Tim Harder
Tim Harder
> cargo-c provides `-uninstalled.pc` files for that purpose, they have to be used explicitly as `--static` thought. Well it didn't seem like it needed that since things still built, at...
Sounds fair, you can close this if you want then.
Currently the pkgcore resolver doesn't handle most of the cases involving blockers in an automatic fashion. In other words, if it hits a blocker for a certain version usually it...
It's on my todo list that I'm slowly ticking through as my time/motivation allows.
At a quick glance this appears to relate to a key value of '_mtime_' being written instead of 'MTIME' for recently created binpkgs which then try to get fixed back...
Note that both of those commits are to my private repo and also don't fully fix the issue yet.
pkgcheck now includes REQUIRED_USE checking so issues like that should be flagged by CI setup, e.g. https://qa-reports.gentoo.org/output/gentoo-ci/77e4233/output.html#dev-python/kombu. Hopefully this will get people to conform to our standards. :smile:
Yeah, I'm not overly fond of the current config design in certain places. Perhaps @ferringb can explain some of the history/reasons behind this.
I forgot to mention another point. Normally when using the various commandline tools such as pmerge, the exception chains created by raise_from() will be shown "properly" by dump_error() from pkgcore.util.commandline....
How would doing sys.excepthook in pkgcore.util.commandline or pwrapper handle raw API usage cases, e.g. https://github.com/pkgcore/pkgcore/issues/82 ?