The web in a box - a next generation web framework for the Perl programming language

IRC log for #mojo, 2015-08-14

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 sri well... you don't know 100%
00:00 sri that show is such a roller coaster ride
00:01 nicomen is it aired o netflix or hbo?
00:01 Grinnz USA network
00:01 Grinnz i dont think the first episode is online anymore :(
00:02 nicomen USA network is a channel?
00:02 Grinnz stupid non-netflix networks
00:02 Grinnz yes
00:03 sri you'd expect it to be a netflix show
00:03 sri or even amc
00:05 Grinnz USA network has some good shows, but yeah its not really their usual stuff
00:06 sri oh my, i might have totally misinterpreted the last episode and mr robot is totally not real :o
00:06 * sri IS SOOO CONFUSED
00:06 Grinnz lol
00:06 sri yea, that's the kind of show it is
00:08 sri no, i'm 75% sure he's real -.-
00:10 lluad joined #mojo
00:35 cpan_mojo Mojo-Pg-2.08 by SRI https://metacpan.org/release/SRI/Mojo-Pg-2.08
00:40 PryMar56 joined #mojo
00:41 sri lol https://www.indiegogo.com/projects/realise-minas-tirith#/story
00:41 sri 1.8 billion seems like a steal
00:41 Grinnz to make a city, yeah thats ambitious
00:43 genio The last episode I saw was pretty messed up with the drug dealers and his girlfriend.  I haven't had the time to watch the others yet
00:43 sri assuming any of those people is real ;p
00:44 cpan_mojo Mojo-SQLite-0.011 by DBOOK https://metacpan.org/release/DBOOK/Mojo-SQLite-0.011
00:44 sri i guess mr robot could be real and elliot imaginary
00:48 dotan1 joined #mojo
00:51 Snelius joined #mojo
00:52 bjoernfan joined #mojo
00:57 meshl joined #mojo
01:20 * sri thinks a mojolicious + postgresql book could sell pretty well
01:21 sri there's so few decent postgresql introductions, would be a great way to steer people towards perl :)
01:22 sri along the lines of those hipster mongodb + express books
01:23 jberger That's a great idea
01:24 sri "web development with grownup technologies"
01:25 sopanshewale joined #mojo
01:29 jberger Like "learn x the hard way"
01:29 jberger This could be "build your site with technologies that will still exist in 5 years"
01:30 bpmedley So, write the book in Wingdings and force the user to manually type and convert the font to read?
01:30 jberger HEY WINDINGS IS A PERFECTLY VALID STANDARD!
01:31 nicomen spent a couple of hours debugging now, turns out a CDATA section that was not closed (but closed soon after following another opening) was f"#?%ing my xml. xmllint was happy. arg
01:32 nicomen should've checked against a schema
01:32 jberger Wow, lint fail
01:32 nicomen had I indented stuff nicer, it would've been more obvious
01:32 nicomen my own fault
01:34 mst ow, yeah, I rely on xmllint a lot too
01:35 nicomen re: wingdings, you know what's not fair? noone ever made a comic serif
01:36 nicomen oh someone did
01:36 nicomen i like it http://idsgn.org/posts/comic-sans-comic-serif/
01:36 sri nicomen: language
01:38 nicomen haha, ok, sorry
01:49 jberger sri: ok, give me that example for the selector again?
01:50 sri jberger: you won't believe it
01:50 jberger oh nm, http://irclog.perlgeek.de/mojo/2015-08-12#i_11046721
01:50 jberger what, you just did it?
01:50 sri i was just looking up the section in the css4 selectors spec
01:50 sri and they changed it again
01:50 jberger :o
01:51 sri the one i wanted is gone
01:51 sri http://www.w3.org/TR/selectors4/#matches
01:52 sri at least i can't find anything about selectors stating with combinators anymore
01:52 jberger http://www.sadtrombone.com/
01:53 sri wait what
01:53 sri subject selectors are back Oo
01:53 sri http://www.w3.org/TR/selectors4/#subject
01:54 sri ohoh
01:55 sri the date on the spec is wrong
01:55 * sri is so confused again
01:55 * jberger waits
01:56 * jberger reads the code anyway, its worth a refresher
01:56 nicomen 75% chance the spec is real
01:56 sri they moved the spec it seems
01:56 sri https://drafts.csswg.org/selectors-4/#relational
01:57 jberger oh, from the examples I get it
01:57 jberger wow, I've wanted that so many times!
