| [04:00:09] | * hra38192639[m] has quit (Quit: You have been kicked for being idle) |
| [04:00:09] | * AllanMacGregor[m has quit (Quit: You have been kicked for being idle) |
| [04:00:10] | * steinfarm[m] has quit (Quit: You have been kicked for being idle) |
| [04:00:10] | * kunigunde[m] has quit (Quit: You have been kicked for being idle) |
| [04:00:10] | * harry[m]12 has quit (Quit: You have been kicked for being idle) |
| [04:00:10] | * gunter[m] has quit (Quit: You have been kicked for being idle) |
| [04:00:11] | * JanSonntag[m] has quit (Quit: You have been kicked for being idle) |
| [05:05:55] | <skipper_is[m]> | Nothing seems like it needs elevated privilege |
| [05:05:56] | <skipper_is[m]> | On farmOS 1.x, is there a way of checking what permissions affect a page? I've got a quick form that is only showing up for user 0, and cannot figure out why... |
| [06:46:30] | <mstenta[m]> | skipper_is: in Drupal 7 the access rules are defined in hook_menu |
| [06:48:57] | <mstenta[m]> | Oh but just remembering, for quick forms the permission was defined in hook_farm_quick_forms() |
| [06:49:02] | <mstenta[m]> | Eg: https://github.com/farmOS/farmOS/blob/c25722650406f98349a30f1a28a72a782c... |
| [06:50:23] | <mstenta[m]> | Oh is it possible you don't have any permissions declared in that hook? |
| [10:59:09] | <symbioquine[m]> | Is there interest in the [Dev call](https://farmos.discourse.group/t/interest-in-an-ongoing-programming-deve...) today? I don't have specific topics, but I'm thinking of calling in and seeing if other folks do... |
| [10:59:40] | <mstenta[m]> | i'll be there :-) |
| [10:59:59] | <paul121[m]> | I have a few things on my mind if no one else brings anything :D |
| [11:00:06] | * SpencerOnazi[m] has quit (Quit: You have been kicked for being idle) |
| [11:00:06] | * davd[m] has quit (Quit: You have been kicked for being idle) |
| [11:00:06] | * and712[m] has quit (Quit: You have been kicked for being idle) |
| [11:00:08] | * gretel[m] has quit (Quit: You have been kicked for being idle) |
| [11:00:08] | * mindcls[m] has quit (Quit: You have been kicked for being idle) |
| [11:00:49] | <mstenta[m]> | paul121: maybe i'll see if @SirSundays wants to join :-) |
| [11:01:08] | <mstenta[m]> | (debugging some l10n stuff right now in https://github.com/farmOS/farmOS/issues/472) |
| [11:01:55] | <paul121[m]> | yes! they joined the dev call once before! haha ping Jan Sonntag |
| [11:39:20] | <mstenta[m]> | <mstenta[m]> "symbioquine paul121 : I wonder..." <- symbioquine paul121 I opened a PR for this: https://github.com/farmOS/farmOS/pull/473 |
| [11:39:48] | <mstenta[m]> | Although, I just found this option for `matrix` strategy workflows: https://docs.github.com/en/actions/learn-github-actions/workflow-syntax-... |
| [11:39:55] | <mstenta[m]> | maybe we should just keep them in the same workflow and temporarily set that to `false` instead? |
| [11:40:23] | <mstenta[m]> | it sounds like that would allow the postgres and mariadb tests to continue even if sqlite3 fails |
| [11:40:28] | <mstenta[m]> | which is really the pressing issue... |
| [11:41:48] | <paul121[m]> | interesting. but also important to know if postgres passes but mysql fails! |
| [11:42:19] | <mstenta[m]> | yea, that would still be visible |
| [11:42:57] | <mstenta[m]> | the issue i'm constantly frustrated by is that sqlite3 fails pretty early (when this happens), and i have to keep rerunning the job to see if the other two work |
| [11:44:08] | <paul121[m]> | ah yeah |
| [11:46:23] | <symbioquine[m]> | What about `jobs.<job_id>.continue-on-error`? |
