[00:13] * Chipaca EODs [08:06] ｇｏｏｏｏｏｏｄｍｏｒｎｉｎｇ！ [11:08] ¡Muy buenos días a todos! [12:23] Chipaca, please remember https://github.com/canonical/charmcraft/pull/89 [12:23] PR charmcraft#89: Store release [12:25] Chipaca, jam, is there a way to "re-send the work to Travis from GH"? [12:25] I need to do that here: https://github.com/canonical/charmcraft/pull/89 [12:25] PR charmcraft#89: Store release [12:25] facubatista: let me look [12:25] if you log into travis you can ask it to rebuild, IIRC [12:25] jam, I don't see the work in travis [12:26] s/work/job/whatever [12:26] facubatista: https://travis-ci.org/github/canonical/charmcraft/builds/710773290 ? [12:26] or is that the wrong commit [12:27] Chipaca, that it's not listed here: https://travis-ci.org/github/canonical/charmcraft/builds [12:27] facubatista: https://travis-ci.org/github/canonical/charmcraft/pull_requests [12:28] oh [12:28] the commit hashes don't match so i dunno [12:29] but it's got two builds 22 hours ago, which matches the two commits from 22 hours ago :) [12:31] if I tell it to "see this commit in github" I see a diff like looks like my PR, but with a message "This commit does not belong to any branch on this repository." [12:31] so it looks it's a travis commit? [12:31] anyway, I'll trigger a re-build of one of the jobs to see if that pings GH correctly [12:31] did you rebase something? [12:38] jam, nop [12:39] meanwhile, this: https://github.com/canonical/charmcraft/pull/93 [12:39] PR charmcraft#93: Store status [12:39] (but note that it depends on 89 [12:39] ) [12:44] facubatista, is there anything blocking 89 other than @Chipaca ? [12:44] nop [12:45] * Chipaca blocks all the things [12:47] Chipaca, if you answer "ehhhh... maybe? :-)" in a review conversation it makes me try to find the new commit changes to see if you changed or not that particular line [12:47] Chipaca, it's better IMO if you just say "no" (and continue talking) or "yes" (and I can assume you changed as suggested) [12:59] facubatista: my apologies. You are right. [12:59] Chipaca, no problemo :) [12:59] facubatista: I changed it to a warning, but I don't know how much sense it makes, because I am not able to create the conditions that would cause that line of code to be exercised [13:06] Chipaca, ack [13:07] facubatista: there are a couple of 'ifs' there that aren't covered because the docs say they can happen but not how :) [13:09] heh, jam, good idea about filtering, why didn't i look at that [13:10] Chipaca, for the OSError that we want to log? [13:10] ye [13:19] PR operator#361 opened: ops/testing.py: Basic support for checking pod spec [13:33] facubatista, jam, we standupping? [13:33] Chipaca, I'm sitdowning [13:34] going [13:54] we're going to grow a shared lib between ops and charmcraft aren't we [13:54] facubatista, reading through your outline I think it flows fairly well. [13:54] obviously needs fleshing out, but the basic flow seems good [13:54] jam, thanks! [14:03] facubatista: venv is not always there fwiw [14:03] why am i replying to reviews on irc [14:03] * Chipaca chastises himself [14:03] Chipaca, because you want immediate feedback ? [14:06] we can talk here or in a HO about some specifics from the review (higher bandwidth), but I'd need more context [14:15] facubatista, jam, do neither of you have your name in gecos? [14:18] it worked in my machine [14:18] that's: getent passwd $USER | cut -d: -f5 | cut -d, -f1 [14:18] fwiw [14:44] Chipaca, jam, store-status is formally ready for review (store-release landed and merged into it) [15:02] facubatista: WDYT of https://paste.ubuntu.com/p/8NqQx9mp9P/ ? [15:06] Chipaca, I'm not sure... we do need to add a "long help", but not sure if the docstring is the proper place to put it. [15:06] we should dedent that at least [15:07] the default formatter does that already [15:07] which we're replacing [15:07] we are? [15:07] we are [15:07] ah [15:07] there's no way to conform to Canonical specifities using argparse help constructor [15:07] facubatista: the one we have does dedent though :) [15:37] facubatista, jam, virtualenv in the template README: env, or .env? [15:37] env [15:37] k [15:38] I don't think we should create hidden stuff if possible [15:38] Chipaca, I've always done venv [15:38] I think that is the default if you just "python3 -m venv" [15:39] jam: ENV_DIR is required for venv as well [15:40] OTOH 'env' might confuse people into thinking it's environment, not virtualenv-ish [15:43] I'm fine with it just being visible [15:44] visible is a terrible name for a virtual environment directory [15:44] jajaj [15:44] jam: i just hit the 'flake8 checks all the files in the venv' thing [15:44] adding a .flake8 :) [15:45] Virtual Included Super Important Basic Large Environment [15:50] \o/ [16:14] facubatista: what's the pytest equivalent of assertRaises? [16:15] Chipaca, pytest.raises() [16:15] Chipaca, it have a `match` arg where you can pass a regex to check for the exception message [16:16] I've been comparing the exact text in a lot of places by doing `str(cm.value)`, though [16:16] (cm is the context manager) [16:16] with pytest.raises(ValueError) as cm: [16:21] Chipaca, for the `init` command, what about a "versions" file? [16:27] facubatista: ? [16:27] facubatista: what do you mean a versions file? [16:27] Chipaca, a file named 'version' [16:27] facubatista: that sounds like a bad idea to me :) [16:27] facubatista: because i'd expect 'charmcraft build' to generate that [16:28] facubatista: as per charmcraft#37 [16:28] Issue charmcraft#37: Populate the 'version' file as part of 'charmcraft build' [16:28] Chipaca, ok :) [16:30] Chipaca, mmm... however, what we'll put in the 'automatic version file'? we only have the git hash, users will want the version to be 1.0, 1.1 [16:31] I'm +1 for build to produce something if not there, but a good formed project probably should provide its own information there [16:34] facubatista: sounds like something we should figure out soon :) maybe comment on that bug? [16:41] done [17:34] facubatista: got a sec? [17:35] Chipaca, got two! [17:35] facubatista: standup meet plz [18:04] Chipaca, "series" in the metadata.yaml file is mandatory for the Store [18:14] Chipaca, jam, updated https://gist.github.com/facundobatista/e279a6e5adac93bb6b048cbf5a9fe96f [18:14] Chipaca, jam, unless "English corrections" or any specific comment from you, it would be done [18:50] Chipaca, also proposed https://github.com/canonical/charmcraft/pull/94 [18:50] PR charmcraft#94: Store commands autocomplete [22:32] facubatista: is the store really not answering anything other than 'rejected' when it doesn't like an upload? [22:53] facubatista: charmcraft writes everything to stederr :-( [22:53] * Chipaca files a bug [22:54] Chipaca, they landed an improvement these days [22:54] Chipaca, we can fix the issue I opened for when they provide more info [22:54] k [22:54] Chipaca, regarding stderr, we need to decide what we'll send to stdout and stderr, check the "messages to the user" doc [22:55] facubatista: the expected output of a command is never stderr [22:56] Chipaca, what about errors? [22:56] anyway, the doc [22:56] * facubatista is not really here [23:12] * Chipaca EODs