*** zbenjamin_ is now known as zbenjamin | 01:29 | |
*** frinring_ is now known as frinring | 07:03 | |
sledges | PSA: SFOS collaboration community meeting at #mer-meeting on the hour (in ~13mins) | 07:42 |
---|---|---|
*** ecloud is now known as ecloud_wfh | 07:46 | |
KeeperoftheKeys | I missed the meeting right :/ | 10:30 |
KeeperoftheKeys | how do I change the RPM being built from a arch to noarch (python only) | 18:30 |
KeeperoftheKeys | the sdk complains that it should be built as noarch but only has options for i486 or arm as far as I can tell | 18:31 |
bionade24 | KeeperoftheKeys: Isn't noarch just a specification in the .spec file? | 18:42 |
mal | yes, it's just how you define the package in spec | 18:44 |
mal | you set "BuildArch: noarch" for the package in spec | 18:45 |
mal | KeeperoftheKeys: ^ | 18:45 |
KeeperoftheKeys | wait but the spec file is generated based on the .pro and the .yaml so where do I set it in those? | 18:54 |
KeeperoftheKeys | (thanks btw) | 18:54 |
bionade24 | KeeperoftheKeys: Most people use the .spec file directly, especially if it's a gnu make or cmake based project. Sorry, I don't know how to this in the .yaml. but should be similar. | 18:56 |
KeeperoftheKeys | OK I'll try again, before I had the spec file that shipped with gpodder and then the SDK would fail because it detected that it was not an autogenerated spec file | 18:59 |
bionade24 | KeeperoftheKeys: You probably can turn this off somewhere | 19:04 |
*** Venemo_XA2_ is now known as Venemo_XA2 | 19:17 | |
KeeperoftheKeys | How about preventing python bincode to be included? | 20:16 |
bionade24 | KeeperoftheKeys: What should that help? It's not binary code, it's bytecode and works with every interpreter (like in java) | 21:12 |
KeeperoftheKeys | just smaller less cluttered package | 21:13 |
bionade24 | KeeperoftheKeys: Really that much smaller? Don't believe so, only faster in execution first time. And that matters. | 21:15 |
KeeperoftheKeys | I have 2 additional files for each .py file and the package as generated without bytecode is ~100k vs ~200k for with bytecode | 21:16 |
KeeperoftheKeys | (.pyc and .pyo) | 21:16 |
KeeperoftheKeys | (uncompressed size ~400k vs ~700k | 21:17 |
bionade24 | KeeperoftheKeys: In today's times completely irrelevant | 21:19 |
KeeperoftheKeys | call me OCD but I feel like I'm shipping users a bag full of garbage if I'm sending them bytecode... | 21:20 |
KeeperoftheKeys | anyhow found it :) | 21:21 |
bionade24 | KeeperoftheKeys: Just be shure then that the .pyo and .pyc files get generated when a user runs it the first time, sometimes there are issues | 21:25 |
bionade24 | KeeperoftheKeys: When you're OCD you should use python wheels with zstd compression for packaging, then you would have <100K | 21:26 |
KeeperoftheKeys | Why I just want to distribute a set of python scripts? without a whole load of clutter | 21:28 |
bionade24 | KeeperoftheKeys: I just wanted to optimize your ideas :) | 21:29 |
KeeperoftheKeys | Why isn't Sailfish.Media 1.0 allowed? | 21:32 |
bionade24 | KeeperoftheKeys: Sailfish Media API get's stabilized soon and then it'll be allowed | 21:35 |
bionade24 | KeeperoftheKeys: You can read about it in the last #mer-meeting log | 21:35 |
KeeperoftheKeys | great :) | 21:44 |
KeeperoftheKeys | I only see KeepAlive mentioned... oh well :/ | 21:49 |
KeeperoftheKeys | stupid question maybe but what is the "preferred" method for changelogs in sfos apps? a markdown file? textfile? in the yaml for RPM? through harbour somehow? | 22:13 |
KeeperoftheKeys | when RPM validator prints a dependency error does it always first print a line which seemingly says that /bin/sh is required by the project? | 23:42 |
KeeperoftheKeys | (It's just that I never gave it such a dependency and it also is not in any file in the RPM as far as I can tell using grep) | 23:43 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!