x11/libfm: fix x11-fm/pcmanfm segfault on start.- Add core SLAVEPORT

[rigoletto]

Summary

PR 234814

The last pcmanfm version is segfaulting for some users on start. The submitted patch fix this issue and also improve the port with a core FLAVOR for -qt versions.

Can core be a FLAVOR or should it be splitted in a SLAVEPORT?

Test Plan

Test it with sysutils/bsdisks.

[rigoletto]

Updating patch as per submitter.

[tcberner]

libfm-gtk should probalby start as an svn cp of libfm.

x11-fm/pcmanfm/Makefile

19

^ maybe its time to swtich 3 to default too?

[duchateau.olivier_gmail.com]

Make Gtk3 default flavor is bad idea, because its support is not good shape. There are lots of warnings due to deprecated methods. Moreover import of exo (from Xfce project) is very old too (taken from 0.10.x branch, upstream needs to use the 0.12.x branch, it is compatible with both version).

[tcberner]

In D18840#403614, @duchateau.olivier_gmail.com wrote:

Make Gtk3 default flavor is bad idea, because its support is not good shape. There are lots of warnings due to deprecated methods. Moreover import of exo (from Xfce project) is very old too (taken from 0.10.x branch, upstream needs to use the 0.12.x branch, it is compatible with both version).

OK, in that case, strike that :)

[rigoletto]

Do svn cp of libfm.

[rigoletto]

Everything building now, and Qt related ports bumped. \o/

[mat]

Mmmm, why create a slave port instead of adding a first, say, core, flavor to libfm?

[rigoletto]

In D18840#404565, @mat wrote:

Mmmm, why create a slave port instead of adding a first, say, core, flavor to libfm?

That was exactly the doubt. The first diff version was flavor only, but I wasn't sure if core could be a flavor. ^^

[mat]

The only string a flavor cannot be is all. Otherwise, it can be anything maching /[a-z0-9_]/. See Mk/bsd.port.mk lines 1474-1486.

[mat]

The only thing that comes to mind that might not work and require a slave port is the dependency of some flavors to another, but I don't think it is the case, I don't remember if I ever tested that.

[rigoletto]

In D18840#404669, @mat wrote:

The only thing that comes to mind that might not work and require a slave port is the dependency of some flavors to another, but I don't think it is the case, I don't remember if I ever tested that.

Yes, we had a problem with it in D16387, reminding:

In D16387#348253, @mat wrote:

It seems you forgot to add revert the pkg-plist deletion.

In D16387#348109, @lbdm_privacychain.ch wrote:

Done.

Converting libfm-extra to FLAVOR (libfm@extra) resulted on:

Warning: (x11/libfm@gtk3): Error: x11/libfm incorrectly depends on itself.

Thanks! ^^

This does not look like a ports framework message, but a poudriere one. Poke @bdrewery.

[rigoletto]

However my doubt was more related with the concept differences of flavors and sub-packages (slaveports in practice), like happened with transmission-* when flavoring everything (cli, daemon, gtk, qt, web) in a single port instead of separated slaveports for cli, daemon, gtk versions, qt versions, and web in D16387.

Thanks!

[duchateau.olivier_gmail.com]

In D18840#404669, @mat wrote:

The only thing that comes to mind that might not work and require a slave port is the dependency of some flavors to another, but I don't think it is the case, I don't remember if I ever tested that.

This is why I created slave port for core, otherwise poudriere complains (GTK flavors requires GLib/Gio core component of libfm).

[rigoletto]

With -qt versions now being unrelated there is no need for this anymore.

Tank you.