| 00:50 | <TabAtkins> | annevk: For Document and ASCII case-insensitive, should I be reffing DOM or HTML? |
| 03:27 | <Domenic> | This is a good question |
| 03:27 | <Domenic> | I guess Document should probably be DOM |
| 03:27 | <Domenic> | HTML is partial |
| 03:28 | <Domenic> | Maybe we should consolidate the string stuff in one of the two specs |
| 06:15 | <smaug> | annevk: the min 1s timeout in background tabs isn't spec'ed anywhere, right? |
| 06:15 | <smaug> | it is just spec'ed that UA is allowed to do something like that |
| 06:16 | <annevk> | smaug: think so |
| 06:16 | <smaug> | k |
| 07:56 | <annevk> | zcorpan: I guess I'll leave that PR to Domenic |
| 07:56 | <annevk> | zcorpan: seems like he knows what needs doing |
| 07:56 | <zcorpan> | sounds like a good idea :-) |
| 07:57 | <annevk> | Anything else someone wants me to look at? I'm back to being fulltime available for the remainder of this week |
| 09:16 | <zcorpan> | annevk: why add a separate file instead of changing cross-origin-objects.html directly? |
| 09:17 | <annevk> | zcorpan: because bholley wanted points for throwing "whatever" exceptions |
| 09:18 | <annevk> | zcorpan: long term it should just be one file |
| 09:18 | <zcorpan> | annevk: ok |
| 12:06 | <mkwst> | tobie: Ping for a quick specref question. You around? |
| 12:07 | <tobie> | mkwst: yes |
| 12:08 | <mkwst> | tobie: https://w3c.github.io/webappsec-csp/ is pointing to CSP1 for some IDL references. I'm sure half of this is a bikeshed issue, but I think the other half is specref/W3C not being clear about the relationship between CSP{1,2,3}. |
| 12:09 | <mkwst> | https://specref.herokuapp.com/bibrefs?refs=CSP shows an `aliasOf` CSP1, for instance. CSP{2,3} don't seem to fit into that chain. Is this a red herring, or can I poke at the data somewhere to link these together? |
| 12:10 | tobie | looking |
| 12:13 | <tobie> | mkwst: so… |
| 12:14 | <tobie> | mkwst: the aliasing bit comes from W3C |
| 12:15 | <tobie> | what should it alias to? CSP2 or CSP3? |
| 12:15 | <tobie> | (it could also be a bug in the import script, LMK check) |
| 12:16 | <mkwst> | Ideally, CSP would mean CSP3. And CSP2 and CSP1 would be obsoleted. |
| 12:16 | <mkwst> | I have no idea how the W3C represents this data, so I'm happy to believe that that's the issue. :) |
| 12:17 | <annevk> | tobie: seems DOM xref "call a user object's operation" to IDL broke |
| 12:17 | <tobie> | mkwst: in RDF, what else? |
| 12:17 | <tobie> | :D |
| 12:18 | <mkwst> | tobie: Ah. Not JSONLD? I thought that was the new hotness. |
| 12:18 | <tobie> | mkwst: think there is one such API that happened since |
| 12:19 | <tobie> | annevk: yeah, I think the apostrophes caused issues |
| 12:19 | <annevk> | tobie: they don't in bikeshed |
| 12:19 | <annevk> | tobie: I guess it's fairly new so I could change DOM, but it seems nicer not to have to add an lt="" |
| 12:20 | <tobie> | annevk: let me look into this |
| 12:20 | <annevk> | ta |
| 12:22 | <tobie> | mkwst: so… the redirects on TR point to CSP2 while the ones in the RDF file point to CSP1 |
| 12:22 | <mkwst> | tobie: who should I poke? |
| 12:22 | <mkwst> | plh? |
| 12:23 | <tobie> | mkwst: no there's an email address, let me check |
| 12:25 | <tobie> | mkwst: sysreq⊙wo |
| 12:26 | <tobie> | mkwst: you want to tell them to change the following in https://www.w3.org/2002/01/tr-automation/tr.rdf: |
| 12:26 | <mkwst> | Ok. And I'll say "The RDF file is wrong. You know. The RDF file. That you use. For the thing." |
| 12:26 | <mkwst> | Thanks. ;) |
| 12:26 | <tobie> | mkwst: here's the content they'll want to modify: |
| 12:27 | <tobie> | https://www.irccloud.com/pastebin/I4XHs8XV/ |
| 12:27 | <tobie> | You can always /cc me on the mail: tobie⊙cc |
| 12:28 | <tobie> | mkwst: that said, I'm not sure this will magically fix your IDL issue |
| 12:28 | <tobie> | mkwst: as I'm not sure how exactly Tab handles this in Bikeshed |
| 12:28 | <mkwst> | Like I said, I'm sure half of it is me holding Bikeshed wrong. |
| 12:29 | <mkwst> | But the CSP1 reference is weird enough that I'm sure the W3C is somehow responsible. :) |
| 12:29 | <tobie> | mkwst: well, that needed fixing anyway |
| 12:29 | <tobie> | mkwst: plus you might ask them to fix the redirects on TR while you're at it |
| 12:29 | <tobie> | mkwst: it would be nice to have more consistency there, tbh |
| 12:31 | <mkwst> | is there something I can do about it, other than writing this email? |
| 12:31 | <mkwst> | like, some structure I (or Bikeshed?) should be adding to the published documents? |
| 12:36 | <tobie> | annevk: this seems to be a line break issue in the original spec |
| 12:36 | <mkwst> | (Emailed. Thanks for the help, tobie!) |
| 12:36 | <tobie> | mkwst: np |
| 12:37 | <jez9999> | Is this spec governed by WhatWG: https://developer.mozilla.org/en-US/docs/Web/API/HTML_Drag_and_Drop_API |
