Skip to contentSkip to conversation

AUTOREV with SRCREV_module? (Re: [OE-core] [PATCH] u-boot: set git repo branch with SRCBRANCH)

DISCUSSION Mikko Rapeli 2 participants 3 messages Apr 4, 2025 1 message not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw

Hi,

poky-bleeding.bbclass only changes main SRCREV to AUTOREV but not SRCREV_module etc variables.

How could support for setting all SRCREV_module variables be implemented?

I tried to grep for some clues how to get all SRCREV_* variables and change them, but didn't find any. I think this should be rather simple.

Code only needs to find all SRCREV_* variables and change them to AUTOREV. I'm missing something to find examples of this...

Cheers,

-Mikko

#2 lore raw

I have to note that poky-bleeding is not tested in any way, so you probably first want to stop using it, or otherwise propose a test plan.

Alex

Quoted 29 lines · Mikko Rapeli via lists.openembedded.org

Hi,

poky-bleeding.bbclass only changes main SRCREV to AUTOREV but not SRCREV_module etc variables.

How could support for setting all SRCREV_module variables be implemented?

I tried to grep for some clues how to get all SRCREV_* variables and change them, but didn't find any. I think this should be rather simple.

Code only needs to find all SRCREV_* variables and change them to AUTOREV. I'm missing something to find examples of this...

Cheers,

-Mikko

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#214340): https://lists.openembedded.org/g/openembedded-core/message/214340
Mute This Topic: https://lists.openembedded.org/mt/112082024/1686489
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

#3 lore raw

Hi,

Quoted 3 lines · Alexander Kanavin

I have to note that poky-bleeding is not tested in any way, so you probably first want to stop using it, or otherwise propose a test plan.

poky-bleeding.bbclass could have some trivial tests. I can try to draft them. I don't expect oe-core to test it with real SW components. I plan to use it with some SW components I care about. Finding and fixing issues from bleeding edge pre-release builds can help improve releases and reduce maintenance and integration work later on.

This could be more work, or the same amount of work but just done by developers closer to the SW component upstream which could help distro maintainers. I find poky-bleeding.bbclass useful.

FWIW, my question seems to be answered by SRCREV_FORMAT and patch will be:

meta-poky/classes/poky-bleeding.bbclass +4 -0
@@ -12,6 +12,10 @@ python pokybleeding_version_handler () {
         if "pseudo" in bpn:
             bb.warn("Here 5 %s %s" % (d.getVar("PN"), bpn))
         d.setVar("SRCREV", "${AUTOREV}")
+        srcrev_format = d.getVar("SRCREV_FORMAT")
+        if srcrev_format:
+            for multi_scm in d.getVar("SRCREV_FORMAT").split("_"):
+                d.setVar("SRCREV_%s" % multi_scm, "${AUTOREV}")
         if "+git" not in d.getVar("PV"):
             d.appendVar("PV", "+git")
 }

Cheers,

-Mikko