Options:
- # Session Start: Fri Aug 30 00:00:01 2013
- # Session Ident: #css
- # [00:00] * Quits: krit (~krit@public.cloak) (Client closed connection)
- # [00:01] * Quits: rhauck (~Adium@public.cloak) (Ping timeout: 180 seconds)
- # [00:19] * Quits: glenn (~gadams@public.cloak) (Client closed connection)
- # [00:25] <SimonSapin> TabAtkins: very nice (issue index)
- # [00:25] <SimonSapin> also, self-links were just empty until I force-refreshed
- # [00:25] <TabAtkins> Yeah, that happens. Cached CSS.
- # [00:59] * Quits: lmclister (~lmclister@public.cloak) (lmclister)
- # [01:02] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [01:03] * Joins: zcorpan (~zcorpan@public.cloak)
- # [01:09] <shepazu> I'm going to ask a CSS question here… sorry if this is the wrong place, and happy for advice on where to ask the question
- # [01:09] <shepazu> I have an empty span with contenteditable
- # [01:09] <shepazu> no text node
- # [01:09] <shepazu> in Chrome, it retains its width and height
- # [01:10] <shepazu> but in FF, it collapses into a little sliver, vertically
- # [01:10] * Quits: zcorpan (~zcorpan@public.cloak) (Ping timeout: 180 seconds)
- # [01:10] <shepazu> I have no idea what's causing it, or how I should fix that in FF
- # [01:10] <shepazu> so, it basically has no height
- # [01:10] <shepazu> any ideas?
- # [01:12] <TabAtkins> contenteditable is underspecified and crazy.
- # [01:12] <TabAtkins> Easiest thing to do is generate the span with an in it initially, I think.
- # [01:13] <shepazu> I don't think it's a contenteditable issue (though I agree with you)
- # [01:13] <shepazu> I think it's just how a span is rendered without content, if it has margins, padding, bg color, etc
- # [01:14] <TabAtkins> Right, Chrome's behavior is the crazy (but more useful) one.
- # [01:14] <TabAtkins> But still, just put something in it.
- # [01:14] * Quits: teoli (~teoli@public.cloak) (Client closed connection)
- # [01:16] <shepazu> TabAtkins, but then when they click on it to edit, they'll have to erase that bit, or I'll have to, which seems like a pain… there's no way to just have a min-height or something?
- # [01:16] * shepazu has been trying different solutions
- # [01:16] <TabAtkins> No, can't set min-height on inlines.
- # [01:17] <shepazu> hmm.. setting min-height and inline-block seemed to work… let's see if that causes problems
- # [01:17] <shepazu> actually, min-height works alone
- # [01:17] <shepazu> I'll swing with that for a bit, thanks!
- # [01:17] <shepazu> sorry to bug this channel
- # [01:23] <TabAtkins> Yes, inline-block will cause problems if the span can get long enough to wap.
- # [01:23] <shepazu> no, somehow that wasn't true :(
- # [01:24] <TabAtkins> wrap
- # [01:24] <shepazu> hmm
- # [01:26] <TabAtkins> Maybe use ::before with an nbsp in it, with its width set to 0?
- # [01:26] <TabAtkins> and its display set to inline-block
- # [01:33] * Joins: zcorpan (~zcorpan@public.cloak)
- # [01:58] <shepazu> TabAtkins, seems inline-block doesn't cause problems, so I'll go with that
- # [01:58] <shepazu> thanks!
- # [01:59] <TabAtkins> Like I said, it'll cause problems as soon as the span gets long enough to wrap, because it *won't* wrap.
- # [01:59] <TabAtkins> Or rather, the entire span will drop to the next line, and then when it's long enough to overflow a line, wrap internally.
- # [01:59] <TabAtkins> But if you're just doing something short, then yeah, inline-block is fine.
- # [02:00] <shepazu> TabAtkins, for this case, it seems like it's mostly ok
- # [02:01] <shepazu> it's not always short, but popping to a new line isn't terrible here
- # [03:04] <TabAtkins> plinss: We need a "callback" dfn type in the API.
- # [03:04] <TabAtkins> Adding it to Bikeshed now.
- # [03:04] <plinss> what's it for?