02:00 nicomen is there a difference between '> foo' and 'foo' inside the has() context?
02:03 jberger I would assume the difference is the same as 'bar > foo' and 'bar foo'
02:04 sri correct
02:04 sri 'foo' is basically ' foo'
02:04 sri and ' ' is a combinator
02:06 nicomen ah right
02:06 sri i think Mojo::DOM::CSS can already tokenize '> foo' and you'll have to find a way to match the combinator
02:07 sri (at the end, since the selector AST is inverted, it starts matching at the end and works its way backwards up the tree)
02:09 sri basically tokenized into [[tag => 'foo'], [combinator => '>']]
02:10 sri for 'bar > foo' you get [[tag => 'foo'], [combinator => '>'], [tag => 'bar']]
02:11 sri then it searches for a foo tag, then uses the combinator to get new candidates (parent), and matches the bar tag
02:13 sri attributes and pseudo classes work similar to tag matching
02:14 * jberger tries to wrap his head around this code
02:14 jberger I wonder if I've ever read it before?
02:14 sri i imagine the combinator stuff is a bit tricky to follow
02:16 sri combinators (parent, sibling...) just give you more candidates for matching, tag, attributes, pseudo-classes actually match against elements
02:17 * jberger reads furiously
02:18 sri i would try a few selectors and dumper $group in _compile
02:18 sri regexes are precompiled there and stuff, dumper should be good for getting a feel for it
02:19 jberger a silly thought: could this pseudoclass put a marker somewhere and then continue matching as if normal, then when matched return the marked spot rather than the end of the chain?
02:20 jberger it kinda seems like that is what it is doing
02:20 genio When giving results to a callback, is it preferred to $self->$cb($err,$res) which makes the object instance the first parameter or just plain old $cb->($err,$res)?
02:21 sri i don't think the pseudo class is important, it can just be a generic feature of the engine, like making $dom->at('> foo') work
02:21 jberger genio: most of my code assumes an "invocant" to the callback
02:22 sri there are multiple markers, and yes, i think a marker will be involved
02:23 sri if it requires a new one i'm not sure about
02:23 rwp joined #mojo
02:25 thowe joined #mojo
02:26 nicomen would running has on a copy of the dom with a new root with a success/failure be too slow?
02:27 nicomen -has
02:28 kaare joined #mojo
02:33 noganex joined #mojo
02:35 sri yes
02:36 sri if it's not easy with a new marker i could imagine a new op
02:36 sri like, if the tokenizer finds a combinator at the beginning, it adds a special op before the combinator that contains a reference to the current root element
02:37 sri could be very easy
02:38 jberger that's actually what I was just thinking of
02:38 jberger ::this
02:38 sri :)
02:40 sri i do still wonder if it could be done without big additions
02:40 nicomen :scope is what the w3c calls it no?
02:40 sri "big" being relative
02:41 sri even a scope op wouldn't be more than a few lines
02:41 jberger yeah, that looks right
02:41 jberger so implement the :scope op then add it if a combinator is first
02:42 jberger then all we need is the way to mark the return value as the :scope if it was a :match
02:42 jberger maybe that can just be grokked from the compiled form too though
02:42 jberger :has I guess
02:55 sri i don't think you need a marker
02:57 sri :has should end up with a "... if _select(...);" i imagine
02:58 sri similar to how :not works with _match()
02:58 sri the reentrant stuff already exists
03:01 irqq_ joined #mojo
03:07 jberger why doesn't adding this line work in sub _pc? : return 1 if $class eq 'scope';
03:11 inokenty-w joined #mojo
03:34 jberger oh, hmmm, ok I get that one
03:35 buu joined #mojo
03:39 davido_ joined #mojo
03:50 zivester joined #mojo
04:05 jberger ok, I'm losing the forest for the trees
04:05 jberger can someone check me here
04:05 jberger for :scope
04:06 jberger we can say that this should work, right?