| 12:38 | <tobie> | mkwst: regarding the shortname -301-> shortname[LEVEL], this is done in a completely ad-hoc manner |
| 12:38 | <annevk> | jez9999: it's part of html.spec.whatwg.org/multipage |
| 12:39 | <tobie> | mkwst: generally this is reflected automatically in the rdf file, but sometimes not |
| 12:40 | <tobie> | annevk: the x-ref seems to WFM |
| 12:41 | <tobie> | annevk: did you fix things already? |
| 12:41 | tobie | is confused |
| 12:43 | <annevk> | tobie: I get a warning locally when generating DOM |
| 12:43 | <tobie> | annevk: oh |
| 12:43 | <jez9999> | annevk: i see that it defines the event handlers that must be supported by all HTML elements, but is there anywhere where it actually defines the behaviour of the drag/drop API events? |
| 12:44 | <tobie> | annevk: what does the warning say? |
| 12:44 | <annevk> | tobie: "No 'dfn' refs found for 'call a user object's operation'" |
| 12:45 | <annevk> | tobie: it seems this isn't really introduced by you though, since afaict this should just work |
| 12:45 | <tobie> | annevk: I mean, it works now... |
| 12:45 | <tobie> | annevk: so I don't know |
| 12:45 | <annevk> | tobie: I'm guessing Shepherd is doing something wrong hmmmmm |
| 12:46 | <tobie> | annevk: might have to do with line breaks that I removed |
| 12:47 | <tobie> | annevk: there was a line break in the middle of the text before the bikeshed conversion |
| 12:47 | <tobie> | annevk: so if something along the pipeline is sensitive to this, it might be causing the problem |
| 12:47 | <annevk> | tobie: that'd be a very weird |
| 12:48 | <annevk> | tobie: filed https://github.com/tabatkins/bikeshed/issues/815 |
| 12:49 | <jez9999> | Suggestion: https://html.spec.whatwg.org/#datatransferitem |
| 12:49 | <jez9999> | a DataTransferItem should allow the code to see whether it is a file or folder. |
| 12:50 | <jez9999> | alternatively, the DataTransferItemList interface could allow Chrome's .getAsEntry() method, which can be a FileEntry or a DirectoryEntry |
| 12:50 | <jez9999> | see: https://developers.google.com/web/updates/2012/07/Drag-and-drop-a-folder-onto-Chrome-now-available?hl=en |
| 12:50 | <annevk> | jez9999: there should be a processing model somewhere in the spec |
| 12:50 | <jez9999> | processing model? |
| 12:50 | <annevk> | jez9999: that defines when events are dispatched (and where) |
| 12:51 | <annevk> | jez9999: as for new features, best to file GitHub issues for those |
| 12:51 | <jez9999> | annevk: yeah i'm talking about the definition of the event received by ondrop |
| 12:51 | <jez9999> | where's the place on github to file? |
| 12:51 | <annevk> | jez9999: and keep https://wiki.whatwg.org/wiki/FAQ#Is_there_a_process_for_adding_new_features_to_a_specification.3F in mind, though if a feature already has an impl it definitely deserves to be filed |
| 12:51 | <annevk> | jez9999: https://github.com/whatwg/html/issues/new is the place |
| 12:51 | <jez9999> | ok |
| 13:03 | <jez9999> | annevk: https://github.com/whatwg/html/issues/1777 |
| 14:06 | <annevk> | JakeA: can https://github.com/w3c/ServiceWorker/blob/master/Makefile be removed? |
| 14:10 | <JakeA> | annevk: hah, yeah. *gone* |
| 14:12 | <annevk> | JakeA: also, it seems "assets" in spec/ can be removed? |
| 14:12 | <annevk> | JakeA: and the contribution guidelines should no longer talk about the web components framework? |
| 14:15 | <annevk> | JakeA: means you can get rid of .gitmodules |
| 14:18 | <JakeA> | annevk: agreed. Fixing now. Also want to make https://w3c.github.io/ServiceWorker/ the spec url |
| 14:18 | <annevk> | JakeA: that'd be nice |
| 14:19 | <JakeA> | annevk: I don't have admin rights, I guess I'm not allowed them, but will ask w3 folks to make the changes |
| 14:20 | <Ms2ger> | MikeSmith, give the man access already |
| 15:25 | <JakeA> | annevk: done the cleanup, and waiting on a switch to be flipped for moving the spec to / |
| 15:25 | <JakeA> | https://github.com/w3c/ServiceWorker/issues/966#issuecomment-246709306 |
| 15:28 | <annevk> | JakeA: saw that, hope it works out |
| 15:28 | <annevk> | and thanks |
| 15:58 | <Domenic> | https://browser-issue-tracker-search.appspot.com/ looks useful |
| 15:59 | <Domenic> | Hmmm still pretty WIP I guess |
| 15:59 | <Domenic> | https://browser-issue-tracker-search.appspot.com/?q=same-origin%20data%20URL |
| 16:56 | <jez9999> | http://stackoverflow.com/questions/15363632/background-size-contain-but-dont-scale-up |
| 16:56 | <jez9999> | are there any plans to fix this? it seems crazy that there is no way to do this in CSS |
| 17:07 | <jyasskin> | Hey TabAtkins, https://html.spec.whatwg.org/multipage/dom.html#document has data-dfn-lt="", but `bikeshed refs --text Document` still finds it. |
| 17:42 | <TabAtkins> | Ugh, I'll bet Shepherd's heuristic mode is just looking for that being falsey and using the text instead. I'll bug plinss. |