- # [03:05] <TabAtkins> Callbacks in WebIDL.
- # [03:05] <TabAtkins> They're distinguished syntax now.
- # [03:05] <TabAtkins> Example in Font Load Events.
- # [03:06] <plinss> ah, ok. So I probably need to add support in my IDL parser too then...
- # [03:06] <TabAtkins> Yeah.
- # [03:06] <plinss> ok, adding it now
- # [03:06] <TabAtkins> Is your IDL parser exposed in a module-y way? I'd like to avoid forcing people to specify for='' attrs when it's obvious from the IDL.
- # [03:07] <plinss> it's not in a standalone module just yet, but it is isolated from all the other code
- # [03:07] <plinss> I'm happy to move it somewhere so that we can just share it
- # [03:08] <TabAtkins> Yeah, a separate file (just to ensure that you don't accidentally have internal linkages anyway) would be great.
- # [03:08] <TabAtkins> So I can just dump it in bikeshed's /lib directory.
- # [03:09] <plinss> I believe it's already good for that…
- # [03:09] * Quits: rhauck1 (~Adium@public.cloak) ("Leaving.")
- # [03:09] <plinss> http://hg.csswg.org/dev/shepherd/file/tip/python/shepherd/idlparser.py
- # [03:09] <TabAtkins> Ah, then that's fine.
- # [03:15] <TabAtkins> plinss: How do you feel about an "event" type?
- # [03:15] <plinss> hmm
- # [03:16] <plinss> is it just subclasses of 'Event'? or something different?
- # [03:17] <TabAtkins> I dunno, I just know that Font Load Events has some event definitions in it.
- # [03:18] <plinss> I don't necessarily see where we need to do anything special about it at the moment...
- # [03:18] <plinss> when I start tracking IDL constructs, you'll be able to see the base interfaces
- # [03:18] <TabAtkins> Ah, to answer your question better, no, I mean the event names.
- # [03:19] <plinss> ah
- # [03:19] <TabAtkins> Preferably with automatic punning between "foo" and "onfoo" in the autolinker.
- # [03:19] <plinss> hmm, maybe
- # [03:21] <plinss> I don't see any event <dfn>s or other ids bound to the individual events in font load events...
- # [03:21] <TabAtkins> Right, because I'm rewriting it at the moment.
- # [03:21] <plinss> ok then...
- # [03:21] <TabAtkins> The definitions are all kinds of funky right now.
- # [03:21] <plinss> so is it going to have a <dfn> then?
- # [03:21] <TabAtkins> Yes.
- # [03:21] <TabAtkins> (At the moment they're "defined" with an <a id>.)
- # [03:22] <plinss> then sure, let's add a dfn type
- # [03:23] <TabAtkins> Hm. I wonder now how to categorize it.
- # [03:23] <TabAtkins> I guess events have to have unique names already.
- # [03:23] <TabAtkins> So no need for a for=''
- # [03:42] * heycam is now known as heycam|away
- # [04:34] <plinss> TabAtkins: ok, event and callback support is online
- # [04:34] <plinss> and the idl parser now parses callbacks, so grab the new one
- # [05:09] * Quits: zcorpan (~zcorpan@public.cloak) (Ping timeout: 180 seconds)
- # [05:51] * Quits: cabanier (~cabanier@public.cloak) ("Leaving.")
- # [05:52] * Joins: cabanier (~cabanier@public.cloak)
- # [05:52] * Quits: cabanier (~cabanier@public.cloak) ("Leaving.")
