#microformats 2013-11-07

2013-11-07 UTC
KevinMarks joined the channel
#
aaronpk tommorris: try gitlab! I have a chunk of repos on my own gitlab server
#
aaronpk http://gitlab.org/
Atamido, tantek, brianloveswords, snarfed and hober joined the channel
#
snarfed tommorris: just tried out mf2py
#
snarfed looks like a good start!
#
snarfed feel free to steal a few more test cases from https://github.com/snarfed/activitystreams-unofficial/tree/master/testdata if you want
#
@EmotiCodeDotNet HTML - hCard First, Middle, and Last Full Name HTML Markup http://www.emoticode.net/html/hcard-first-middle-and-last-full-name-html-markup.html #html #snippet (twitter.com/_/status/398351358149287936)
#
tantek edited /section (+752) "add HTML5 Doctor article that admits section causes most questions, easy to use incorrectly" (view diff)
tobyink joined the channel
#
tantek edited /section (+500) "add recent paciellogroup article about continued misuse of section element" (view diff)
#
tantek edited /section (+471) "add 2011 article noting confusion and lack of value for section" (view diff)
tswicegood joined the channel
#
KevinMarks Reveal.js uses section sensibly for slides http://kmpreso.herokuapp.com/?theme=serif#/8
pfefferle joined the channel
#
tantek KevinMarks - does it? It could just use <div>
#
KevinMarks Each slide makes sense as a section. You could just use div for nav too
#
tommorris making steady progress with http://code.tommorris.org/mf2py - now passes p- and u- properties.
#
tommorris edited /microformats2 (+106) "/* Implementations */ adding Python impl" (view diff)
#
tommorris edited /microformats2-parsing (+7) "/* Implementations */ adding Python impl" (view diff)
barnabywalters, adactio, tobyink, Hixie_, barnabywalters_ and twisted` joined the channel
#
tommorris hey barnabywalters - works progressing nicely on mf2py.
#
barnabywalters morning tommorris — you’re working on a python parser? great!
#
barnabywalters I’m wrangling python all day every day at work, so let me know if I can help at all
#
tommorris barnabywalters: http://code.tommorris.org/mf2py - patches welcome ;)
#
tommorris I'm mostly using it as a way to pass the time on my commute
#
barnabywalters cool, forking…
#
barnabywalters wow, bitbucket takes ages to fork projects
#
@ANDHRATEACHERS Msg2/2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem (twitter.com/_/status/398432250544410625)
iwaim_ and pfefferle joined the channel
#
@NaseerEseva ® Msg.2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem (twitter.com/_/status/398437918760398849)
#
@AiitaNews ® Msg.2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem (twitter.com/_/status/398438039334039552)
#
@GNGSMS * Msg2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem (twitter.com/_/status/398439014027771904)
#
@KHASEEM4AP Msg2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem (twitter.com/_/status/398439207376814080)
#
@VICTORJ13536519 Msg.2> *Conduct>TET *Upgaradtion of Pandts&PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem (twitter.com/_/status/398441572955463680)
#
@TUTAKTL Msg.2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem" (twitter.com/_/status/398442601499459584)
brianloveswords joined the channel
#
@PRASADNEWS Upgaradtion of Pandt, PETs CM agreed on HCards forMpl, AidedTchrs (in2ndPhas),TET& Upgradation &Examine Spl tchrs problames. (twitter.com/_/status/398445920389701632)
#
@Praveen53661150 STF DVK: Upgaradtion of Pandt, PETs CM agreed on HCards forMpl, AidedTchrs (in2ndPhas),TET& Upgradation &Examine Spl tchrs problames." (twitter.com/_/status/398446489048842240)
#
@UTFMCL Msg2/2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem (twitter.com/_/status/398448039733776384)
#
@PRASADINFOSYS Msg2/2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem (twitter.com/_/status/398448438901104641)
#
@JVRINFO Upgaradtion of Pandt, PETs CM agreed on HCards forMpl, AidedTchrs (in2nd Phase). TET& Upgradation &Examine Spl tchrs Problames." (twitter.com/_/status/398448727452844032)