04:06 jberger perl -Ilib -Mojo -E 'say x(q[<foo><bar>hi</bar></foo>])->at("foo")->at(":scope bar")->text'
04:06 jberger and equivalently
04:06 jberger perl -Ilib -Mojo -E 'say x(q[<foo><bar>hi</bar></foo>])->at("foo")->at(":scope > bar")->text'
04:32 davido___ joined #mojo
04:37 davido_ joined #mojo
04:47 melo joined #mojo
04:52 dustinm joined #mojo
05:07 jberger sri: this makes a minimal patch for scope: http://pastie.org/10349969
05:08 jberger a better test in _pc might be return $current eq $tree if $class eq 'scope';
05:08 bwf joined #mojo
05:08 jberger but then you have to pass $tree two levels down (not too bad though)
05:08 jberger but I can't think of when this wouldn't work
05:09 jberger anyway, from there its just a matter of injecting an implicit :scope at the front of the selector group if it would have started with a combinator
05:12 jberger hmmmm, need it in _parent too
05:16 jberger oh, _parent can trivially be implemented as _ancestor
05:36 good_news_everyon joined #mojo
05:36 good_news_everyon [mojo] jberger created css_scope (+1 new commit): http://git.io/v3XUv
05:36 good_news_everyon mojo/css_scope 1f49264 Joel Berger: enable :scope pseudoclass and relative selectors...
05:36 good_news_everyon left #mojo
05:37 jberger sri: ^^ no tests or doc yet, but I'm tired and that seems to be working
05:37 jberger so I'm calling it a good start
05:38 jberger by removing _ancestors (which you can do with or without the rest of my patch) it only adds 5 lines
05:38 jberger and maybe some of it is golfable
05:39 jberger sorry, removing _parent (see I am tired!)
05:43 jberger (also ignore my first pastie, it was crap)
05:43 jberger nn
05:51 sopanshewale joined #mojo
06:13 berov1 joined #mojo
06:20 dotandimet joined #mojo
06:39 damaya joined #mojo
06:39 Fink joined #mojo
06:57 Kogurr joined #mojo
07:20 dotandimet joined #mojo
07:24 eseyman joined #mojo
07:52 berov1 joined #mojo
07:52 Vandal joined #mojo
07:54 berov joined #mojo
08:42 zivester joined #mojo
09:27 dod joined #mojo
09:28 meshl joined #mojo
09:32 dod joined #mojo
09:41 punter joined #mojo
09:49 berov joined #mojo
09:49 zivester joined #mojo
09:51 jontaylor joined #mojo
09:52 Mikey joined #mojo
10:16 y1mmm joined #mojo
10:18 dod joined #mojo
11:01 neilhwatson joined #mojo
11:06 arpadszasz joined #mojo
11:20 eitz joined #mojo
11:22 mattp joined #mojo
11:37 jontaylor joined #mojo
11:48 berov left #mojo
11:49 dotandimet joined #mojo
11:56 jontaylor joined #mojo
12:04 jontaylor joined #mojo
12:18 arpadszasz joined #mojo
12:19 hernanGOA joined #mojo
12:36 dvinciguerra joined #mojo
12:56 marcusr joined #mojo
13:04 hummeleBop joined #mojo
13:10 ajr_ joined #mojo
13:16 jberger after a night's sleep I now worry about this mechanism: https://github.com/kraih/mojo/blob/1f49264ea95987869eb48847e4f6eed2da76f86a/lib/Mojo/DOM/CSS.pm#L31-L32
13:17 * sri just started looking at it
13:17 sri interesting you've combined _parent and _ancestors
13:18 sri is that less code even in master?