- # [06:19] * heycam|away is now known as heycam
- # [09:03] * heycam is now known as heycam|away
- # [09:07] * Joins: cabanier (~cabanier@public.cloak)
- # [09:07] * Joins: teoli (~teoli@public.cloak)
- # [09:41] * Joins: Ms2ger (~Ms2ger@public.cloak)
- # [09:49] * Joins: zcorpan (~zcorpan@public.cloak)
- # [11:01] * Quits: teoli (~teoli@public.cloak) (Client closed connection)
- # [11:34] * Joins: teoli (~teoli@public.cloak)
- # [11:59] * Joins: myakura (~myakura@public.cloak)
- # [12:08] * Quits: myakura (~myakura@public.cloak) (Client closed connection)
- # [12:09] * Joins: myakura (~myakura@public.cloak)
- # [12:16] * Quits: myakura (~myakura@public.cloak) (Ping timeout: 180 seconds)
- # [12:50] * Joins: myakura (~myakura@public.cloak)
- # [13:17] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [13:38] * Joins: zcorpan (~zcorpan@public.cloak)
- # [13:46] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [13:57] * Joins: zcorpan (~zcorpan@public.cloak)
- # [13:58] * Quits: teoli (~teoli@public.cloak) (Client closed connection)
- # [14:03] * Joins: teoli (~teoli@public.cloak)
- # [14:05] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [14:16] * Quits: myakura (~myakura@public.cloak) (Client closed connection)
- # [14:17] * Joins: myakura (~myakura@public.cloak)
- # [14:24] * Quits: myakura (~myakura@public.cloak) (Ping timeout: 180 seconds)
- # [14:29] * Joins: myakura (~myakura@public.cloak)
- # [14:44] * Quits: myakura (~myakura@public.cloak) (Client closed connection)
- # [14:44] * Joins: myakura (~myakura@public.cloak)
- # [14:46] * Joins: myakura_ (~myakura@public.cloak)
- # [14:46] * Quits: myakura (~myakura@public.cloak) (Client closed connection)
- # [14:56] * Joins: zcorpan (~zcorpan@public.cloak)
- # [15:02] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [15:10] * Joins: plh (plehegar@public.cloak)
- # [15:12] * Quits: myakura_ (~myakura@public.cloak) (Client closed connection)
- # [15:12] * Joins: myakura (~myakura@public.cloak)
- # [15:19] * Quits: myakura (~myakura@public.cloak) (Ping timeout: 180 seconds)
- # [15:53] * Joins: myakura (~myakura@public.cloak)
- # [16:04] * Joins: zcorpan (~zcorpan@public.cloak)
- # [16:08] * Quits: myakura (~myakura@public.cloak) (Client closed connection)
- # [16:08] * Joins: myakura (~myakura@public.cloak)
- # [16:09] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [16:09] * Joins: abucur (~Adium@public.cloak)
- # [16:14] * Joins: lmclister (~lmclister@public.cloak)
- # [16:15] * Quits: myakura (~myakura@public.cloak) (Ping timeout: 180 seconds)
- # [16:31] * Joins: myakura (~myakura@public.cloak)
- # [16:33] * Quits: myakura (~myakura@public.cloak) (Client closed connection)
- # [16:34] * Joins: myakura (~myakura@public.cloak)
- # [16:41] * Quits: myakura (~myakura@public.cloak) (Ping timeout: 180 seconds)
- # [17:25] * Joins: myakura (~myakura@public.cloak)
- # [17:46] * Quits: myakura (~myakura@public.cloak) (Client closed connection)
- # [17:51] * Joins: zcorpan (~zcorpan@public.cloak)
- # [17:53] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [18:19] * Joins: zcorpan (~zcorpan@public.cloak)
- # [18:21] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [18:37] * Quits: abucur (~Adium@public.cloak) ("Leaving.")
- # [18:51] * Joins: rhauck (~Adium@public.cloak)
- # [19:11] * Joins: zcorpan (~zcorpan@public.cloak)
- # [19:18] * Quits: teoli (~teoli@public.cloak) (Client closed connection)
- # [19:27] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [19:28] * Quits: darktears (~darktears@public.cloak) (Client closed connection)
- # [19:28] * Joins: darktears (~darktears@public.cloak)
- # [19:56] * Joins: teoli (~teoli@public.cloak)
- # [20:20] * Joins: tantek (~tpod@public.cloak)
- # [20:30] * Quits: tantek (~tpod@public.cloak) ("Colloquy for iPod touch - http://colloquy.mobi")
- # [20:37] * Joins: zcorpan (~zcorpan@public.cloak)
- # [20:44] * Quits: zcorpan (~zcorpan@public.cloak) (Ping timeout: 180 seconds)
- # [21:10] * Quits: Ms2ger (~Ms2ger@public.cloak) ("nn")
- # [21:22] * Quits: lmclister (~lmclister@public.cloak) (lmclister)
- # [22:06] <TabAtkins> plinss: You said you prefer the <dfn>s to be in the IDL block, right?
