| 02:23 | <MikeSmith> | CSP disown-opener is really not supported in any UAs yet? |
| 02:23 | <MikeSmith> | https://developer.mozilla.org/en-US/docs/Web/HTTP/CSP#Browser_compatibility |
| 05:35 | <annevk> | MikeSmith: didn't that have a lot of issues? |
| 05:36 | <annevk> | MikeSmith: maybe just one: https://github.com/w3c/webappsec-csp/issues/194 |
| 05:56 | <MikeSmith> | annevk: didn’t know ー wasn’t following it so close |
| 05:56 | MikeSmith | looks |
| 05:58 | <MikeSmith> | hmm yeah |
| 05:58 | <MikeSmith> | so yeah I guess that’s why it’s not been implemented |
| 09:47 | <smaug____> | jgraham: who knows about wpt.fyi? |
| 09:48 | <smaug____> | would it be possible to get it to run also Nightly? |
| 09:48 | <annevk> | foolip: ^^ |
| 09:53 | <Ms2ger> | Doesn't it? |
| 09:55 | <jgraham> | smaug____: Yeah, it's being worked on |
| 09:55 | <smaug____> | great |
| 09:55 | <jgraham> | We have the data, but are missing the code to upload the data to the UI |
| 09:55 | <smaug____> | oh |
| 09:56 | <jgraham> | It's being worked on aiui |
| 10:02 | <foolip> | smaug____: https://wpt.fyi/results/?label=experimental |
| 10:02 | <foolip> | It's been available for a week or so, just not anncounced |
| 10:03 | <smaug____> | NICE |
| 10:03 | <smaug____> | er, nice |
| 10:03 | <foolip> | jgraham: that's from buildbot. are Taskcluster builds reliable yet so we can start getting the results from there soonish? |
| 10:04 | <foolip> | Beware of https://github.com/web-platform-tests/wpt.fyi/issues/188 if copying URLs |
| 13:50 | <Domenic> | TabAtkins: you up and ready to merge https://github.com/tabatkins/bikeshed/pull/1239 ? |
| 14:07 | <wanderview> | zcorpan: I was looking at this... https://bocoup.github.io/wpt-disabled-tests-report/ |
| 14:07 | <wanderview> | zcorpan: are you checking that every annotation you parse is actually a real test? in the mozilla tree we actually have stale annotations for tests that don't exist any more |
| 14:09 | <zcorpan> | wanderview: I'm not checking that |
| 14:10 | <wanderview> | ok |
| 14:32 | <TabAtkins> | Domenic: Aight, merging. |
| 14:32 | <Domenic> | Perfect timing :) |
| 14:32 | <Domenic> | Thank you!! |
| 14:33 | <TabAtkins> | I should have made you a collaborator so you could do this yourself, silly me. Just invited you. |
| 14:33 | <TabAtkins> | annevk: You want in, too? That way y'all can merge whatever boilerplate stuff y'all need. |
| 14:37 | <annevk> | ah cool |
| 14:46 | <MikeSmith> | TabAtkins: about the syntax highlighter, I’ve still not been able to get output from it no matter what I test with ー including very simple HTML snippets and very simple CSS snippets |
| 14:46 | <TabAtkins> | Yeah, looking into that now. Very weird, since it definitely works on my end. ^_^ |
| 14:47 | <MikeSmith> | giving it CSS seems to cause it hit some bug and fail completely: |
| 14:47 | <MikeSmith> | File "/Users/mike/workspace/html-build/highlighter/highlighter/pygments/pygments/lexer.py", line 432, in _process_token |
| 14:47 | <MikeSmith> | 'token type must be simple type or callable, not %r' % (token,) |
| 14:47 | <MikeSmith> | AssertionError: token type must be simple type or callable, not Token.Comment |
| 14:48 | <MikeSmith> | TabAtkins: OK, could well be something broken in my environment |
| 14:49 | <MikeSmith> | FYI I have Pygments 2.2.0 installed from pip |
| 14:49 | <MikeSmith> | and Python 2.7.15 |
| 14:51 | <TabAtkins> | I'm using a local Pygments anyway, that doesn't matter. |
| 14:51 | <MikeSmith> | k |
| 14:53 | <TabAtkins> | MikeSmith: Ah, one problem is that I didn't set this up to be called directly, I wrote it as a module to be used by other Python code. Easy fix. ^_^ |
| 14:54 | <MikeSmith> | oh |
| 14:54 | <MikeSmith> | cool |
