Options:
- # Session Start: Tue Jan 01 00:00:00 2008
- # Session Ident: #whatwg
- # [00:22] * Quits: weinig (n=weinig@adsl-76-203-73-177.dsl.pltn13.sbcglobal.net)
- # [00:25] * webben_ is now known as webben
- # [00:39] <gsnedders> http://geekninja.blogspot.com/2007/12/html5s-canvas-tag-are-we-using-it.html
- # [00:41] * Joins: weinig (n=weinig@c-71-198-176-23.hsd1.ca.comcast.net)
- # [00:50] <kig> as webcore is lgpl and canvas is a part of webcore and lgpl has a patent release clause...
- # [00:57] <Hixie> Philip`: .width doesn't just reflect the content attribute?
- # [00:59] <Philip`> Hixie: http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%0D%0A%3Cimg%20src%3Dimage%20style%3Dwidth%3A100%25%20onload%3Dw(this.width)%3E
- # [00:59] <Philip`> At least Opera and Firefox return the rendered width, and I think I remember IE doing that too
- # [00:59] <Hixie> well crap
- # [00:59] <Hixie> does html5 require that?
- # [00:59] <Philip`> I think it does
- # [01:00] <Philip`> HTML5 says "The DOM attributes height and width must return the rendered height and width of the image, in CSS pixels, if the image is being rendered, and is being rendered to a visual medium, or 0 otherwise. [CSS21]"
- # [01:01] <Philip`> (As far as I could see when I last wanted to do this, it's impossible to find the actual bitmap size of the image, which is a bit irritating)
- # [01:02] <Philip`> ((except for maybe adding it to the document and having it rendered and then seeing how big it is, but that's nasty))
- # [01:11] <Hixie> well cool, glad the spec is right
- # [01:12] <Hixie> and yes 'width' would be a way to do it then
- # [01:13] * Joins: jruderman (n=jruderma@c-67-180-15-227.hsd1.ca.comcast.net)
- # [01:19] * Joins: G0k (n=hmason@c-67-164-171-32.hsd1.co.comcast.net)
- # [01:56] <gsnedders> kig: only LGPL v.3. LGPL v.2 doesn't.
- # [01:58] <kig> 2 has a "if you can't redistro this code due to you wanting to destroy competitors with patent wars, you can't redistro this code"
- # [01:58] <kig> i.e. if you do distro the code, you don't want to patentwar
- # [01:59] <kig> but i don't know
- # [01:59] <kig> the canvas source files have a different license header too, wtf's up with that
- # [02:00] <kig> usual marketing bs? "webcore is LGPL super open source yay <smallprint>except for the parts that we edited</smallprint>"
- # [02:01] <gsnedders> kig: no, it doesn't say that. it says if you have restrictions that contradict the license, you can't redisribute it.
- # [02:01] <G0k> kig: i think basically the stuff they took from KTML is lgpl, but the new stuff is mostly BSD
- # [02:01] <gsnedders> kig: that doesn't affect Apple.
- # [02:01] <gsnedders> kig: their own stuff is BSD licensed, which is more lax than LGPL
- # [02:02] <G0k> *KHTML
- # [02:02] <gsnedders> kig: yeah, the canvas stuff is two clause BSD license.
- # [02:03] * kig pops a new beer
- # [02:03] <gsnedders> kig: the marketing bs is, "WebCore is LGPL super open source yay <smallprint>some stuff is even more free!</smallprint>"
- # [02:03] <gsnedders> kig: but there's nothing directly about patents in the LGPL
- # [02:04] <gsnedders> (nor the BSD license — the license header in the canvas files is the entire license)
- # [02:06] <kig> lgpl2 has "Therefore, we insist that any patent license obtained for a version of the library must be consistent with the full freedom of use specified in this license."