#
@UTFPONNUR Msg 2 *Upgaradtion of Pandt,PETs *CM agreed on HCards for Mpl,AidedTchrs (in2ndPhase),TET&Upgradation&Examine SplTchrs Problem (twitter.com/_/status/398449229862940672)
#
@utfchittoor Msg2/2 *Conduct of TET *Upgradation of Pandt, PETs CM agreed on HCards for Mpl,AidedTchrs(in2ndPhas),TET&Upgradtin&Examine SplTchrs Problems (twitter.com/_/status/398449617450180608)
snarfed and TallTed joined the channel
#
@gsrmaths Upgaradtion of Pandt, PETs CM agreed on HCards forMpl, AidedTchrs (in2ndPhas),TET& Upgradation &Examine Spl tchrs problems." (twitter.com/_/status/398457145017909248)
#
@urduteachersknr Msg.2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem"" (twitter.com/_/status/398459041917444096)
#
@TEACHERS3AP Upgaradtion of Pandt, PETs CM agreed on HCards forMpl, AidedTchrs (in2ndPhas),TET& Upgradation &Examine Spl tchrs problames." (twitter.com/_/status/398460944046514176)
#
@STUKOLLIPARA Msg.2 *Conduct of TET *Upgaradtion of Pandt,PETs CM agreed on HCards forMpl,AidedTchrs(in2ndPhas),TET&Upgradation&Examine SplTchrs Problem" (twitter.com/_/status/398461653554954241)
#
@pigsonthewing @MagnusManske If you'd like to include hCard microformat, I can provide sample markup. (twitter.com/_/status/398461874809090048)
#
@UTFNRTAREA Msg2/2 *Conduct of TET *Upgaradtion of Pandt PETs CM agreed on HCards forMpl AidedTchrs(in2ndPhas) TET& Upgradation& Examine SplTchrsProblem (twitter.com/_/status/398462181777219584)
#
@UTFAP Msg2/2 *Conduct of TET *Upgaradtion of Pandt PETs CM agred on HCards forMpl AidedTchrs(in2ndPhas) Examine Spl tchers problems (twitter.com/_/status/398464878190411776)
#
@UPENDRAPRTU RT @PRASADNEWS: Upgaradtion of Pandt, PETs CM agreed on HCards forMpl, AidedTchrs (in2ndPhas),TET& Upgradation &Examine Spl tchrs problames. (twitter.com/_/status/398466071318908928)
jgay joined the channel
#
@UTFSANGAM Msg2/2 *Conduct of TET *Upgradation of Pandt, PETs CM agreed on HCards for Mpl,AidedTchrs(in2ndPhas),TET&Upgradtin&Examine SplTchrs Problems (twitter.com/_/status/398473185986756608)
#
@STUKOLLIPARA Upgaradtion of Pandt, PETs CM agreed on HCards forMpl, AidedTchrs (in2ndPhas),TET& Upgradation &Examine Spl tchrs problames." (twitter.com/_/status/398475962930962432)
snarfed joined the channel
#
snarfed morning!
#
snarfed barnabywalters, aaronpk: i had a few spare min earlier, so i started to run php-mf2-shim and activitystreams-unofficial on the same fb and twitter pages, just out of curiosity to see the diff
#
snarfed i had trouble running php-mf2-shim both with and without composer, almost certainly due to my inexperience w/php
#
snarfed i'll try again tonight, but just fyi if you all want to try too, since i'm sure you can do it in your sleep
#
snarfed feel free to diff against https://facebook-activitystreams.appspot.com/ and https://twitter-activitystreams.appspot.com/
#
snarfed i'll post a gist here if i get it working tonight
#
@Kiran0907 RT @PRASADNEWS: Upgaradtion of Pandt, PETs CM agreed on HCards forMpl, AidedTchrs (in2ndPhas),TET& Upgradation &Examine Spl tchrs problames. (twitter.com/_/status/398483223661252610)
#
barnabywalters snarfed: oh cool, just trying it out now
#
snarfed twitter api seems flaky/slow today. blame the ipo, clearly.
#
barnabywalters snarfed: mf2-json output looks good to me. one small fix: p-author shouldn’t be in the h-card type list
#
snarfed thanks, will fix
#
barnabywalters what’s with the weird looking uids? are they actually more useful than URIs?
#
snarfed i was particularly curious to see how it compares to the shims, not even for validation so much as just to see the variety
#
barnabywalters I’ll post a gist of the php-mf-shim output for my tweets — I don’t think there’s a huge difference
#
snarfed tag uris? eh, maybe not. i started using them a long time ago when some FB objects had no dedicated permalink, and others had multiple
#
snarfed not married to them
#
snarfed thanks!
#
barnabywalters https://gist.github.com/barnabywalters/7357443
#
snarfed re p-author, just to confirm, that's only a problem in the json, right?