13:18 sri :scope as a psudo class is not something i would have done
13:18 * sri would have made it a special op
13:20 zivester joined #mojo
13:20 sri oh, it does work in master
13:22 good_news_everyon joined #mojo
13:22 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/v3MO4
13:22 good_news_everyon mojo/master 58d601c Sebastian Riedel: use a little less code in Mojo::DOM::CSS
13:22 good_news_everyon left #mojo
13:22 sri now i'm ruining the line stats of your patch :o
13:23 Grinnz lol
13:23 sri jberger++ # though
13:24 sri actually, the entire pseudo class stuff seems pointless
13:25 sri you can just match a scope op in _selector
13:25 sri no need to pass $tree around
13:25 sri at least not yet
13:26 sri jberger: does '~ foo' work?
13:29 good_news_everyon joined #mojo
13:29 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/v3Msa
13:29 good_news_everyon mojo/master 4f5f966 Sebastian Riedel: slightly more consistent arguments
13:29 good_news_everyon left #mojo
13:29 davido__ joined #mojo
13:33 sri hmm, reading the css4 selectors spec again, i guess :scope makes sense
13:34 sri actually, makes perfect sense
13:35 sri E:scope - an E element being a designated reference element
13:35 sri for the selector list in the docs
13:35 nicomen \o/
13:36 sri now the question becomes if we should do it already
13:36 jontaylor joined #mojo
13:37 sri currently we have one css4 selector http://mojolicio.us/perldoc/Mojo/DOM/CSS#E-foo-bar-i
13:37 sri it was added because it became a common problem
13:39 sri jberger: oh, and i agree, now after reading it all, i worry about the exact same thing as you :)
13:39 sri those two lines
13:40 sri everything else looks good
13:45 sri oh, btw. it looks like :scope has already been accepted by browser vendors https://developer.mozilla.org/en-US/docs/Web/CSS/%3Ascope
13:45 sri chrome 20, firefox 21 and safari 7
13:47 sri seems like a safe bet
13:48 sri jberger: with docs and tests it will get a +1 from me
13:48 sri maybe you'll even find a better solution for those two lines ;p
13:52 sri hahahaha http://i.imgur.com/Q3cUg29.gif
13:58 sri doesn't $selectors->[2][0][0] have to be something like $selectors->[$pos + 1][0][0]?
13:59 sri seems broken now
14:13 sri foe selectors like '> foo > bar > baz'
14:13 sri s/e/r/
14:18 gryphon joined #mojo
14:18 Repaster joined #mojo
14:20 * sri pokes jberger with a stick
14:34 Grinnz_ http://penny-arcade.com/comic
14:40 jberger Oh yes it should probably be relative to $pos
14:41 * jberger just dropped his car at the shop
14:41 Grinnz_ jberger: did it bounce?
14:43 jberger Basically
14:43 jberger New mechanic already discovered that old mechanic forgot to connect the engine fan
14:43 * jberger bangs head on desk
14:48 njlg joined #mojo
14:59 sri that kills the jberger
14:59 Grinnz_ heh
15:06 jberger joined #mojo
15:16 Kogurr joined #mojo
15:17 meshl joined #mojo
15:22 genio jberger: What's wrong with your car?
15:27 jberger it overheated on the expressway last night
15:27 jberger also the AC hasn't been working
15:27 jberger turns out that when you don't have the engine fan powered those things happen
15:28 jberger it was exciting, it was the full steam pouring out of the hood experience
15:29 jberger I had been going to chicago.pm, so when I came home instead I worked on the css stuff
15:31 sri yay?
15:31 genio There is no direct drive fan, just an electric one?  What kind of car is this?
15:32 genio There should be no cause to overheat if air is properly flowing through the radiator while in motion.  If you're stopped in traffic, yes, the fans are necessary.  A thermostat not opening properly is likely the problem if you were in motion
15:33 genio hopefully you didn't mess up your head gasket during the overheating process (unlikely, but possible).