- # [02:06] <kig> in the preamble
- # [02:07] <G0k> yeah i disagree about that part gsnedders
- # [02:07] <G0k> iirc the LGPL basically says you implicitly license your patent rights to users of the code
- # [02:09] <Lachy> Happy New Year everyone
- # [02:10] <G0k> happy 2008 to gmt-1 and earlier people
- # [02:12] * Quits: nickshanks (n=nickshan@cpc2-clif1-0-0-cust535.nott.cable.ntl.com) (Read error: 110 (Connection timed out))
- # [02:21] * Quits: ROBOd (n=robod@89.122.216.38) ("http://www.robodesign.ro")
- # [02:25] * Quits: G0k (n=hmason@c-67-164-171-32.hsd1.co.comcast.net)
- # [02:29] * Joins: doublec (n=Chris_Do@203-97-173-6.cable.telstraclear.net)
- # [02:34] * Quits: tantek (n=tantek@70-13-176-76.area2.spcsdns.net) (Read error: 104 (Connection reset by peer))
- # [02:48] * Quits: kingryan (n=kingryan@dsl092-002-056.sfo1.dsl.speakeasy.net)
- # [02:50] * Quits: Lachy (n=Lachlan@cm-84.215.54.100.getinternet.no) ("Leaving")
- # [02:51] * Joins: Lachy (n=Lachlan@cm-84.215.54.100.getinternet.no)
- # [02:52] * Quits: Lachy (n=Lachlan@cm-84.215.54.100.getinternet.no) (Client Quit)
- # [02:52] * Joins: Lachy (n=Lachlan@cm-84.215.54.100.getinternet.no)
- # [03:07] * Quits: dbaron (n=dbaron@pool-72-94-185-124.phlapa.fios.verizon.net) ("8403864 bytes have been tenured, next gc will be global.")
- # [03:59] * Joins: tantek (n=tantek@70-13-200-58.area2.spcsdns.net)
- # [04:36] * Quits: tndH (i=Rob@adsl-83-100-132-169.karoo.KCOM.COM) ("ChatZilla 0.9.79-rdmsoft [XULRunner 1.8.0.9/2006120508]")
- # [04:48] * Joins: dglazkov (n=dglazkov@adsl-074-229-248-021.sip.bhm.bellsouth.net)
- # [05:00] * Joins: roc (n=roc@121-72-34-26.dsl.telstraclear.net)
- # [05:26] * Quits: roc (n=roc@121-72-34-26.dsl.telstraclear.net)
- # [05:57] * MacDomeOut is now known as MacDome
- # [06:01] * Quits: tantek (n=tantek@70-13-200-58.area2.spcsdns.net)
- # [06:04] * Quits: dglazkov (n=dglazkov@adsl-074-229-248-021.sip.bhm.bellsouth.net) ("durr...")
- # [07:20] * Joins: roc (n=roc@121-72-34-26.dsl.telstraclear.net)
- # [07:25] * Quits: doublec (n=Chris_Do@203-97-173-6.cable.telstraclear.net) (Read error: 110 (Connection timed out))
- # [07:28] * Quits: roc (n=roc@121-72-34-26.dsl.telstraclear.net)
- # [08:50] * Joins: tantek (n=tantek@70-13-107-175.area2.spcsdns.net)
- # [08:56] * Joins: roc (n=roc@121-72-34-26.dsl.telstraclear.net)
- # [10:10] * Quits: wakaba (n=w@77.137.148.210.dy.bbexcite.jp) (Read error: 110 (Connection timed out))
- # [10:19] * Joins: ROBOd (n=robod@89.122.216.38)
- # [10:28] * Joins: doublec (n=Chris_Do@203-97-173-6.cable.telstraclear.net)
- # [10:30] * Quits: roc (n=roc@121-72-34-26.dsl.telstraclear.net)
- # [10:38] * Joins: anne-mac (n=annevk@5352CE6F.cable.casema.nl)
- # [11:25] * Joins: nickshanks (n=nickshan@cpc2-clif1-0-0-cust535.nott.cable.ntl.com)
- # [12:35] * MacDome is now known as MacDomeSleep
- # [12:36] * Quits: anne-mac (n=annevk@5352CE6F.cable.casema.nl) (Read error: 110 (Connection timed out))
- # [12:38] * Quits: doublec (n=Chris_Do@203-97-173-6.cable.telstraclear.net) ("ChatZilla 0.9.79-rdmsoft [XULRunner 1.8.0.9/2006120508]")
- # [12:54] * Quits: inimino (n=inimino@c-75-71-88-233.hsd1.co.comcast.net) (Read error: 113 (No route to host))
- # [13:05] * Quits: MacDomeSleep (n=eric@c-69-181-78-198.hsd1.ca.comcast.net)
- # [13:05] * Joins: hdh (n=hdh@58.187.89.61)
- # [13:07] <gsnedders> kig: the preamble isn't the license. it has no legal meaning.
