Opened 5 months ago
Last modified 3 weeks ago
#70274 new defect
gtk3 @3.24.41_2+quartz: review active_variants check against glib2
| Reported by: | cooljeanius (Eric Gallager) | Owned by: | |
|---|---|---|---|
| Priority: | Normal | Milestone: | |
| Component: | ports | Version: | 2.9.3 |
| Keywords: | Cc: | mohd-akram (Mohamed Akram), mascguy (Christopher Nielsen) | |
| Port: | gtk3 |
Description (last modified by ryandesign (Ryan Carsten Schmidt))
Basically, see discussion on commit [d0aca0012c93f2bf9c7f83c46d15040ad2b83d78/macports-ports]. I can understand doing the variant check one way around for the missing symbol, but that doesn't necessarily mean it's necessary to do it the other way around, too. I have gtk3 installed with +quartz and glib2 installed with +x11 and this is breaking upgrading of all of my gnome ports. Getting issue #63680 fixed would probably resolve this.
Change History (10)
comment:1 follow-up: 3 Changed 5 months ago by mohd-akram (Mohamed Akram)
comment:2 Changed 5 months ago by ryandesign (Ryan Carsten Schmidt)
| Description: | modified (diff) |
|---|
comment:3 follow-up: 4 Changed 5 months ago by cooljeanius (Eric Gallager)
Replying to mohd-akram:
Just install glib2 +quartz?
I have no way of knowing if that might break whichever port might have required me to install glib2 +x11 in the first place, though...
comment:4 Changed 3 months ago by cooljeanius (Eric Gallager)
Replying to cooljeanius:
Replying to mohd-akram:
Just install glib2 +quartz?
I have no way of knowing if that might break whichever port might have required me to install glib2 +x11 in the first place, though...
Update: it's possible that I only have glib2 installed with +x11 just because I installed it prior to [45c59fe0a66e382927fbf6fe629d37898eb42b6b/macports-ports], but even if that's the case, that doesn't mean it's safe for me to remove the variant, though, as some port I installed afterwards could have had an active_variants check for +x11 on it that I've forgotten about... it'd be nice if you could do an active_variants search in reverse, i.e., when viewing a port's variants, see what other ports require those variants... but I guess that'd just be resurrecting #126 again...
comment:5 follow-up: 6 Changed 3 months ago by cooljeanius (Eric Gallager)
Ah, maybe my situation is pulseaudio's fault, since it has a default +x11 variant, and depends on glib2...
comment:6 Changed 5 weeks ago by cooljeanius (Eric Gallager)
Replying to cooljeanius:
Ah, maybe my situation is pulseaudio's fault, since it has a default +x11 variant, and depends on glib2...
...and this in turn might be because of multimon-ng @1.2.0+debug+pulseaudio+x11...
comment:7 Changed 5 weeks ago by cooljeanius (Eric Gallager)
looks like the libepoxy active_variants check added in [a85cc86813b8a49cd1422b5cf7ba01d1711ac623/macports-ports] may require review, too...
(also this applies to gtk3-devel as well as just gtk3)
comment:8 Changed 5 weeks ago by cooljeanius (Eric Gallager)
I opened PR 26062 to address this: https://github.com/macports/macports-ports/pull/26062
comment:9 follow-up: 10 Changed 5 weeks ago by mohd-akram (Mohamed Akram)
Neither multimon-ng nor pulseaudio specifies an active variant for glib2. We should probably figure out what's causing the problem for you before attempting a fix.
comment:10 Changed 3 weeks ago by cooljeanius (Eric Gallager)
Replying to mohd-akram:
Neither multimon-ng nor pulseaudio specifies an active variant for glib2. We should probably figure out what's causing the problem for you before attempting a fix.
What's causing the problem is that unnecessary active_variants checks have been blocking the upgrading of all my gtk-based ports for over 4 months now, leaving my MacPorts installation in an extremely broken state. I am done postponing my attempt at a fix; it's been long enough already.
Just install glib2 +quartz?