| [11:47:24] | <mstenta[m]> | saw that too... |
| [11:47:43] | <mstenta[m]> | but i think we still do want to see red when sqlite fails |
| [11:47:57] | <mstenta[m]> | otherwise we'll miss actual sqlite test failures |
| [11:50:58] | <symbioquine[m]> | It seems like that would depend how well those "failed, but continued" jobs are represented in the UI... |
| [11:51:13] | <symbioquine[m]> | If they're just shown as "passed" then I agree. |
| [11:51:23] | <mstenta[m]> | mm yea |
| [11:51:26] | <mstenta[m]> | that's my assumption |
| [11:52:06] | <mstenta[m]> | i'm opening another PR with the `fail-fast` approach... |
| [11:57:33] | <mstenta[m]> | https://github.com/farmOS/farmOS/pull/474 |
| [11:57:47] | <mstenta[m]> | it is cleaner than the first PR, which I like |
| [11:57:58] | <mstenta[m]> | and keeps everything in the same file |
| [12:03:41] | <skipper_is[m]> | <mstenta[m]> "Eg: https://github.com/farmOS/..." <- I do indeed have it declared `'permission' => 'create farm_maintenance log entities',` |
| [13:07:04] | <skipper_is[m]> | Also, does anyone use Github Copilot? I just had a go and wow, it is very impressive! |
| [13:55:10] | <paul121[m]> | I have not! how do you try it out? |
| [13:55:44] | <paul121[m]> | also symbioquine I was thinking of the Cloud optimized geotiffs - just remembered: https://www.cogeo.org/ |
| [13:55:49] | <paul121[m]> | s/symbioquine/skipper_is/ |
| [13:56:12] | <paul121[m]> | sorry your internet wasn't working at symbioquine ! |
| [13:56:19] | <skipper_is[m]> | https://copilot.github.com/ |
| [13:56:28] | <symbioquine[m]> | Yeah, it's still dropping in and out :( |
| [13:56:38] | <symbioquine[m]> | I'm connected through my phone right now. |
| [13:56:51] | <skipper_is[m]> | Essentially, you start typing, and it makes suggestions - like any other IDE so far, but the suggestions are like entire functions in some cases... |
| [13:57:55] | <skipper_is[m]> | ACTION uploaded an image: (24KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/ODYlsVGQvWH... > |
| [13:58:30] | <skipper_is[m]> | So because my previous form element was a battery checkbox, it has suggested that my next option should be a textfield to record the battery level |
| [13:58:39] | <mstenta[m]> | oh yea i heard about this! |
| [13:58:42] | <skipper_is[m]> | Like... the entire element has been suggested! |
| [13:59:24] | <mstenta[m]> | man... what is programming going to look like when my kids are grown up 🤯 |
| [13:59:38] | <skipper_is[m]> | Just write a brief description of what you want and it'll do the code... |
| [13:59:46] | <paul121[m]> | omg...... |
| [13:59:58] | <skipper_is[m]> | "Program that records information from the farm" and it'll write the whole of FarmOS 10.x |
| [14:00:43] | <paul121[m]> | but really, "create a {blank} asset type" would really be useful! |
| [14:01:28] | <skipper_is[m]> | The next suggestion after the battery level option was prestart brake test, followed by oil and fuel levels... |
| [14:01:32] | <skipper_is[m]> | How does it know what a prestart check is! |
| [14:04:01] | <skipper_is[m]> | It's also just suggested all the related prestart $form_state[xx] for the submit function as well... |
| [14:04:07] | <skipper_is[m]> | It is terrifying, and will make me a lazy programmmer. |
| [14:04:48] | <paul121[m]> | crazy |
| [15:19:38] | * natewrench123456 has joined #farmos |
| [15:19:59] | * natewrench123456 has quit (Max SendQ exceeded) |
