=== mwhudson_ is now known as mwhudson === mwhudson_ is now known as mwhudson === matsubara-afk is now known as matsubara === mrevell is now known as mrevell-lunch === mrevell-lunch is now known as mrevell [14:01] * sinzui looks about [14:01] #startmeeting [14:01] hello everyone and welcome to this week's ameu reviewers meeting. who's here today? [14:01] Meeting started at 09:01. The chair is sinzui. [14:01] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [14:02] me === flacoste_afk is now known as flacoste [14:02] me [14:02] me [14:02] hmm, we may be an hour early [14:02] me [14:02] me [14:02] me [14:02] sinzui, I don't think so [14:02] me [14:02] me [14:03] me [14:03] danilo has a bed headache so he might not come. [14:03] me [14:03] s/bed/bad/ [14:03] dunno if he's in bed ... [14:04] we all have laptops. I assume that some of you are attending this meeting from your bed in you pajamas [14:05] [TOPIC] agenda [14:05] * Roll call [14:05] * Action items [14:05] * UI review call update [14:05] * Peanut gallery [14:05] New Topic: agenda [14:05] pyjamas are a redundant nicety [14:05] [Topic] Action items [14:05] New Topic: Action items [14:05] * barry to get with mrevell on guidelines migration from old wiki to new [14:05] ^ I do not think this has happened [14:06] No, it hasn't. My apologies, it's still on my "to do" [14:06] mrevell: thanks [14:06] I will keep it on the agrenda [14:06] * intellectronica and barry to draft guidelines for drive-by cleanups [14:06] sorry, i didn't find the time to do this. please keep it on the list, i will do it next week when there's a bit more time [14:06] I will [14:06] [Topic] UI review call update [14:06] New Topic: UI review call update [14:07] Gosh this weeks meeting was pretty quick. [14:08] yeah, the only interesting item was about helping developers get up to speed with integrating widgets [14:08] we decided to talk to tim, who reported some difficulties and has started documenting the process [14:09] if any developer still hasn't done any integration work for lazr js widgets, and wants some help and/or wants to help documenting the steps they take, you are most welcome [14:09] thanks intellectronica. [14:10] [Topic] Peanut gallery [14:10] New Topic: Peanut gallery [14:10] b.t.w francis regularly sends meeting notes for that call now. do we really need this section in this meeting? [14:10] Indeed I was looking at those notes [14:11] I think we are only concerned that the team is reading them [14:11] we can do a quiz [14:12] lightning reviews of the week that was [14:12] thanks sinzui, you should always chair ;) [14:12] I'll bring this matter up with the antipodeans. [14:13] As I think about all the talent whose attention I assume I have, I want advice on how to test something I have seen in reviews. [14:13] me [14:14] I do not like story tests that verify a link it not present; they are not a part of the story [14:15] We have tales formatters that ensure the links is rendered correctly (or not at all) and we have permission decorators for links [14:15] So as long as the link is defined correctly and the approved formatter is used, I do not want to vague test that some user does not see the link... [14:16] But there is a problem with out links and views that we are *not* tesing [14:16] lp.testing.menu has a helper that will verify all the links in the menu have a view. During the 3.0 release I found many that did not! [14:17] I used the helper to find and remove the bad links [14:17] Last week I found a new problem that I do not know how to test... [14:17] We have links and views that disagree about permission. [14:18] eg. the link is public, the view is edit. [14:19] in my case, though, there was a companion view that as launchpad.View. I had to audit many links and views. [14:19] Q: Is it possible to test that a link and its view have the same permission? [14:19] * sinzui thanks everyone for their patience [14:21] sinzui: perhaps I'm missing your point why tests for non-existence in stories are bad (for yure they are noisy), but they can ensure that the right permission decorator is used. [14:21] ...non-existience of links... [14:21] adeuring: checking for non-existence in a story is wrong. [14:23] sinzui: where else would you do do this test? [14:23] adeuring: we can test the links elsewhere. in fact, the menu helper could be used to verify all links for permission, but that will not help with the link to view permission [14:24] adeuring: testing contract and implementation details belong in browser/tests [14:24] ok [14:24] sinzui: i think i understand why you dislike testing this kind of stuff in stories, but you should consider the benefits. i think they outweigh the problems with this approach [14:25] doctests are easier to write and read, and most importantly much much easier to maintain, because they include a lot more context [14:25] The testrbowser is not testing tales, and stories should not know about markup. [14:25] sinzui: why? [14:26] intellectronica: browser/tests can be unit or doc they can even employ the test browser. [14:26] sinzui: that's going back to the chaos we had before the tree re-org [14:27] intellectronica: Stories are integration tests that verify that a user can traverse many pages to accomplish a task. it is slow and labourious to use it to test details that have many factors at play at a high level [14:28] sinzui: is this about