15:35 till joined #mojo
15:35 jberger no I was in traffic
15:35 jberger and that was the problem
15:35 jberger mechanic is looking over the car now
15:36 till Hi, is this the intended behaviour with PUT request method?
15:36 till http://pastebin.com/BYZCCfHV
15:36 Repaster Repasted text from till: http://fpaste.org/255213/14395665/
15:36 jberger 2011 Saab 9-3 convertible
15:36 till why does PUT not change to GET after redirect (like POST)?
15:38 Grinnz_ http://programmers.stackexchange.com/questions/99894/why-doesnt-http-have-post-redirect/99966#99966
15:39 Grinnz_ also, redirect of anything other than GET or HEAD seems fishy
15:39 sri till: RFC 7231
15:40 Grinnz_ one of the comments on that answer mentions 7231
15:40 till Google Chrome sends a GET request after redirect
15:41 sri jberger: do you want to keep working on :scope or should i give cleaning it up a try?
15:41 jberger you are welcome to, I'm $working currently
15:41 sri hard work seems done
15:41 jberger I wouldn't be able to look until tonight
15:41 sri allright, after reading the spec i'm kinda interested now :)
15:42 genio https://rhn.redhat.com/errata/RHSA-2015-1623.html
15:42 jberger I don't care about git stats either, do as you see fit
15:42 jberger (I'll take a note in the Changes)
15:42 Grinnz_ till: https://tools.ietf.org/html/rfc7231#section-6.4.3
15:42 sri i've been meaning to look into making commits under other peeoples names... but that seems a bit odd
15:43 sri s/e//
15:43 sri "BUT I DID NOT COMMIT THAT!1"
15:43 Grinnz_ lol
15:43 jberger for tweaking up a working commit I don't see a problem with that, but yeah, some people might take it the wrong way
15:44 till ok thanks
15:44 sopanshewale joined #mojo
16:10 sri actually, it was $selectors->[$pos]
16:21 sri jberger: ohoh, think i've found a big flaw
16:22 sri ':scope + foo' cannot work
16:22 sri since it starts by searching children of the current scope
16:22 sri and to find siblings, it needs to start with the parent
16:23 jberger isn't that "out of scope" though?
16:23 jberger I would think that that is a bit of a non-sensical selector
16:23 sri you'd think so, but the selectors spec references all combinators
16:23 jberger hmmmm
16:23 jberger yeah
16:23 sri it's logical, but the spec seems unclear
16:24 jberger well can that be added to the _siblings stuff the way I added it to _ancestors
16:24 jberger ?
16:24 dod1 joined #mojo
16:27 sri seems super complicated
16:28 sri https://gist.github.com/anonymous/a5df9f736b9e033964f4
16:28 sri that's how far i got
16:28 sri second test fails
16:28 dod joined #mojo
16:30 sri actually i'm not sure it can work like that, since the initial match is already supposed to be a sibling
16:30 sri and we never even get the initial match
16:35 sri the spec makes me believe we don't have to support it
16:35 sri but i can't find a clear explanation
16:37 sri siblings are cleanrly outside the "scope"
16:37 sri this is hard, lets go shopping
16:40 sri https://drafts.csswg.org/selectors-4/#absolutizing
16:40 sri the rules are odd, but the results seem to match what we have now
16:40 * Grinnz_ doesnt even understand css3 that well
16:41 sri d'oh!
16:42 sri https://drafts.csswg.org/selectors-4/#relational
16:42 sri "dt:has(+ dt)"
16:42 sri so that was that
16:44 sri i give up for now
16:45 sri jberger: ^^
16:46 jberger I'm ok on punting on sibling behavior
16:47 * jberger would rather be shopping
16:47 * jberger is writing ansible roles, which is more fun than I thought
16:49 sri this confuses the hell out of me https://drafts.csswg.org/selectors-4/#evaluating-selectors
16:49 sri "If the selector is scope-contained, the selector match list is immediately filtered to contain only elements that are either the scoping root or descendants of the scoping root."