| [16:04:53] | <paul121[m]> | mstenta: what would you think about moving the codesniffer test out to another workflow? for the same reason to same time, unit testing can continue despite codesniffer errors |
| [17:09:59] | <mstenta[m]> | Hmm I don't think it will make much difference - codesniffer takes less than 10 seconds to run, compared to ~8 mins for tests (using this one as an example: https://github.com/mstenta/farmOS/runs/4731417964?check_suite_focus=true) |
| [17:10:36] | <paul121[m]> | s/same/save/ |
| [17:10:49] | <mstenta[m]> | but perhaps we could split it out to a separate job within the same workflow? |
| [17:10:56] | <mstenta[m]> | kinda like we did with the sqlite stuff? |
| [17:11:01] | <paul121[m]> | ah but I mean when the codesniffer fails. it prevents the unit tests from running |
| [17:11:09] | <mstenta[m]> | or i mean: set it up so that it won't stop other stuff |
| [17:11:11] | <mstenta[m]> | ah yea |
| [17:11:52] | <mstenta[m]> | remember the commit i showed on the dev call today? that split the docker build and test matrix into two separate "jobs"? |
| [17:12:13] | <mstenta[m]> | maybe we could move the codesniffer to it's own job too, which runs in parallel with the tests, but tests don't depend on it |
| [17:13:12] | <paul121[m]> | right, because that needs the dev image |
| [17:13:26] | <mstenta[m]> | ah yea |
| [17:13:47] | <mstenta[m]> | i can try that out in my `2.x-docker` branch |
| [17:13:54] | <mstenta[m]> | probably pretty simple |
| [17:23:28] | <mstenta[m]> | https://github.com/mstenta/farmOS/commit/a1770c44edb8c9cba7470a415b7c96b... |
| [17:23:56] | <mstenta[m]> | testing now (the `docker run` command might need some adjustment, we'll see...) https://github.com/mstenta/farmOS/actions |
| [17:25:36] | <mstenta[m]> | * see...) https://github.com/mstenta/farmOS/actions/runs/1665151218/runs/1665151218 |
| [17:25:53] | <mstenta[m]> | s/a1770c44edb8c9cba7470a415b7c96b605a29925/137d967f92a441a1df7ec3315e78ecb472d5843c/ |
| [17:26:01] | <mstenta[m]> | pretty neat how GitHub visualizes it: |
| [17:26:11] | <mstenta[m]> | ACTION uploaded an image: (19KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/AethPQXVAxb... > |
| [17:34:23] | <mstenta[m]> | Hmm one drawback to this approach: I'm finding that the uploading/downloading of the artifacts between jobs is a non-trivial step... it took 2 min 5 s to upload(!!) and ~1.5 min to download and load it in the next job |
| [17:36:28] | <mstenta[m]> | i wonder if that will be like that consistently |
| [17:41:05] | <mstenta[m]> | seems like the upload took a long time just to get started (kept showing 0% for a long time) |
| [17:42:15] | <paul121[m]> | oh interesting |
| [17:42:42] | <mstenta[m]> | oof yea it's pretty painful how slow this is |
| [17:45:12] | <mstenta[m]> | it works though! https://github.com/mstenta/farmOS/runs/4732982774?check_suite_focus=true |
| [17:46:49] | <mstenta[m]> | oof but it adds almost 4 minutes to the run! 😳 |
| [17:46:55] | <mstenta[m]> | that's crazy |
| [17:49:28] | <mstenta[m]> | that might be a dealbreaker for this strategy |
| [17:51:20] | <mstenta[m]> | well... ok... so i just realized that we aren't compressing the docker image tarballs... 🤦 |
| [17:51:36] | <mstenta[m]> | i'll give that a try and see if it makes a difference... might help, but probably not that much |
| [17:52:41] | <paul121[m]> | yeah, reading into it a little bit too, and it seems like they compress everything already |
| [17:52:55] | <mstenta[m]> | oh |
| [17:53:54] | <mstenta[m]> | (pushed a commit to test anyway) |
| [18:30:54] | <paul121[m]> | that's a bummer tho :-( |
| [18:30:56] | <paul121[m]> | looked great |