- # [13:07] <gsnedders> Lords Of HTML, forgive me, for I said "alt tag".
- # [13:09] <kig> so use the opengl canvas instead..
- # [13:11] <Philip`> Why would that be less likely to be affected by patents?
- # [13:12] <kig> because it's opengl
- # [13:14] <kig> unless the patent is the usual heh heh patent of the form "hypertext document space in which there may be pixels and which may be programmatically changed"
- # [13:14] <kig> which it probably is
- # [13:14] <Philip`> But it's OpenGL inside a web browser, which maybe is an innovative new idea and patentable
- # [13:15] <kig> MPL doesn't have a patent release?
- # [13:15] <Philip`> (I can't see why 2D canvas does anything that isn't trivially obvious, since it's just a wrapper around boring standard 2D graphics libraries)
- # [13:17] <kig> because some legal eagle at apple said so
- # [13:18] <kig> waiting for adobe to sue all browser makers for having vector graphics that is not flash/pdf
- # [13:19] <kig> and my catmull-rom spline drawer needs more work
- # [13:20] <Lachy> open source software is generally incompatible with patents, so I doubt Mozilla has any patents for their opengl canvas implementation
- # [13:20] <gsnedders> kig: if you read the whole post, they have no issue if it moved to a standards body with a patent policy (i.e., the W3C). Once FPWD is published there, after 90 days, Apple's requirements are met
- # [13:20] <Philip`> kig: I'd guess the W3C patent whatsits make it hard for Adobe to complain about patents on things like SVG
- # [13:20] <Philip`> Lachy: People other than Mozilla still could, though
- # [13:21] <Philip`> Lachy: Also, some Mozilla software does have patents
- # [13:21] <Philip`> http://developer-cluster.mozilla.org/en/docs/SpiderMonkey_Internals:_Thread_Safety#Patent
- # [13:22] <Lachy> yeah, royalty free patents are ok
- # [13:25] <Philip`> kig: Is there any chance your SVG renderer could output things (triangles?) for OpenGL to draw? I'd quite like to have SVG embedded in X3D embedded in XHTML...
- # [13:26] <kig> if you have a tesselator/renderer/magic for quadratic/cubic bezier splines, sure
- # [13:26] <kig> cubic'll suffice, quadratics can be done with those
- # [13:27] <Philip`> That sounds not impossible
- # [13:27] <kig> http://www.mdk.org.pl/2007/10/27/curvy-blues
- # [13:27] <kig> gradients and stroking may be, ah, more interesting
- # [13:28] <Philip`> I'd guess gradients wouldn't be too hard with pixel shaders
- # [13:30] <kig> guess i should move all the rendering primitives under the same backend, then implementing the ops in that would get you going
- # [13:31] <kig> or write a canvas context that calls opengl
- # [13:34] <kig> best performance by caching the paths generated by svg, compile into vertex array
- # [13:34] <kig> s/generated by/in/
- # [13:36] <kig> now i'm compiling the SVG d="..." -paths into [['moveTo', [0,4]], ['lineTo', [400,230]], ...] and for (var i=0;i<segs.length;i++) ctx[segs[i][0]].apply(ctx, segs[i][1])
- # [13:38] <kig> but with ogl you could store the geometry in a VBO, which should make it fly
- # [13:39] <kig> http://glimr.rubyforge.org/cake/canvas.html#CatmullRom hooray!