16:50 jberger at least this document shows that my take on the problem is correct
16:50 jberger "Note that if the selector is scoped, the scoping root is automatically taken as the :scope element, so it doesn’t have to be provided explicitly unless a different result is desired."
16:51 jberger and "If the selector is a relative selector, the set of :scope elements must not be empty."
16:51 sri it conflicts with the example "dt:has(+ dt)" though"
16:52 jberger yes your quote does seem to directly conflict with that
16:52 jberger again, I say we keep this and punt on relative sibling selectors
16:52 jberger this is going to be marked as experimental anyway right?
16:52 jberger ancestors is nicely defined
16:53 jberger siblings makes no sense and has conflicting examples
16:53 sri i was wondering if scope-contained and scope-filtered might have something to do with it
16:53 sri scope-filtered is for javascript apis, and scope-contained for stylesheets
16:54 sri "If the selector is scope-filtered, then after the selector is finished processing, the selector match list must be filtered to contain only elements that are either the scoping root or descendants of the scoping root."
16:54 sri that sounds exactly like the scope-contained description
16:54 jberger it does to me too
16:55 ajr_ joined #mojo
16:56 jberger the standards author: http://i0.kym-cdn.com/photos/images/original/000/305/210/19e.jpg
16:56 jberger also my previous mechanic
16:56 jberger :-P
16:58 sri ok, this one makes more sense
16:58 sri "scope-filtered: With this method of scoping, a selector matches an element only if the element is within the scope, even if other components of the selector are outside the scope. (A scoping element is considered to be in scope.)"
16:59 sri the difference seems to be that scope-filtered can use elements outside the scope during matching
16:59 sri but only actually match elements in scope
16:59 sri while scope-contained can't touch elements outside the scope
17:00 sri both totally rule out sibling matching though
17:00 sri that example must be wrong
17:00 mst CSS appears to be slowly turning into XPath
17:03 jberger everything inevitably becomes XML right?
17:03 mst and/or lisp
17:03 Grinnz_ all we can do is slow the conversion
17:03 mst XML is mostly just lisp, angle brackets, and stupidity shaken together in a bag, after all
17:05 jberger someone must have made an xml denoted list variant by now right?
17:05 mst also there's SXML/SXSLT
17:05 mst which are actually quite cute
17:05 sri this is beautiful https://www.youtube.com/watch?v=w8fQ-0uw-l0
17:07 sri "Wow! So you took a dark, violent, post apocalyptic world and added it to Mad Max Fury Road"
17:11 sopanshewale joined #mojo
17:28 irqq joined #mojo
17:31 njlg joined #mojo
17:44 * sri is unsure about :scope, if it wasn't for the incosnistent spec i'd give it a +1 :S
17:47 jberger remember, it isn't actually about :scope
17:47 jberger its about :has()
18:04 disputin joined #mojo
18:11 jnbek joined #mojo
18:14 PryMar56 joined #mojo
18:16 sri yea, the one with the bad example :S
18:33 jberger showed a cow orker (front end), he drooled
18:39 sri if only that one example wasn't there
18:41 jberger is there a list we can email for clarification?
18:42 sri https://github.com/w3c/csswg-drafts/pulls
18:42 sri haha, maybe send a pull request removing it
18:49 arpadszasz joined #mojo
18:49 sri Feedback:
18:49 sri www-style@w3.org with subject line “[selectors] … message topic …”
18:50 stephen joined #mojo
18:56 amon joined #mojo
19:04 njlg joined #mojo
19:12 disputin joined #mojo
19:14 jnbek joined #mojo
19:15 PopeFelix Stupid XML question.  not mojo related, but there's no traffic on xml-toolkit.  <Foo xmlns:bar="http://uri.fake" xmlns:baz="http://otheruri.fake" /> defines two namespace aliases, but doesn't actually say that any unqualified elements below are in either namespace, right?
