Aaron Puchert

See the [Wiki page about Package dependencies](https://en.opensuse.org/openSUSE:Package_dependencies#OBS_Caveat): > [W]hen evaluating BuildRequires, the RPM parser of the BuildService runs in an environment where subshell expansions are disabled. You'll see warnings like...

> Macro expansion is delayed Specifically of `%define`. With `%global` the macro should be expanded once at the place of definition. [Some guides recommend](https://docs.fedoraproject.org/en-US/packaging-guidelines/#_global_preferred_over_define) the latter over the former for...

> Note that [#7591 (comment)](https://github.com/openSUSE/open-build-service/pull/7591#issuecomment-525836832) explains why "everything in one page" is not a good thing. That's indeed ugly, but ~100 builds isn't all too common, I'd say. And perhaps...

> build results are just useless to be honest: the request page shows you the build results of HEAD (which may be 'blocked') not of the sources submitted. Their usefulness...

> > several users expressed that they would wish to prevent or delay BOINC startup as a service to allow prerequisites such as graphics drivers and WiFi networking to initialise...

It looks like memory corruption, and indeed with AddressSanitizer: ``` ERROR: AddressSanitizer: stack-buffer-overflow on address 0x7ffc15f71b30 at pc 0x7ff221a20ae9 bp 0x7ffc15f717c0 sp 0x7ffc15f717b8 READ of size 3 at 0x7ffc15f71b30 thread...

I think I've found the issue. In `hasMethodLike` from clang/lib/Analysis/LifetimeTypeCategory.cpp we use `CXXRecordDecl::forallBases` to determine if any base has a certain `CXXMethodDecl`. To that end, we pass a lambda that...

> I'm not sure what the proper solution is, because it seems like we want to look into templates here. Maybe we should just ignore the return value of `forallBases`...

I got the same error. As the message says, the module `QtQuick.XmlListModel` is missing. The [Qt docs](https://doc.qt.io/qt-5/qtquick-xmllistmodel-qmlmodule.html) tell us that the module is part of “Qt XML Patterns”, so I...

See also #71, which is exactly the same issue.