- # [13:42] <Philip`> kig: Would it be better/worse to compile into eval('function(ctx){ctx.moveTo(0,4), ...}')
- # [13:43] <kig> haven't tried, that's really evil though :)
- # [13:43] <Philip`> Could you do constant-velocity movement along the spline?
- # [13:45] <kig> when i find the algo to compute a segment's length
- # [13:45] * Philip` likes evil
- # [13:52] * Joins: tndH_ (i=Rob@adsl-83-100-132-169.karoo.KCOM.COM)
- # [13:52] * tndH_ is now known as tndH
- # [13:57] <kig> i had a fleeting thought of parsing truetype fonts with javascript and compiling them into quadraticCurveTo:s but then i took a look at the specs
- # [13:58] <Philip`> Could something like FreeType provide you with a list of curves?
- # [13:58] <kig> it could
- # [13:58] <Philip`> (then precompute the JS-compatible font representation on a server)
- # [14:02] <Philip`> Do you happen to know if Opera's drawImage(svgimage, ...) can draw SVG text onto a canvas?
- # [14:02] <Philip`> (I can't remember if I ever tested that or not)
- # [14:02] <kig> haven't tried, i guess it should
- # [14:03] <Philip`> I think it refused to draw foreignContent onto canvas, so HTML text won't work :-(
- # [14:05] <kig> if you look at the svgparser transformed text (e.g. coord systems.. -> skew) things with a webkit nightlie, it has madness (the text element has -webkit-transform matrix)
- # [14:13] <kig> time to read freetype docs
- # [14:14] <kig> oh oh idea
- # [14:14] <kig> compute the JS-compatible font representation on a server on the fly
- # [14:16] <kig> ok, it's not much of a change, is it :|
- # [14:17] <kig> or have a list of fonts, query with js / server-side include
- # [14:19] <kig> font server ..
- # [14:22] <Philip`> You'd probably want the font server to only send the characters are needed, and not the other tens of thousands of glyphs in the original font file
- # [14:23] <Philip`> (so it couldn't just be a static file server)
- # [14:24] <Philip`> ((I'm assuming the JS representation of the font would be pretty inefficient, so you couldn't send everything, but maybe I'm wrong))
- # [14:26] <kig> gzipped array of numbers, shouldn't be much larger than the original font file (? only one way to find out...)
- # [14:26] * Joins: Hemebond (n=Hemebond@ip-118-90-36-41.xdsl.xnet.co.nz)
- # [14:27] <Philip`> Is the original font file just a list of curves, rather than something more programmatic?
- # [14:27] * Joins: dglazkov (n=dglazkov@adsl-074-229-248-021.sip.bhm.bellsouth.net)
- # [14:27] <Philip`> (ignoring all the hinting and stuff, which presumably could never work in SVG-in-canvas)
- # [14:28] <kig> truetype fonts have the curves and then some nutty bytecode system for doing the hinting afaik
- # [14:28] <kig> i don't really know
- # [14:29] <kig> i guess they have something extra to handle all the different types of type
- # [14:32] * Quits: gsnedders (n=gsnedder@host86-138-198-209.range86-138.btcentralplus.com) ("Partying in teh intarwebs")
- # [14:35] <Philip`> (Is there a better touchpad driver for Windows in Boot Camp? I don't like taking twenty seconds to right-click on a folder just because right-clicking requires two fingers on the touchpad which then makes the window scroll wildly up and down even if I keep my fingers almost stationary)
- # [14:48] * Joins: webben_ (n=benh@91.84.19.165)
- # [14:54] * Quits: hdh (n=hdh@58.187.89.61) (Remote closed the connection)
- # [14:59] * Quits: webben (n=benh@nat/yahoo/x-10fac8bad40e840f) (Read error: 110 (Connection timed out))
- # [15:04] * Quits: psa (n=yomode@71.93.19.66) (Remote closed the connection)
- # [15:06] * Quits: ROBOd (n=robod@89.122.216.38) (Read error: 110 (Connection timed out))
- # [16:10] * Quits: dglazkov (n=dglazkov@adsl-074-229-248-021.sip.bhm.bellsouth.net)
- # [16:46] <kig> just what i needed: http://www.geometrictools.com/Documentation/MovingAlongCurveSpecifiedSpeed.pdf
- # [16:52] * Joins: gsnedders (n=gsnedder@host86-138-198-209.range86-138.btcentralplus.com)
- # [17:01] * Joins: hdh (n=hdh@58.187.89.61)
- # [17:06] * Joins: anne-mac (n=annevk@5352CE6F.cable.casema.nl)
- # [17:18] <anne-mac> Philip`, where in ECMAScript does it state that \0 is to be stripped?
