this problem.

@dibodov : Thanks I have rebuilt the package localy with the archive https://code-industry.net/public/master-pdf-editor-5.4.38-qt5-all.amd64.tar.gz and everything works fine. You do not need to use the bash script as the desktop file is modified to launch it directly.

@Scimmia : could you issue a pkrel "2" with the right archive?

You can use either .deb or tarball:
https://code-industry.net/public/master-pdf-editor-5.4.38-qt5-all.amd64.tar.gz
https://code-industry.net/public/master-pdf-editor-5.4.38-qt5-all.amd64.deb

Both ship with Qt binaries.

Use this shell file:
/opt/master-pdf-editor-5/masterpdfeditor5.sh

As far I can tell the undefined symbol '_ZN10QMutexPool17globalInstanceGetEPKv' refers to this method QMutexPool::globalInstanceGet(void const*). The class QMutexPool seems to be not part of the QT API. Looking at qtbase source code the class QMutexPool was removed when moving from 5.13.2 to 5.14.0. Obviously, It is not good practice to use a method that is not part of the API. This is somewhat disconcerting.

The master-pdf-editor-5.4.38-qt5-all.amd64.deb ships with Qt 5.9.8 and therefore should run fine. The tarball used for this package "master-pdf-editor-5.4.38-qt5.amd64.tar.gz" however ships without qt.