- # [22:06] <TabAtkins> I'm wondering if it's more useful for attribute and method definitions to be at the point of the actual definition, and just have the IDL link to them.
- # [22:46] <TabAtkins> plinss: Also wondering how to categorize method parameters.
- # [22:55] <TabAtkins> Hm, though. Many IDL constructs are most naturally defined in the IDL block, like dicts.
- # [22:55] <TabAtkins> Maybe just take it on a case-by-case.
- # [22:57] <TabAtkins> That is, if you have a prose definition, make that the actual <dfn>, and link to it from the IDL. If you just use something in the IDL and don't really define it elsewhere, like a dict for some method's argument, just dfn it there in the IDL.
- # [23:05] <plinss> TabAtkins: yeah, put the <dfn>s where they make the most sense
- # [23:07] <plinss> the nice thing about when they're in in the IDL is that I can always tell what they are. If they're not, the I only do IDL lookups if they have some magic trigger
- # [23:07] <TabAtkins> plinss: kk. I'll write up some docs for this eventually in Bikeshed with best practices.
- # [23:07] <plinss> yes please
- # [23:07] <plinss> I don't do anything with method params at the moment
- # [23:07] <plinss> do we have dfns for them anywhere?
- # [23:08] <TabAtkins> I'm adding them to Font Load Events, so I can link up to it. ^_^
- # [23:08] <TabAtkins> Dunno if I'm going overboard or not, though.
- # [23:08] <plinss> heh, ok, so I guess I need to parse those too now...
- # [23:08] <plinss> maybe, it depends on the prose
- # [23:08] <TabAtkins> Yeah.
- # [23:08] <TabAtkins> It ends up with a funky nested <dfn>, interestingly enough.
- # [23:09] <TabAtkins> <dfn method for="FontLoader" title="loadFont()">loadFont(<dfn argument for="FontLoader/loadFont()">params</dfn>)</dfn>
- # [23:09] <plinss> if there's a good textual definition of a param, then it should have a dfn
- # [23:09] <TabAtkins> (Using the same convention as descriptor values, where a two-level for='' value is slash-separated.
- # [23:09] <TabAtkins> )
- # [23:09] <plinss> hmm, for the nested, can't you skip the 'for'? (and take it from context)
- # [23:09] <TabAtkins> Yeah.
- # [23:09] <TabAtkins> Hm, yeah, I could.
- # [23:10] <TabAtkins> I'll go ahead and make that change - it would reduce verbosity a lot.
- # [23:10] <plinss> ok, I'll go ahead and add the argument type
- # [23:10] <plinss> (and parse them in the IDL)
- # [23:10] <TabAtkins> And I'm using "argdef" for it.
- # [23:10] <plinss> ok
- # [23:11] <plinss> I used 'callbackdef' and 'eventdef' for the other two
- # [23:12] * Quits: darktears (~darktears@public.cloak) ("Linkinus - http://linkinus.com")
- # [23:12] <plinss> any other IDL types while we're at it?
- # [23:12] <TabAtkins> Yeah, same here.
- # [23:12] <TabAtkins> I'm calling them as I see them, but I think we've nearly slurped up all the IDL constructs. ^_^
- # [23:12] <TabAtkins> I mean, eventually we might need to hit extended attributes or something.
- # [23:13] <plinss> those are probably going to be aspects of the other constructs
- # [23:14] <plinss> there are 'typedef's and 'exception's left...
- # [23:15] <plinss> we should maybe add exception at least...
- # [23:16] <plinss> typedef conflicts with css typedefs...
- # [23:16] <plinss> not sure what to do about that one
- # [23:22] <TabAtkins> idltype?
- # [23:22] <TabAtkins> That's dumb, but whatever.
- # [23:23] <TabAtkins> <dfn typedef id="typedefdef-foo">foo</dfn>
- # [23:26] * Quits: plh (plehegar@public.cloak) ("Leaving")
- # [23:32] <plinss> ick
- # [23:36] <TabAtkins> Yeah, that was a joke. ^_^
- # [23:36] <plinss> I was hoping so...
- # Session Close: Sat Aug 31 00:00:00 2013
The end :)