- # [17:18] <anne-mac> Philip`, presumably the Acid3 test bases that on some statement?
- # [17:29] * Joins: zcorpan (n=zcorpan@c-cb21e353.1451-1-64736c12.cust.bredbandsbolaget.se)
- # [17:34] * Joins: dglazkov (n=dglazkov@adsl-074-229-248-021.sip.bhm.bellsouth.net)
- # [17:39] <Philip`> anne-mac: As far as I'm aware, it shouldn't be stripped - it's just a normal character, so "foo\0bar" != "foo"
- # [17:40] <Philip`> (except Opera (and no other browser) tends to strip off everything after the first \0 presumably because it's treated like a C string)
- # [17:40] <Philip`> (and so I'm assuming Opera is wrong, because it's the odd one out and it doesn't seem sensible behaviour)
- # [17:41] <anne-mac> other browsers strip \0 characters
- # [17:42] <anne-mac> I guess one can claim that treating \0 as terminating character is wrong, but stripping it out isn't in the specs either
- # [17:43] <Philip`> I don't see it being stripped out in any browser
- # [17:43] <Philip`> (except Opera)
- # [17:45] <Philip`> (...and IE - oops)
- # [17:46] <anne-mac> Firefox and Safari at least strip it when doing alert('x\0x')
- # [17:46] <Philip`> http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%0A%3Cbody%20onload%3Dw(document.getElementById('fo%5C0o'))%3E%0A%3Cspan%20id%3Dfo%3E%3C%2Fspan%3E%0A%3Cdiv%20id%3Dfoo%3E%3C%2Fdiv%3E%0A
- # [17:48] <Philip`> Oh, that sounds more likely to be a problem with alert than with the general DOM API system
- # [17:48] <anne-mac> https://bugzilla.mozilla.org/show_bug.cgi?id=310037
- # [17:49] <Philip`> alert('x\0x') in Safari on Windows outputs just "x", and FF3 on Windows says "xx"
- # [17:49] <anne-mac> Safari on Mac says "xx" ...
