[01:24] <sergiusens> zyga: consider that we both run the same snap and did "snap run --shell", if it had been staged, it should be failing for me too
[05:52] <zyga> sergiusens: I must have misunderstood then; I will check later
[05:57] <brlin> Found it mildly interesting https://www.irccloud.com/pastebin/TZuveJwP/
[06:33] <brlin> I need someone to proofread https://forum.snapcraft.io/t/system-options/87#heading--refresh-hold
[06:45] <zyga> hey brlin
[06:45] <zyga> how are you doing?
[06:48] <zyga> brlin: so does refresh hold set the time or a boolean flag?
[06:48] <zyga> brlin: I think having an example would be useful
[06:54] <brlin> zyga: It set the time in a certain RFC format AFAICT, I'll craft a example.
[07:40] <zyga> brlin: thank you
[07:40]  * zyga is very sleepy today
[11:45] <zyga> pedronis: the seed is indeed broken
[11:46] <zyga> pedronis: I poked will cooke with the details
[11:46] <zyga> pedronis: since it is snapd seeding itself that is affected sending a fixed snapd to the core snap won't help as snapd keeps panicking
[11:46] <zyga> pedronis: we would need to update snapd via the debian package to address this
[11:47] <zyga> pedronis: with regards to seed damage: I would suggest that we do ignore nil values and carry on as I implemented
[11:47] <zyga> this has higher chance of actually working than snapd perpetually erroring on the seed process
[11:48] <zyga> alternatively desktop should fix the seed file post-install
[11:48] <zyga> not great 19.04 release though, I hope it's just from alpha images
[15:03] <luciom> Hello, I am trying to create a snap for gtk-gnutella. This is the script: https://github.com/luciomarinelli/gtk-gnutella/blob/master/snap/snapcraft.yaml the compile runs well, however what I get is an empty snap package. It looks like the "make install" step is not performed. What am I missing?
[16:46] <brlin> luciom: The part is using custom build script, the autotools plugins specify null prefix and set the DESTDIR to point to SNAPCRAFT_PART_INSTALL by default but that customization won't apply to custom build script.  Replace the build script with equivalent commands (aside from those autotools plugin run for you, which can be triggered by running `snapcraftctl build`) in the override-build scriptlet or reimplement the entire build
[16:46] <brlin> step entirely
[16:47] <brlin> s/step entirely/step completely/
[16:47] <brlin> You may check out what command the autotools plugin are running at the source: https://github.com/snapcore/snapcraft/blob/master/snapcraft/plugins/autotools.py
[17:05] <luciom> brlin, hello if I remove the override-build part i get the error: "Failed to run 'autoreconf -i' for 'gtk-gnutella': Exited with code 1. Verify that the part is using the correct parameters and try again."
[17:06] <cmatsuoka> luciom: hi, I didn't see you were here and answered in the forum post
[17:07] <cmatsuoka> luciom: it seems that gtk-gnutella doesn't like DESTDIR and uses INSTALL_PREFIX instead
[17:07] <cmatsuoka> luciom: I added a suggestion there
[17:08] <luciom> cmatsuoka, thanks I just read your forum post, I am going to try it out
[17:11] <luciom> cmatsuoka, if I add the override-build string you suggest it gives me error "Issues while validating snapcraft.yaml: mapping values are not allowed in this context on line 14, column 21" which refers to the | symbol
[17:12] <cmatsuoka> luciom: I just tried with this yaml file: https://pastebin.ubuntu.com/p/jVyPBQWckG/
[17:12] <cmatsuoka> it builds the package, but it still has many runtime libraries missing
[17:17] <cmatsuoka> luciom: added one more note about SDL to the forum post
[17:17] <cmatsuoka> I don't think gtk-gnutella needs SDL
[17:24] <luciom> cmatsuoka, I used your script in pastebin and the process went forward, the snap is no longer empty!
[17:26] <cmatsuoka> \o/
[17:55] <luciom> cmatsuoka, it built without errors on my virtual machine, however on spancraft.io still errors: https://build.snapcraft.io/user/luciomarinelli/gtk-gnutella/534538
[18:06] <cmatsuoka> luciom: I can't see the log, could you paste the error message somewhere?
[18:09] <cmatsuoka> gotta go, will be back later