#
barnabywalters snarfed: yep, the HTML should still have class="p-author h-card"
#
snarfed thanks for the gist, glad to have more data points
#
snarfed cool
#
barnabywalters snarfed: no problem! I need to make php-mf2-shim easier to use — maybe a command line tool like the main parser has
#
snarfed hell yes! i tried to adapt parse-mf2 but failed. only spent 10m on it though. someday i'll Learn PHP For Real
#
snarfed someday
#
barnabywalters oh, I’m so stupid — intertubes now has seamless support for mf2-shim
#
barnabywalters so you can go to http://waterpigs.co.uk/intertubes/
#
barnabywalters make a “fetch posts” module with a twitter.com address in, connect it to output or press refresh, and it’ll give you the parsed JSON on the right
#
barnabywalters well, not the raw parsed json — filtered to only include h-entries and normalised
#
snarfed ahhh perfect!
#
snarfed thanks
#
barnabywalters should work the same for facebook, but due to FBs limitations it can only parse public post permalink pages
#
barnabywalters and it might not even do that properly yet, as fb requires UA faking before it shows anything :/
#
barnabywalters which is where your tool is super-useful :)
#
snarfed definitely
jgay, lonimurar, tobyink, jschweinsberg and tantek joined the channel
#
lonimurar tantek: any idea how I can reach benward? I sent him a query about a week ago, he hasn't responded yet (and been idle since)
tobyink and JohnBeales joined the channel
#
tommorris tantek, barnabywalters: are we doing anything intelligent with value-class-pattern in uf2?
#
tommorris like, if I've got .p-tel > .type && .p-tel > .value, we just discard the type, right?
#
barnabywalters tommorris: well, the spec says to support it
#
barnabywalters but type doesn’t exist any more
#
tommorris so, it's just value
#
barnabywalters yep
#
tommorris surely the wiki should now say that VCP is in-practice deprecated in microformats2
#
tommorris what happens if we've got multiple .value classes?
#
barnabywalters currently I’m concatenating them on everything but dt-* properties
#
aaronpk this discussion is relevant
#
aaronpk see #indiewebcamp
#
barnabywalters dt-* is a special case — php-mf2 loops through all the .values and looks for whole or parts of datetimes
#
barnabywalters then it makes the most complete datetime it can from the parts
#
barnabywalters this is to support things like <span class="dt-published"><span class="value">8:00</span> on <span class="value">2013-12-02</span></span>
#
aaronpk barnabywalters: can you help cweiske in #indiewebcamp with that?
#
tommorris Hmm. surely that should be a time and date element? ;)
#
aaronpk tommorris: only in html5
#
tommorris oh christ, I'm going to have to implement an HTML5 dt parser. I hate dts
#
barnabywalters tommorris should work either way, I think I’m using <time> intelligently in value-class
#
tommorris There are things that aren't HTML5? I thought we'd just retroactively turned all HTML5 into HTML
#
barnabywalters tommorris: parser not necessary, list of patterns on the wiki
#
aaronpk xhtml is still strict
#
barnabywalters finds link
#
barnabywalters http://microformats.org/wiki/value-class-pattern#Date_and_time_parsing
#
tommorris aaronpk: ah, I'm running everything through html5lib whatever it pretends to be
#
aaronpk tantek: is this page basically deprecated? is that what the orange note is trying to say? http://microformats.org/wiki/datetime-design-pattern
#
tantek exactly what it says
#
aaronpk no not really
#
aaronpk it starts with an "if"
#
barnabywalters tommorris: see https://github.com/indieweb/php-mf2/blob/master/Mf2/Parser.php#L347
#
tantek right, "If you cannot use the <time> element"
#
tommorris that reminds me - for backwards compat, might it be an idea to treat acronym as equiv to abbr for uf2 parsing. ;)
#
tantek I'll amend it to make a stronger statement
#
barnabywalters it’s a bit ugly
#
aaronpk how about "Deprecated: Everything below this notice is deprecated and is archived here for historical purposes"
#
tommorris iirc acronym has been deprecated but some pedants will still use it (much as some people will still try and use XHTML 1.x and deny that we're in a tagsoup world)
#
aaronpk "Deprecated: Everything below this notice is deprecated and is archived here for historical purposes. If you cannot use the <time> element, see the [[value-class-pattern]]."
#
aaronpk much more concise and to the point
#
tantek deprecated isn't strong enough
#
tantek editing
#
tommorris deprecated plus the community strongly recommends you stay the hell away lest you let the demonic forces of old mailing list discussions come pouring out from the sleeping archives of yesteryear
#
barnabywalters that’s some pretty terrifying imagery :/
#
tommorris I deal with overwrought wikipedians. sometimes rhetorical excess is the only way to get through to them.
#
aaronpk oh right I keep forgetting that deprecated actually means "express disapproval of" and not the way it's used in software development
#
tantek aaronpk - even in software development, deprecated means won't be supported much longer.