- # [17:50] <Philip`> "Version 3.0.4 (523.15)"
- # [17:51] * Philip` doesn't know if newer versions changed it
- # [17:51] <anne-mac> Version 3.0.4 (5523.10)
- # [17:51] <Philip`> That sounds futuristic
- # [17:51] <anne-mac> it's a literal copy
- # [17:52] * Philip` wonders what the extra 5 means
- # [18:02] <Philip`> WebKit nightly on Windows seems to still say "x"
- # [18:03] <anne-mac> maybe stripping \0 is done at the OS level on Mac and Windows doesn't do that
- # [18:10] * Quits: zcorpan (n=zcorpan@c-cb21e353.1451-1-64736c12.cust.bredbandsbolaget.se) (Read error: 110 (Connection timed out))
- # [18:12] * Joins: ROBOd (n=robod@89.122.216.38)
- # [18:17] * Joins: zcorpan (n=zcorpan@c-cb21e353.1451-1-64736c12.cust.bredbandsbolaget.se)
- # [18:17] * Joins: jgraham (n=james@81-86-215-9.dsl.pipex.com)
- # [18:45] <gsnedders> Philip`: extra 5 is on leopard, but not in UA string
- # [18:47] <gsnedders> anne-mac: seems to be within OS string type for alerts
- # [19:01] * Quits: Lachy (n=Lachlan@cm-84.215.54.100.getinternet.no) (Read error: 104 (Connection reset by peer))
- # [19:01] * Joins: Lachy_ (n=Lachlan@cm-84.215.54.100.getinternet.no)
- # [19:12] * Quits: zcorpan (n=zcorpan@c-cb21e353.1451-1-64736c12.cust.bredbandsbolaget.se) (Read error: 110 (Connection timed out))
- # [19:12] * Quits: anne-mac (n=annevk@5352CE6F.cable.casema.nl) (Read error: 110 (Connection timed out))
- # [19:16] * Quits: Lachy_ (n=Lachlan@cm-84.215.54.100.getinternet.no) (Read error: 104 (Connection reset by peer))
- # [19:16] * Joins: Lachy__ (n=Lachlan@cm-84.215.54.100.getinternet.no)
- # [19:21] * Parts: hdh (n=hdh@58.187.89.61)
- # [19:28] * Joins: G0k (n=hmason@c-67-164-171-32.hsd1.co.comcast.net)
- # [19:40] <hsivonen> zcorpan: "If you the automatic choice of parser ..." fixed. thanks
- # [19:50] * Quits: nickshanks (n=nickshan@cpc2-clif1-0-0-cust535.nott.cable.ntl.com)
- # [20:07] * Joins: MacDome (n=eric@c-69-181-78-198.hsd1.ca.comcast.net)
- # [20:16] * Quits: bzed (n=bzed@devel.recluse.de) (Remote closed the connection)
- # [20:16] * Joins: bzed (n=bzed@devel.recluse.de)
- # [20:17] * Joins: zcorpan (n=zcorpan@c-cb21e353.1451-1-64736c12.cust.bredbandsbolaget.se)
- # [20:18] <zcorpan> http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%3Cstyle%3E[align%3D%22A%22]%2C%20[align%3D%22B%22]%2C%20[align%3D%22i%22]%2C%20[align%3D%22%C4%B1%22]%20{%20background%3Ayellow%20}%3C%2Fstyle%3E%3Cp%20align%3Da%3Ea%20A%3Cp%20align%3Da%3EB%20b%3Cp%20align%3D%C4%B0%3E%C4%B0%20i%3Cp%20align%3DI%3EI%20%C4%B1
- # [20:19] <G0k> oo
- # [20:20] <zcorpan> um, make that http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%3Cstyle%3E[align%3D%22A%22]%2C%20[align%3D%22b%22]%2C%20[align%3D%22i%22]%2C%20[align%3D%22%C4%B1%22]%20{%20background%3Ayellow%20}%3C%2Fstyle%3E%3Cp%20align%3Da%3Ea%20A%3Cp%20align%3DB%3EB%20b%3Cp%20align%3D%C4%B0%3E%C4%B0%20i%3Cp%20align%3DI%3EI%20%C4%B1
- # [20:23] * Quits: G0k (n=hmason@c-67-164-171-32.hsd1.co.comcast.net)
- # [20:23] <zcorpan> http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%3Cstyle%3E[align%3D%22A%22]%23a%2C%20[align%3D%22b%22]%23b%2C%20[align%3D%22i%22]%23c%2C%20[align%3D%22%C4%B1%22]%23d%20{%20background%3Ayellow%20}%3C%2Fstyle%3E%3Cp%20id%3Da%20align%3Da%3Ea%20A%3Cp%20id%3Db%20align%3DB%3EB%20b%3Cp%20id%3Dc%20align%3D%C4%B0%3E%C4%B0%20i%3Cp%20id%3Dd%20align%3DI%3EI%20%C4%B1 even...