| 14:54 | <MikeSmith> | annevk: about that wattsi .travis.yml, I added that stub to the master branch because I’ve been putting together a patch for doing an actually-working version, on another branch, and testing it. But as far as I can see, the way that Travis+Github integration is set up, Travis never gets triggered unless you have a .travis.yml on the master branch |
| 14:55 | <annevk> | MikeSmith: I don't think that's the case |
| 14:55 | <annevk> | MikeSmith: you can flip some switch inside Travis I think that makes it start monitoring the repo |
| 14:58 | <MikeSmith> | yeah I tried that and it didn’t seem to work |
| 14:58 | <MikeSmith> | I guess I was doing it wrong |
| 14:58 | <MikeSmith> | it only started working after I added the .travis.yml on master |
| 15:00 | <MikeSmith> | anyway I’m stalled on actually getting it working even on the branch, due to not having a good way to get the fpc binaries. When I have time next, I think I’ll just copy them to my own server so that travis can grab them from there |
| 15:00 | <MikeSmith> | over HTTPS |
| 15:06 | <Domenic> | MikeSmith: hmm I thought you could get them from apt these days? |
| 15:06 | <Domenic> | MikeSmith: at least in debian/docker we successfully do `apt-get update && apt-get install -y ca-certificates curl rsync git unzip fp-compiler default-jre` |
| 15:10 | <MikeSmith> | Domenic: when I tried that on Travis it installed fpc 2.6 something ー not the 3.0.x we need |
| 15:11 | <Domenic> | Hmm :(. Including the update step? Apparently Travis used to do updates automatically but now requires opt-in |
| 15:11 | <MikeSmith> | well I think Travis still only has Ubuntu Trusty as the latest distro version, and the latest version of fpc available for that is 2.6.x |
| 15:12 | <Domenic> | Ah I see :( |
| 15:12 | <Domenic> | I guess you could get around it with Docker if you wanted to |
| 15:12 | <MikeSmith> | or maybe there’s some other magic I need to do (I didn’t try doing the update step) |
| 15:12 | <MikeSmith> | yeah |
| 15:12 | <Domenic> | But hosting the binaries is also easy |
| 15:13 | <MikeSmith> | right, I reckon just hosting the binaries will be the least trouble |
| 16:30 | <annevk> | Domenic: I think the snapshot integrity problem is shared by Bikeshed documents |
| 16:30 | <annevk> | Domenic: we only copy the .bs resource after all, we don't copy any of the additional inputs |
| 16:30 | <Domenic> | annevk: we copy their extra files at least. |
| 16:31 | <annevk> | Domenic: only at publication time |
| 16:31 | <Domenic> | Sure. What other time would we copy them? |
| 16:32 | <annevk> | Domenic: publish RD; modify one of the non-.bs sources on master; modify RD; |
| 16:32 | <annevk> | Domenic: modified RD will now incorperate the non-.bs source changes |
| 16:32 | <Domenic> | If you right-click an image in a commit snapshot of console, you get taken to a snapshot image, not the living one. |
| 16:32 | <Domenic> | Ah, I see. |
| 16:33 | <Domenic> | So maybe we need a folder for each review draft, with all the sources inside :/ |
| 16:33 | <Domenic> | Hopefully this could be done by just modifying review.sh |
| 16:35 | <annevk> | It needs some input about which sources it needs to copy |
| 16:36 | <annevk> | E.g., visualize.py for the Encoding Standard |
| 16:36 | <annevk> | I'm not a 100% sure this is worth it though |
| 16:36 | <annevk> | I'd rather just not modify RDs |
| 19:04 | <jwalden> | yeesh |
| 19:05 | <jwalden> | someone adds me to the wpt repository in some fashion, I come back a day or two later to 157 notifications |
| 19:17 | <Domenic> | Yeah, having that happen on a US holiday was a bit rough on my inbox |
| 19:18 | <Domenic> | But, at least I got a brief glimpse of zcorpan_ burning down all the disabled tests |
| 20:27 | <TabAtkins> | jwalden: "filter messages like these" => "label as wpt" => click on wpt label => archive all ^_^ |
| 20:29 | <jsbell> | "My inbox is what now? That's more privacy policy updates than I was expecting." |