| Summary: | gnome2.eclass: call ${EROOT}${GLIB_COMPILE_SCHEMAS} from eclass | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Pacho Ramos <pacho> |
| Component: | [OLD] GNOME | Assignee: | Gentoo Linux Gnome Desktop Team <gnome> |
| Status: | RESOLVED WONTFIX | ||
| Severity: | normal | CC: | esigra |
| Priority: | Normal | Keywords: | NeedPatch |
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Bug Depends on: | |||
| Bug Blocks: | 262490 | ||
Well the idea is that a couple of ebuilds are using locally compiled schemas to run unittests and it should be an eclass thing. It is currently not the case. I haven't explored much how this could be done so if anyone has time for this, just propose your patch. The main issue I was seeing is that looks like we need to specify a folder for looking to relevant schemas, but that folder changes on every affected package then... I still don't see how this could be "automatized" a bit more to be moved to the eclass (probably as a default src_test phase) :/ I think I changed my mind on this. I saw something in a build.log yesterday that made me think that upstream should actually provide a rule that does this job since it's this is the proper way to run unittests (not touching live system). This is the kind of problem that should be seen on make distcheck on a system without the schema installed yet. I'll see how I can come up with such patch for affected packages. Is this wontfix then? :) (for eclass I mean) Yes, let's not clutter ourselves with bugs that don't need actions from us directly. |
I have seen multiple ebuilds with: # FIXME: this should be handled at eclass level "${EROOT}${GLIB_COMPILE_SCHEMAS}" --allow-any-name "${S}/schema" || die But I don't know how are we expected to call it from eclass