- # [20:29] * Joins: roc (n=roc@121-72-34-26.dsl.telstraclear.net)
- # [21:12] * Quits: zcorpan (n=zcorpan@c-cb21e353.1451-1-64736c12.cust.bredbandsbolaget.se) (Read error: 110 (Connection timed out))
- # [21:32] * Quits: bzed (n=bzed@devel.recluse.de) (Read error: 104 (Connection reset by peer))
- # [21:32] * Joins: bzed (n=bzed@devel.recluse.de)
- # [21:50] <Philip`> If I have some XML (specifically XSLT) which uses QNames in attribute values, and I want to extract a subtree and move it into a new document and want the QNames to continue working, is there a proper way to do that?
- # [21:54] <hsivonen> Should I remove text/xsl support from Validator.nu even in the lax mode?
- # [21:57] <hsivonen> I see application/xslt+xml in mime.types on Leopard. Is that an Apple addition or does Apache now qualify unregistered types in the default list?
- # [21:58] <gsnedders> hsivonen: it'll be Apple
- # [21:58] <hsivonen> ok
- # [21:58] * Quits: ROBOd (n=robod@89.122.216.38) ("http://www.robodesign.ro")
- # [21:59] <gsnedders> hsivonen: Apache policy hasn't changed (and Apple will've taken the Apache version in Leopard before WWDC in June)
- # [22:06] * Joins: Thezilch[FH] (n=fuz007@ip68-111-154-116.sd.sd.cox.net)
- # [22:12] * Quits: weinig (n=weinig@c-71-198-176-23.hsd1.ca.comcast.net)
- # [22:13] <takkaria> http://ewx.livejournal.com/459902.html
- # [22:14] * Quits: Lachy__ (n=Lachlan@cm-84.215.54.100.getinternet.no) (Read error: 104 (Connection reset by peer))
- # [22:14] <MacDome> takkaria: would you like to file a bug? http://webkit.org/quality/reporting.html
- # [22:14] <MacDome> it will end up being a CFNetwork bug
- # [22:14] <MacDome> which dates back from NeXT in part :)
- # [22:14] * Joins: Lachy__ (n=Lachlan@cm-84.215.54.100.getinternet.no)
- # [22:14] <takkaria> heh
- # [22:15] <takkaria> OK, I will
- # [22:15] <takkaria> I bet it's done like that for a reason though
- # [22:15] <MacDome> takkaria: thank you. don't be surprised if your bug is moved into Radar
- # [22:15] <MacDome> since CFNetwork is handled by a non-opensource team
- # [22:15] <MacDome> takkaria: but bugs are the only way to get fixes :) not sure if that's your blog post (I'm assuming not), but blog posts certainly don't get fixed :)
- # [22:15] <takkaria> it's not mine, no
- # [22:16] <MacDome> takkaria: anyway, thanks in advance for the bug~!
- # [22:16] * MacDome goes back to fixing webkit bugs
- # [22:17] <takkaria> I just saw it pasted in #netsurf (channel for a small browser written in C) and thought it would be of interest here
- # [22:17] <MacDome> takkaria: of most interst in bugzilla :) and secondly #webkit, but here works too :)
- # [22:21] * Joins: virtuelv (n=virtuelv@233.80-203-100.nextgentel.com)
- # [22:23] * Quits: Thezilch (n=fuz007@ip68-111-154-116.sd.sd.cox.net) (Connection timed out)
- # [22:30] * Quits: roc (n=roc@121-72-34-26.dsl.telstraclear.net)
- # [22:33] * Joins: zcorpan (n=zcorpan@c-cb21e353.1451-1-64736c12.cust.bredbandsbolaget.se)
- # [22:36] * gsnedders realises he is capitalising MUST and SHOULD and MAY in an email… Need less to do with RFC 2119…
- # [22:44] <takkaria> everyone should use OpenID, I'm sick of registering more than usual today
- # [22:44] <Philip`> Everyone should allow anonymous access
- # [22:47] * Quits: digx (n=rick@c-76-109-201-140.hsd1.fl.comcast.net) ("Leaving...")
- # [22:54] <jgraham> Philip`: That doesn't make sense for e.g. flickr
- # [22:54] <jgraham> (at least, not for the part you currently hav to register for)
- # [22:56] * jgraham notes that the word charset has 7 letters
- # [22:58] <jgraham> Philip`: Did you have a single page copy of the issues list somewhere?
- # [22:59] <Philip`> jgraham: canvex.lazyilluminati.com/misc/cgi/issues.cgi/expand though it's maybe a little slow to generate
- # [22:59] * Joins: weinig (n=weinig@c-71-198-176-23.hsd1.ca.comcast.net)
- # [23:00] <Philip`> That charset thing has been mentioned before
- # [23:00] <jgraham> Ah, I thought it would have been but I thought I should check
- # [23:01] <Philip`> http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2007-June/011596.html
- # [23:01] <Philip`> http://canvex.lazyilluminati.com/misc/cgi/issues.cgi/message/%3Cea09c0d10706011740l797240ebncf31cf1403b0c5cb@mail.gmail.com%3E - hooray, Google has indexed the issues list
- # [23:05] * Philip` should perhaps add a search box
- # [23:11] * Quits: zcorpan (n=zcorpan@c-cb21e353.1451-1-64736c12.cust.bredbandsbolaget.se) (Read error: 110 (Connection timed out))
- # [23:19] * Quits: virtuelv (n=virtuelv@233.80-203-100.nextgentel.com) ("Leaving")
- # [23:22] * Joins: roc (n=roc@121-72-34-26.dsl.telstraclear.net)
- # [23:24] * Quits: roc (n=roc@121-72-34-26.dsl.telstraclear.net) (Client Quit)
- # [23:31] * Joins: rubys (n=rubys@cpe-075-182-087-110.nc.res.rr.com)
- # [23:31] <rubys> Phillip`, hsivonen: fixed. Thanks!
- # [23:35] <Philip`> rubys: http://intertwingly.net/blog/?q=%00%80 looks unintended
- # [23:39] <Philip`> Also, it's annoying how everyone keeps fixing their code so my attempted demonstrations that XML is hard stop working :-p
- # [23:40] <hsivonen> rubys: cool
- # [23:45] <MacDome> Hixie: you reference ES4 with your \3 argument
- # [23:45] <MacDome> I found this in ES3: Informative comments: An escape sequence of the form \ followed by a nonzero decimal number n matches the result of the nth set of capturing parentheses (see 15.10.2.11). It is an error if the regular expression has fewer than n capturing parentheses. If the regular expression has n or more capturing parentheses but the nth one is undefined because it hasn't captured anything, then the backreference always succeeds.
- # [23:45] <Hixie> er i meant es3
- # [23:45] <MacDome> which doesn't quite seem to answer the question
- # [23:45] <MacDome> Hixie: could you be more specific in your ES3 reference?
- # [23:45] <Hixie> sure hold on
- # [23:45] <MacDome> Hixie: thank you
- # [23:48] <rubys> Phillip`: again, fixed. (Sorry! :-))
- # [23:49] * hsivonen just read a paper explaining the dire consequences of backreferences in "regular expressions"
- # [23:52] * Quits: gsnedders (n=gsnedder@host86-138-198-209.range86-138.btcentralplus.com) ("Partying in teh intarwebs")
- # [23:54] <Philip`> rubys: Seems to work - thanks :-)
- # [23:54] <rubys> try some UTF-8. :-)
- # [23:56] <Philip`> http://intertwingly.net/blog/?q=xyzzy%ef%bf%bf
- # [23:57] <Philip`> ...gives a parse error in Opera, not Firefox
- # [23:57] <hsivonen> Philip`: I get the YSoD in Firefox 2
- # [23:58] <hsivonen> not in Minefield, though
- # [23:58] <Philip`> Ah - FF3 automatically converts it to %EF%BF%BD
- # [23:59] <Philip`> (judging by what it shows in the magical address bar thing)
- # Session Close: Wed Jan 02 00:00:00 2008
The end :)