#
tantek but is still supported for temporary backcompat
#
tantek then it's dropped
#
tantek edited /datetime-design-pattern (+233) "be more explicit about abbr datetime-design-pattern being dropped" (view diff)
#
aaronpk yeah true
#
tantek similar thing in specs
#
tantek ok updated
#
aaronpk great that's better
#
tantek thanks for the feedback guys - sorry that led cweiske astray
#
tantek is now off for a quick run.
#
tommorris Hmm. just found a link to upcoming.org in the value-class-pattern spec.
#
aaronpk *tear*
#
tantek tommorris - there are TONS of links to upcoming.org on the microformats wiki
#
tantek not sure exactly what to do about it
#
aaronpk change to archive.org links?
#
tantek if you come up with a good general approach (archive.org?)
#
tantek yeah
#
tommorris yeah, need a wikignome. when I've got a few minutes
#
tantek same with pownce links :/
#
tantek feel free to update those to archive.org links
#
tommorris also, here's a puzzler. <div class="h-card"><span class="p-tel"><span class="value">1234</span><span class="h-card"><span class="p-tel"><span class="value">5678</span></...
#
tommorris when parsing the first h-card rather than the improperly nested second h-card, how do you handle the VCP
#
barnabywalters who nests h-cards with p-tel properties?
#
tommorris barnabywalters: morons. unfortunately, some morons author html
#
barnabywalters so, php-mf2 handles this type of thing by doing depth-first parsing and then preventing already parsed property elements from being parsed again
#
barnabywalters not sure if I’m applying that to value-class yet, but it would handle it
#
tommorris it's not so much a compelling use case argument I'm looking for, but if I'm trying to parse the p-tel of the first h-card, on what basis do I not traverse down to the nesimproperly nested elements?
#
barnabywalters well, technically they’re not improperly nested, it’s just bad data
#
barnabywalters but depth first parsing handles this
#
tommorris probably worth checking against implementations
#
barnabywalters yeah, not sure if the spec actually specifies how to deal with this. I just handled it in the way which seemed most obvious
#
tommorris nicking that might be sensible, but we should probably formalise error handling
#
tantek what's the error?
#
barnabywalters tommorris: again, I don’t think there’s an error there
#
tommorris I'll pop it on the wiki
#
tantek when value-class parsing, nothing below the nested h-* formats shows up
#
tommorris tantek: that should be in the spec. ;)
#
tantek it's not?
#
tantek yes it should if it's not
#
barnabywalters unless the error is that the p-tel for the top level h-card was supposed to be the two .values concatenated
#
tantek or at least an FAQ
#
tommorris I'll be bold and add it to microformats2-parsing.
#
tantek tommorris - it's value-class-pattern specific
#
tantek add it there as an FAQ
#
barnabywalters in which case it’s just broken markup which looks like valid markup
#
barnabywalters and without trying to understand the content it’s impossible to correct, because it is kinda correct
#
tommorris it's valid markup, it just leads to unexpected results for the author.
#
barnabywalters best way to prevent this from happening is to offer good validation tools
#
tantek tommorris - microformats2-parsing already says to parse the element for the value-class-pattern in all the right places
#
tantek no need to edit microformats2-parsing
#
tantek you're talking about a value-class-pattern detail which I thought was already specified there
#
tantek if not, we either should add it explicitly to value-class-pattern, and even if so, add an FAQ to clarify it and make the answer more discoverable
#
tommorris yep, I'll check VCP and add it
#
tantek thank!
#
tommorris ah, it's sort of mentioned in VCP
#
tommorris "Descendants with class of value must not be parsed deeper than one level."
#
barnabywalters added issue for php-mf2: https://github.com/indieweb/php-mf2/issues/34
#
barnabywalters tommorris: oh, as in one HTML tree level?
#
barnabywalters that would fix the problem, seems a little arbitrary
#
barnabywalters anyway, food
krijnh joined the channel
#
tantek it was a deliberate simplification that met the needed use-cases
#
tantek so yeah, tommorris - add an FAQ for your question/example :)
#
aaronpk http://www.w3.org/WAI/wcag-curric/img/shake-a.gif
#
tommorris edited /value-class-pattern (+1261) "a rather headache-inducing question about microformats2 parsing of improperly nested h- objects within p- properties that contain .value elements" (view diff)
JohnBeales and barnabywalters joined the channel
#
KevinMarks Based on experience with feedparser, I'd say add the invalid/inconsistent markup as test cases, labelled as invalid, so that implementations can treat them consistently
#
KevinMarks That's also part of the lesson of html5
#
tommorris KevinMarks: yep. consistent error handling is important especially given the constraints of generic rather than specific microformat parsing
#
tommorris i.e. you can't rely on the vocabulary to give you hints
KevinMarks_, barnabywalters, gavinc, tobyink, kennyluck, tantek and jgay joined the channel