the time it takes to run the test suite? [14:28] intellectronica: We still have chaos since we have developer writing form contract testing in stories [14:28] intellectronica: no, it is about knowing what we are testing and how we test it [14:29] sinzui: but placing the files in a certain place or writing them in a particular format doesn't buy you confidence that you're testing what you need to [14:29] intellectronica: you misunderstand me. [14:30] sinzui: regarding the discrepancy in permission between links and views, it's usually on-purpose, the link is public but the view protected so that the user can see the link, but is asked to log in [14:30] Testing the rules of a link or a form are best done in isolation where you control that is happen. [14:30] sinzui: it's only a problem if the user is logged in and has permission to the link and not the view [14:31] flacoste: after discovering the link issue, I tried the links in production. I got a 403 [14:33] flacoste: I understand your point. I think the approach to invite users to explore a feature is fine [14:33] right [14:35] Well I do not think there is any more to say on this. [14:35] sinzui: maybe this is a good topic for the list? i really think that i don't understand the motivation behind your argument, so i could use a for-dummies explanation [14:36] intellectronica: I will. I can get examples [14:36] sinzui: thanks [14:37] Does anyone else have an issue to announce/discuss [14:37] 5 [14:37] 4 [14:37] 3 [14:37] 2 [14:37] 1 [14:38] thank you everyone. [14:38] #endmeeting [14:38] Meeting finished at 09:38. === salgado is now known as salgado-lunch === rockstar is now known as rockstar-afk === salgado-lunch is now known as salgado === leonardr_ is now known as leonardr === matsubara is now known as matsubara-lunch === rockstar-afk is now known as rockstar === matsubara-lunch is now known as matsubara === salgado is now known as salgado-afk === matsubara is now known as matsubara-afk [21:12] sorry [21:13] * sinzui was distracted [21:13] who wants to talk about reviews? [21:13] sinzui, :) [21:14] hi [21:14] thumper, wgrant, hi [21:14] #startmeeting [21:14] Meeting started at 16:14. The chair is sinzui. [21:14] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [21:14] reviewer meeting? [21:14] hello everyone and welcome to this week's antipodean reviewers meeting. who's here today? [21:14] * rockstar is [21:15] i am here [21:15] * sinzui thinks everyone who will attend has spoken [21:15] me [21:15] [TOPIC] agenda [21:15] New Topic: agenda [21:15] * Roll call [21:15] * Action items [21:15] * UI review call update [21:15] * Peanut gallery [21:16] [Topic] Action items [21:16] New Topic: Action items [21:16] * barry to get with mrevell on guidelines migration from old wiki to new [21:16] ^ Not done yet [21:16] * intellectronica and barry to draft guidelines for drive-by cleanups [21:16] ^ Not done yet [21:16] [Topic] UI review call update [21:16] New Topic: UI review call update [21:16] * sinzui looks for trascript [21:16] intellectronica: yeah, the only interesting item was about helping developers get up to speed with integrating widgets [21:17] intellectronica: we decided to talk to tim, who reported some difficulties and has started documenting the process [21:17] intellectronica: if any developer still hasn't done any integration work for lazr js widgets, and wants some help and/or wants to help documenting the steps they take, you are most welcome [21:17] Since the UI meeting is producing notes for all of us to read, do we need a weekly update in this meeting? [21:17] sinzui: intellectronica tried, but I was busy :) [21:18] sinzui, I think lazr-js could use more documentation in general. [21:19] I think documentation and lack of leadership was a concern for other canonical groups when they were told of use it [21:19] sinzui, so does lazr-js need a leader? [21:19] beuno: quickly recovered by inviting landscape and U1 people to the lazr-js sprint [21:20] rockstar: Without a sense of who is guiding it, many potential contributors are hesitant to join [21:21] moving on [21:21] [Topic] Peanut gallery [21:21] New Topic: Peanut gallery [21:21] * thumper throws a peanut at sinzui [21:22] I brought up the topic of menu and link testing. I do not like the testing that link is not present in a story [21:22] The matter was complicated by a subdiscussions about where and what we test [21:22] we test far too much in stories, yes [21:23] sinzui: i notice that registry has far more tests in browser/tests these days [21:24] We are doing a lot of view tests since we have a tremendous number of them [21:24] Menus are defines in browser, so that is where I expect to so verification of link permissions, state, etc... [21:25] yep, i think it's a good thing [21:25] I think I need to take my concerns to the dev list. [21:25] this reviewers meeting has a slightly different approach to testing than the other one sometimes :-) [21:25] I am certain of two things. We have links to views that do not exist. We have links with permissions different from the view it links to [21:26] seems likely [21:27] Gentleman, do you have any concerns that we should be discussing? [21:27] 5 [21:28] nope [21:28] nope [21:28] 4 [21:28] 3 [21:28] 2 [21:28] 1 [21:28] thank you gentleman. I will compile the transcripts [21:28] thanks sinzui [21:29] #endmeeting [21:29] Meeting finished at 16:29.