19:17 * jberger vaguely remembers xmlns from the bank, cries
19:18 jberger (this was wpf/xaml, you would cry too)
19:18 sri there's specs for that
19:24 * Grinnz_ only knows as much about XML as about HTML
19:30 sri jberger: ok, it is definitely not an accident
19:30 sri i found lots of references in the list archives, sadly most discussions seem to appear offline
19:31 sri stuff like 'el:has(+sibling)' is totally valid
19:31 sri so, :scope and :has support are blocked for now :(
19:32 CromeDome joined #mojo
19:32 berov joined #mojo
19:33 jberger ok well leave the code there, I'll try to take a look sometime soon
19:34 sri that's the problem with draft specs :/
19:36 njlg joined #mojo
19:36 * jberger weeps
19:37 jberger and I can't even push purl into a well
19:37 sri that would be rude
19:38 sri on the plus side, you know how selectors work now \o/
19:38 jberger yeah
19:38 jberger I wonder which other dark corners of the codebase await me
19:39 sri and this got merged too :) https://github.com/kraih/mojo/commit/58d601cae49e8aea0b61a2f8d8d6a4ae49ab8af8
19:42 sri oh, :has seems to be up for removal from the css4 selectors spec :o
19:42 sri there's a vote going on
19:43 jberger WHY GOD?!
19:43 sri looks like it has been in the works since 1999
19:44 jberger http://33.media.tumblr.com/450a8d13c2d070774680758dd151840a/tumblr_inline_miqtxz9Obi1qz4rgp.png
19:45 sri https://lists.w3.org/Archives/Public/www-style/2015Jul/0296.html
19:45 sri well, at least case insensitive selectors is in
19:46 sri (the one we have)
19:46 sri :matches() requires one line to support (opposite of :not)
19:48 jberger https://lists.w3.org/Archives/Public/www-style/2015Jul/0298.html
19:48 jberger argh!
19:48 jberger oh well, if javascript can already do it, I guess a :has css selector isn't needed
19:49 jberger by that logic none of them are needed and we can all just go home!
19:49 sri funny thing, the jquery version of :has is super easy for us to support
19:49 jberger I think they just told us to do that
19:49 sri since there's no scope stuff
19:49 sri https://api.jquery.com/has/
19:50 sri it's just a _select() in Mojo::DOM::CSS
19:50 sri with a true/false check
19:50 sri oops, wrong link
19:51 sri https://api.jquery.com/has-selector/
19:51 sri that one
19:53 jberger ah and here they admit that the "implementation complexity" is only in the JIT
19:54 jberger and for the same reason as us
19:54 jberger going back up the tree
19:54 sri ;p
20:00 jberger there are some good rants on that thread
20:00 jberger some that apply to other 15 year old development projects that haven't materialized :-P
20:01 jberger (I can joke now that xmas is coming right?)
20:04 sri if you don't know Mojo::Template yet, that might also be an interesting dark corner
20:05 sri Mojo::DOM::HTML is surprisingly boring
20:06 sri Mojo::Server::Daemon could be informative too
20:07 sri you do know ioloop, reactor and router already...
20:07 Grinnz_ more reactorssss
20:11 hernanGOA joined #mojo
20:13 jberger I have poked around in template a few times
20:13 jberger and daemon a little bit
20:14 jberger ::HTML is just tag soup right?
20:16 sri yea, just lists of special cases
20:33 njlg joined #mojo
20:40 asm35 joined #mojo
20:42 njlg joined #mojo
20:44 dotandimet joined #mojo
20:44 DaTa joined #mojo
20:53 sri oh, and we still have no proposal for the documentation conventions section!
21:15 njlg joined #mojo
21:26 CromeDome joined #mojo
21:49 disputin joined #mojo
21:50 njlg joined #mojo
21:54 jzawodn_ joined #mojo
21:59 jzawodn joined #mojo
22:16 ajr_ joined #mojo
22:20 guru joined #mojo
22:30 ajr_ joined #mojo
22:57 disputin joined #mojo
23:02 disputin joined #mojo
23:02 meshl joined #mojo
23:06 howitdo joined #mojo
23:59 jontaylor joined #mojo

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary