#instantbird log on 07 20 2010

All times are UTC.

00:04:46 --> micahg has joined #instantbird
00:07:52 <-- micahg has quit (Input/output error)
00:10:55 --> micahg has joined #instantbird
00:30:00 --> skeledrew has joined #instantbird
00:52:15 <-- clokep has quit (Quit: Instantbird 0.3a1pre)
00:52:19 --> clokep has joined #instantbird
01:04:15 <-- GeekShadow has quit (Connection reset by peer)
01:07:16 <-- micahg has quit (Quit: Leaving.)
01:09:44 --> hicham has joined #instantbird
01:34:02 <-- skeledrew has quit (Ping timeout)
01:45:57 --> skeledrew has joined #instantbird
02:23:26 --> DGMurdockIII has joined #instantbird
02:31:24 <-- DGMurdockIII has quit (Quit: get satisfied! • :: www.unitedservers.de ««« (Gamers.IRC) »»» gamersirc.net ::)
03:17:15 <-- hicham has quit (Client exited)
03:24:21 <-- clokep has quit (Quit: Instantbird 0.3a1pre)
05:22:30 --> tymerkaev has joined #instantbird
05:48:53 <-- tymerkaev has quit (Client exited)
05:52:15 <-- flo has quit (Quit: Instantbird 0.3a1pre)
05:55:34 <-- testib has quit (Ping timeout)
06:20:23 <-- skeledrew has quit (Quit: Instantbird 0.2)
07:07:51 --> tymerkaev has joined #instantbird
07:27:29 <-- tymerkaev has quit (Ping timeout)
07:33:32 --> testib has joined #instantbird
07:37:38 --> flo has joined #instantbird
07:37:38 * ChanServ sets mode +qo flo flo 
07:55:09 <-- flo has quit (Quit: Instantbird 0.3a1pre)
07:57:05 <-- testib has quit (Ping timeout)
09:59:04 --> Mic has joined #instantbird
10:05:08 --> flo has joined #instantbird
10:05:08 * ChanServ sets mode +qo flo flo 
10:16:33 --> testib has joined #instantbird
10:30:28 --> GeekShadow has joined #instantbird
10:59:46 --> clokep has joined #instantbird
11:13:54 <-- clokep has quit (Quit: Instantbird 0.3a1pre)
11:13:58 --> clokep has joined #instantbird
11:17:00 <Mic> Is anyone else getting an error when connecting with and jabber.org account?
11:17:10 <Mic> "Invalid authzid" ..
11:23:52 <clokep> Only Jabber account I connect to is GTalk an that was fine for me.
11:33:10 <-- clokep has quit (Quit: Instantbird 0.3a1pre)
11:36:21 --> hicham has joined #instantbird
12:00:29 --> clokep has joined #instantbird
12:54:48 --> Even has joined #instantbird
12:54:49 * ChanServ sets mode +o Even 
13:02:36 <-- Mic has quit (Input/output error)
13:09:02 --> skeledrew has joined #instantbird
13:28:53 --> iBool has joined #instantbird
13:29:04 <-- iBool has left #instantbird ()
13:39:49 --> tymerkaev-afk has joined #instantbird
13:50:08 * tymerkaev-afk is now known as tymerkaev
14:03:27 <-- orlik has quit (Ping timeout)
14:03:30 --> orlik has joined #instantbird
14:22:01 <clokep> Anyone know if bug 207 is still an issue? I'd imagine it is from looking at the code, but just wanted to make sure before I look at it.
14:22:06 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=207 min, --, ---, nobody, NEW, URLs should not contain Smileys
14:22:26 <flo> it hasn't been fixed yet
14:24:26 <clokep> Thanks.
14:38:56 <clokep> Fixed. ;)
14:39:42 <flo> without regressions?
14:40:01 <clokep> No regressions I can find. Any particular test cases you want me to run?
14:40:19 <flo> pastebin the fix? :)
14:40:36 <clokep> Its not quite a "fix" yet. ;)
14:40:57 * flo looks at the huge list of things to do "once 0.2 is released"
14:41:00 <clokep> I pulled some code out to rewrite it and it works, but I'm trying to fit it into a patch now.
14:44:11 <flo> hmm, oh well, it's not so terrible
14:44:15 <flo> only about 250 items
14:46:34 <clokep> I hope they're small and you can check them off easily. :)
14:46:46 <flo> not really ;)
14:46:52 <flo> that's the reason why they aren't checked yet
14:47:08 <flo> I hope I can put lots of them in the "file a bug" category though
14:50:00 <clokep> Mmhmm. Filing bugs takes a while though haha.
14:51:39 <flo> yes
14:51:52 <flo> but it's better to have a bug filed so that anybody can decide to do something about it
14:52:04 <flo> rather than having it sit in my personal list for yeas untouched
14:52:36 <flo> all those that could benefit instantbird, that could be worked on by anybody, and that I'm unlikely to touch in the next 6 months should really be filed :)
14:53:04 <clokep> Sounds good to me.
15:05:52 <clokep> I have a possible patch but it hasn't been tested in Instantbird. :)
15:07:47 <clokep> Is it OK if I assign that to myself?
15:07:59 <flo> if you want to work on it, yes
15:26:11 <clokep> flo: What's the expected behavior of a string like "teXDst"?
15:26:37 <flo> popup a dialog asking the user's opinion? :)
15:26:48 <flo> (seriously, I don't know ;))
15:26:54 <clokep> OK, I wasn't sure.
15:27:11 <clokep> I won't worry about it then.
15:27:37 <flo> I guess some users will expect something and others will expect something else
15:27:51 <clokep> Right, the links is an issue though.
15:29:05 <ibbot> clokep@gmail.com added attachment 316 to bug 207.
15:29:09 <ibbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=207 min, --, ---, clokep, ASSI, URLs should not contain Smileys
15:31:53 <clokep> Posted a patch in the bug, but I'm not sure it'll work until I get home. :)
15:34:10 * flo goes home!
15:34:16 <flo> back later :)
15:34:17 <-- flo has quit (Quit: Instantbird 0.3a1pre)
15:37:06 <-- testib has quit (Ping timeout)
15:41:14 <-- hicham has left #instantbird ()
15:42:13 <clokep> Mic: Sounds like bug 300 would help you easily do what you want in your extension. ;)
15:42:15 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=300 enh, --, ---, nobody, NEW, Add an array of all the displayed messages
15:54:11 <-- Even has quit (Quit: Instantbird 0.3a1pre)
16:04:06 <skeledrew> clokep: could really do with a fix on bug 207. my Pidgin smileys tend to make it pretty glaring. i'm still working on making it protocol specific though...
16:04:09 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=207 min, --, ---, clokep, ASSI, URLs should not contain Smileys
16:06:19 <skeledrew> 300 would also make it easy to make a history from the logs
16:19:34 <-- skeledrew has quit (Ping timeout)
16:23:26 <clokep> skeledrew: Right, well. I'll see if my patch works tonight.
16:37:23 --> mic has joined #instantbird
16:37:27 <mic> hi
16:37:36 <clokep> Hello mic.
16:37:36 <mic> clokep: just one thing before I forget it
16:38:25 * mic is now known as IRCMonkey45469
16:39:12 <clokep> How's it going?
16:39:20 <IRCMonkey45469> How did you fix the "no smilies in links"?
16:39:39 <clokep> The patch is in bug 207, I can paste bin it too if that's easier.
16:39:41 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=207 min, --, ---, clokep, ASSI, URLs should not contain Smileys
16:39:43 <IRCMonkey45469> I'm fine, thanks
16:39:45 <IRCMonkey45469> How are you?
16:40:35 * IRCMonkey45469 is now known as Mic|webclient
16:40:41 <Mic|webclient> ah, ok
16:40:44 <clokep> Fine. Going through some bugs. :P
16:40:58 <Mic|webclient> I have a different 'fix' laying around
16:41:29 <clokep> Pseudo code style I always take "chunks" of the message a time (i.e. "words", things separated by spaces), then I check whether that chunk matches a URL scheme, if it does we skip it, otherwise we make all smiles in the chunk into smileys.
16:41:31 <Mic|webclient> if I recall correctly
16:41:41 <clokep> Ah, why not checked in? ;)
16:42:05 <Mic|webclient> if I recall it correctly it replaced the linkification code
16:42:30 <Mic|webclient> The smilies are replaced by a textModifier iirc
16:42:56 <clokep> OK.
16:43:07 <clokep> My patch hasn't really been checked.
16:43:10 <Mic|webclient> You can't apply several textModifiers to the same content and so only the order matters + you've got the advantage that it's easier to modifiy the link-ing  code
16:43:40 <clokep> I don't touch the linking code, so that should be easy to modify still. :P
16:43:52 <clokep> If its a better solution then post the patch and we can see.
16:44:18 <Mic|webclient> The difference is that there would be no need to check for smilies in urls anymore then
16:45:01 <Mic|webclient> well, maybe .. if it's a html message with already existing links maybe .. 
16:45:17 <Mic|webclient> Right now the link-ing code is a function from the framework
16:45:23 <Mic|webclient>  .. and it has some flaws
16:45:31 <clokep> Hmm...I see.
16:45:43 <Mic|webclient> I have a nice list of test cases where it fails on valid URLs
16:45:46 <clokep> So its being called /after/ the smiley code right now? SO you're saying swapping them would fix it.
16:48:29 <Mic|webclient> http://hg.instantbird.org/instantbird/file/dbebf9abbb07/instantbird/content/convbrowser.xml#l295
16:48:34 <Mic|webclient> if I'm not mistaken
16:49:25 <Mic|webclient> ah, I'm right btw concerning the Smilie replacement: http://hg.instantbird.org/instantbird/file/dbebf9abbb07/instantbird/content/convbrowser.xml#l275
16:50:24 <clokep> But http://hg.instantbird.org/instantbird/file/dbebf9abbb07/instantbird/content/convbrowser.xml#l301 calls it after the links have been done, so I'm not sure what we're discussing.
16:51:14 <Mic|webclient> My replacement for the linkification code is based on a textModifier, like the smilie code
16:52:09 <clokep> Right, but I don't see how that would solve the problem, the links should be found either way.
16:52:11 <Mic|webclient> but you can't apply two different textModifiers to the same text/content, so you can either have a smilie or a link somewhere. Depending on in which order I add them into the textModifiers array
16:52:30 <clokep> Ah.
16:52:41 <Mic|webclient> So find the links first and the parts where they are are "untouchable" now
16:53:00 <Mic|webclient> They are just skipped .. let me find the line
16:53:05 <clokep> But if the links are created currently why would the text modifiers code touch it?
16:53:19 <clokep> Mic: I understand what you're saying, I don't understand how its /different/ from what's being done now.
16:53:46 <clokep> The textModifiers only run on text nodes, but if they were already linked...the modifiers shouldn't touch them.
16:53:58 <Mic|webclient> exactly
16:54:27 <Mic|webclient> since the linkification code is bugged we need to replace it anyways
16:54:45 <Mic|webclient> and by using a textModifier we get rid of the smilie problem in the same moment
16:55:10 <clokep> OK, but why would the smiley code touch them now (assuming its a link that the code finds).
16:55:20 <clokep> That's good, put your patch up then. ;)
16:56:21 <Mic|webclient> That's where the cleaning up and text modifications are applied: http://hg.instantbird.org/instantbird/file/dbebf9abbb07/instantbird/modules/imContentSink.jsm#l269
16:56:38 <Mic|webclient> Here's the skipping of newly added nodes: http://hg.instantbird.org/instantbird/file/dbebf9abbb07/instantbird/modules/imContentSink.jsm#l351
16:57:15 <clokep> So it only skips *newly* added nodes, not all nodes?
16:57:24 <Mic|webclient> Currently the linkification is applied before and not during the cleanup so it is treated like any other part of the message.
16:57:30 <Mic|webclient> Exactly
16:57:45 <clokep> That's what I didn't realize. I thought it just skipped all nodes that weren't text.
16:58:02 <Mic|webclient> it is filtering them and leaves only the "good ones"
16:58:08 <clokep> Right well that sounds like a better solution, want me to assign that bug to you? ;)
16:58:11 <Mic|webclient> and applies the textModifiers to them
16:58:30 <Mic|webclient> My problem is: I still lack a good regexp for matching URLs :D
16:58:37 <clokep> I have one in my code.
16:58:49 <Mic|webclient> let's talk about this later, ok?
16:58:59 <Mic|webclient> I'm need to catch my bus now
16:59:01 <clokep> Yes.
16:59:01 <Mic|webclient> bye
16:59:03 <clokep> OK.
16:59:04 <clokep> Bey.
16:59:08 <-- Mic|webclient has quit (Quit: http://www.mibbit.com ajax IRC Client)
17:19:38 --> skeledrew has joined #instantbird
17:28:40 <-- skeledrew has quit (Ping timeout)
17:39:24 --> flo has joined #instantbird
17:39:24 * ChanServ sets mode +qo flo flo 
17:41:11 <clokep> Mic: I had a really simple one in my code, but it would probably match too much, there's a good one here though: http://immike.net/blog/2007/04/06/5-regular-expressions-every-web-programmer-should-know/
17:45:02 --> testib has joined #instantbird
17:48:25 * tymerkaev is now known as tymerkaev-afk
17:50:27 * flo has a different proposal for the smileys in URLs bug
17:50:34 <flo> links are created before replacing emoticons
17:51:23 <clokep> flo: That's what I thought too, but I'm not sure if my code fully works or not even. ;)
17:51:25 * tymerkaev-afk is now known as tymerkaev
17:51:26 <flo> when going through the DOM subtree of the message, we can detect links, and add a flag if the text is the address of the link. When this flag is present, the emoticon text modifiers should do nothing.
17:51:57 <clokep> Does it need a flag? Can we actually just check the tagName?
17:52:09 * clokep originally did it working with strings, not nodes.
17:52:15 <flo> I'm interested in addressing the <a href="http://yahoo.com">http://google.com/</a> issue too
17:52:48 <clokep> What should be done in that case?
17:52:59 <flo> not sure. Maybe remove the link
17:53:11 <flo> but I think an URL should never be a link to a different URL.
17:54:19 <clokep> What about <a href="http://google.com/somesearch>http://google.com/</a>?
17:54:35 <clokep> (I.e. some long path that maybe you don't want to show in text so you just give the TLD.)
17:57:01 <flo> maybe a better solution would be to replace <a href="A">B</a> by <a href="B">B</a> (<a href="A">A</a>)
17:57:18 <flo> so that no information is lost in the unlikely case that such a thing wasn't a spoofing attempt
17:58:45 <clokep> flo: That sounds like a reasonable idea.
17:58:51 <clokep> Don't want to get rickrolled, do ya? ;)
17:59:05 <flo> ;)
17:59:32 <clokep> This seem like a better (much simpler) idea for the smiles in URL issue? http://pastebin.instantbird.com/301
18:05:39 <clokep> The only issue I see with doing it that way is what if I want to have a smiley? I.e. I want <a href="http://google.com">I <3 Google :)</a>
18:06:07 <clokep> Which is what my patch does. it'll run on the textnode no matter what but attempt to void URLs.
18:18:59 <flo> don't like the smileString function anymore? ;)
18:19:39 <clokep> Not if you're going to sanitize based on tagName. :P
18:20:21 <flo> I thinik the smileString should only be used on plain text, not HTML (as input. Obviously the output will be HTML)
18:21:08 <flo> what about (smileNode.tagName != 'a' || !smileNode.textContent.match( regexp matching the start of an URL) ?
18:21:31 <clokep> But why would you do that if its already linked?
18:21:45 <flo> you "I <3 Google" case
18:21:50 <flo> *your
18:22:39 <clokep> Oh, I see. That could work.
18:23:09 <clokep> But really all you need then is !smileNode.textContent.match(/uriExpression/)
18:23:50 <clokep> Although...now I'll break that case and say I want to link to <a href="http://google.com">http://google.com I <3 Google! :)</a> ;)
18:24:42 <clokep> Oops, not quite what I mean: here we go: <a href="http://google.com/search=XD">http://google.com/search=XD  I <3 Google! :)</a>
18:25:07 <flo> I think that would work better if the smileNode function was actually called though
18:25:11 <clokep> You don't want XD to match but you want the :) to match.
18:25:40 <clokep> Right, but that's why I wrote it as parsing each individual "word" instead of just ignoring things that were in Link tags.
18:26:44 <flo> what about replacing my regexp test by just testing that node.textContent == node.getAttribute("href") ?
18:26:44 <clokep> If only we could control how links are made...:(
18:27:03 <flo> yeah, let's take over the remote client ;)
18:27:22 <clokep> Haha, that'll work fine, but it'll still cause the XD at the end of the URL to be made into a smiley.
18:34:08 <clokep> The way I'm seeing it right now you either need to totally ignore link ('a') nodes or you have to be able to properly parse the inside text of one so that smileys in the url are ignored but can exist if it is not a link url.
18:51:20 --> Mic has joined #instantbird
18:57:20 <Mic> Just rewriting the linkification code doesn't suffice .. 
18:57:51 <clokep> Yeah, it seems a bit more complicated. :(
18:58:06 <Mic> .. since there can be "ready-made" links in incoming messages already
18:58:33 <Mic> well, I mean link tags .. not the most likely nonsense wording I used
18:58:58 <Mic> flo: would it theoretically work to .. 
18:59:15 <Mic> .. put and anchor in one message and a link to this anchor in another?
19:01:01 <Mic> I was thinking about putting anchors to messages where some content has been extracted
19:02:28 <Mic> (that's a different idea now, but still anchor-related) so if I add a window with a list of the content that has been posted so far
19:02:47 <Mic> I could put an option to each to scroll to the place where it has been posted
19:04:30 <Mic> clokep: I hope that you don't mind if I merge your idea into my addon one day?
19:05:05 * tymerkaev is now known as tymerkaev-afk
19:05:11 <clokep> Mic: Not at all! When you first were talking about it I figured that would end up happening. I have enough addons to handle right now. :)
19:05:14 <clokep> Plus a couple in the pipeline.
19:06:24 <Mic> I'm curious what these might be :)
19:08:08 <-- tymerkaev-afk has quit (Ping timeout)
19:08:31 <Mic> two different adresses in the reference attribute and the text node are one thing
19:08:54 <Mic> but you can't possibly catch all the cases, e.g. like <a href="http://yahoo.com">I <3 Google :)</a>
19:12:29 <clokep> Mic: But you can catch that if you parse the text node separately. :P
19:13:14 <clokep> The ones I've onnly done very little ground work on are a little bit of research into translation and a sample protocol in JS based on Markov chains.
19:13:18 <clokep> Since its more fun then Lorem Ipsum.
19:13:19 <Mic> I don't see how
19:14:33 <Mic> and what does "a sample protocol [...] based on Markov chains" mean?
19:14:51 <Mic> Markov chains as I know that are "a random process with no history"
19:14:52 <clokep> You talk to it and it talks back with random text.
19:15:08 <clokep> Yeah basically.
19:15:43 <clokep> Right now its based off of "Ulysses", it returns a unique "sentence" every time that's almost readable. Its interesting
19:16:24 <clokep> We'll see if I get around to make it a protocol as opposed to just a JavaScript object.
19:21:14 <clokep> It had a little bit of a performance issue when I had it make it from "War and Peace" however. :(
19:22:44 <clokep> Mic: I'm not positive, but i think the patch I put in would actually catch the I <3 Google :) and leave the href alone (if there was a smiley in it).
19:23:35 <Mic> The important point in my example was that the href is pointing to Yahoo while the text is mentioning Google
19:24:54 <Mic> Maybe we should add (append?) the real URL in general?
19:26:01 <clokep> Oh, I didn't realize that. Flo mentioned something about that earlier, but I don't think that has anything to do with the bug in mention.
19:26:11 --> skeledrew has joined #instantbird
19:26:37 <Mic> That's true
19:26:45 <clokep> We could add a tooltip? Although that's still a bit hidden.
19:27:05 <clokep> Show the real URL in the statusbar on mouse over?
19:27:05 <Mic> It's shown in the status bar at the moment
19:27:11 <clokep> Oh, OK. :)
19:27:49 <clokep> If you want to be very annoying yuo can have all links where href != textContent have a popup that asks if they want to open the link or not.
19:37:49 <Mic> flo: which protocols would you like to support in URLs?
19:38:19 <Mic> I was thinking about the common ones like https?, ftp, ..
19:38:48 <clokep> (ht|f)ps? ;)
19:38:52 <clokep> Oops.
19:39:00 <Mic> While others might still be useful (e.g. someone sending a skype: link with a phone number
19:39:02 <clokep> (ht|f)tps? rather
19:39:21 <skeledrew> ftps???
19:39:24 <Mic> javascript/chrome/file should be forbidding imo
19:39:28 <skeledrew> secure ftp?
19:39:32 <clokep> I had been using \S+:\/\/\S+ as my regex. :)
19:39:36 <skeledrew> is that possible?
19:39:42 <clokep> skeledrew: You shouldn't be using ftp unless its sftp or ftps.
19:39:50 <skeledrew> hmm
19:40:07 <Mic> clokep: couldn't it be that http|https|ftp would be preferable performance wise? 
19:40:10 <skeledrew> use it over ssh :0
19:40:27 <skeledrew> :)
19:40:31 <clokep> Mic: Probably, but I just wanted to make it confusing.
19:40:36 <clokep> skeledrew: I think that's what sftp is.
19:40:48 <Mic> I have no idea what happens inside a regexp interpreter but I could image a simple either/or rule is easier to match
19:40:49 <skeledrew> ah
19:41:05 <clokep> Probably, its also earier to read.
19:41:21 <clokep> I'm guessing you didn't see this: Mic: I had a really simple one in my code, but it would probably match too much, there's a good one here though: http://immike.net/blog/2007/04/06/5-regular-expressions-every-web-programmer-should-know/
19:41:31 <Mic> Thanks I've seen this
19:41:37 <clokep> OK.
19:41:42 <Mic> that's why I came up with the protocol question
19:42:35 <Mic> He's only supporting https? and ftp could have additional things like "username:passwort@" included in the adress
19:42:37 <clokep> I'd say http, https, ftp, ftps, sftp, gopher, gophers (need to check that one), skype, <all the IM categories, i.e. aim, xmpp, ymsgr, I have a list somewhere>, mailto,....I'm out.
19:42:38 --> DGMurdockIII has joined #instantbird
19:42:55 <skeledrew> !
19:43:00 <Mic> :D
19:43:02 <clokep> Mic: https? makes the s optional.
19:43:03 <Mic> I expected that
19:43:21 <clokep> Mic: You expected what? :P
19:43:32 <clokep> (Or was that actually a question "https?" Haha. Silly Regex)
19:43:35 <Mic> skeledrew 'objecting' with the IM protocols
19:43:46 <Mic> Nope, I was using it to refer to both http and https
19:43:54 <clokep> We should support both.
19:44:16 <clokep> the username:password@ isn't only ftp syntax, its also http auth I believe? But not positive.
19:44:19 <skeledrew> take a while to support all of those
19:44:24 <clokep> Should also support a port number, i.e. http://google.com:80
19:44:40 <Mic> The regexp from the page does that if I saw that right
19:44:40 <clokep> Nah, you just make an array and do .join('|') :)
19:45:19 <clokep> Yes it does, it also makes an attempt at TLD (com, net, org, etc.) but we can probably just do \w+(\.\w+)?
19:45:47 <Mic> I agree
19:46:08 <clokep> (Also wouldn't need to update it for new TLDs)
19:46:14 <Mic> In my opinion we're not looking for valid urls but only for things that resemble one
19:46:28 <clokep> I agree.
19:47:45 <clokep> http://en.wikipedia.org/wiki/Template:URI_scheme
19:48:06 <clokep> git, svn, hg, etc. would be good to have. :)
19:49:31 <clokep> I mean I feel like we can make something that vaguely looks like a uri by just saying like \w+:\/\/(\w:)?(\w@)?\w+\.\w+(\.\w+)+\S+ or something like that.
19:50:02 <Mic> ah.
19:50:34 <Mic> maybe
19:50:39 <Mic> I'm off again
19:50:40 <Mic> bbl
19:50:49 <clokep> I'll try to make a list of things we should match for it.
20:03:05 <clokep> Looks like I lied and gophers is not one. :P
20:13:16 <clokep> Uhhh, so I came up with this Mic: http://regexpal.com/?flags=&regex=%28%5Cw%2B%3A%5C%2F%5C%2F%28%5Cw%2B%28%3A%5Cw%2B%29%3F%40%29%3F%7Cwww%5C.%29%5Cw%2B%5C.%5Cw%2B%28%5C.%5Cw%2B%29%3F%28%3A%5Cd%2B%29%3F%5CS*%7C%5B%5E%5CB%5C.%5Cd%5D%28%3F%3A%28%3F%3A25%5B0-5%5D%7C2%5B0-4%5D%5B0-9%5D%7C%5B01%5D%3F%5B0-9%5D%5B0-9%5D%3F%29%5C.%29%7B3%7D%28%3F%3A25%5B0-5%5D%7C2%5B0
20:14:20 <clokep> And my samples I used for matching: http://pastebin.instantbird.com/302
20:14:32 <clokep> (The top group should be matched, the bottom group should /not/ be matched.)
20:16:01 <clokep> Although it seems from http://en.wikipedia.org/wiki/URI_scheme#Examples that not all of them use the :// in the middle which is kind of the crux of the whole regex, some only use ":"
20:17:44 <skeledrew> should be easy getting this to work with IB (i hope): https://addons.mozilla.org/en-US/firefox/addon/162021/
20:18:36 <clokep> Should be easier in Instantbird then Firefox since it has the hooks for new messages.
20:19:04 <skeledrew> k
20:19:51 <flo> "plus three layers of remote caching on the server-side." uh?
20:20:57 <skeledrew> lol
20:21:05 <clokep> Excessive. :)
20:21:21 <skeledrew> precaution?
20:22:14 <skeledrew> better yet, there's a pure JS version that can be modded
20:22:54 <skeledrew> just put it in a filter for the messages
20:23:08 <clokep> Exactly.
20:24:38 <flo> clokep: I made an Eliza protocol plugin last week
20:24:52 <clokep> I have no idea what Eliza is...
20:25:02 <flo> http://en.wikipedia.org/wiki/ELIZA
20:25:10 <skeledrew> some AI thinie
20:25:15 <skeledrew> *thingie
20:25:22 <clokep> Oh, we were talking about this briefly last week?
20:25:24 <flo> skeledrew: not intelligent though ;)
20:25:25 <clokep> Did you port it to JS?
20:25:35 <skeledrew> nope :(
20:25:37 <flo> I took a (crappy) JS implementation on a website
20:25:44 <flo> and just put it in a JS proto
20:25:57 <clokep> Ah neat.
20:26:31 <flo> I'm not sure if it's worth cleaning up the code
20:27:20 <clokep> Probably not if there are other examples around.
20:28:25 <flo> I rewrote all the parts that were using setTimeout
20:28:49 <flo> and wrapped everything else into a single containing function
20:29:10 <clokep> Sounds like a lot of work. :P
20:29:18 <flo> but as there are lots of warnings about undeclared variables, it may not be enough to have it working right with several accounts at once
20:29:22 <skeledrew> flo: why isn't replytonick autoupdating?
20:29:38 <skeledrew> MinTrayR did
20:29:42 <flo> maybe because I haven't incremented the version number
20:29:49 <flo> is it marked as incompatible?
20:29:50 <skeledrew> k
20:29:55 <skeledrew> so it's manual then...
20:29:59 <skeledrew> no
20:30:11 <skeledrew> just wanted to avoid manual labour
20:31:59 <clokep> According to AIO its compatibly with 0.3a1pre
20:32:04 <clokep> https://addons.instantbird.org/en-US/instantbird/addon/249
20:32:18 <skeledrew> k
20:32:19 <skeledrew> thx
20:32:20 <clokep> Oh, but its experimental, doh.
20:32:23 <clokep> No screenshot still. :)
20:32:50 <skeledrew> lol
20:33:37 <flo> stupid screenshot thing... ;)
20:37:43 <clokep> flo: I compiled js-proto branch this morning btw. It worked. :P
20:38:46 <flo> cool
20:39:45 <clokep> The abstracted JS protocol is pretty neatly done, very simple.
20:46:48 <flo> it's not finished
20:47:03 <flo> it's not possible yet to have JS protocols that have a buddy list ;)
20:47:25 <clokep> Hahah, I see.
20:47:30 <clokep> Good thing Omegle is totally random? :P
20:52:56 <flo> that's the reason why I took it for the first demo
20:53:42 <clokep> To see what needs to be added/if it works type of thign?
20:54:32 <flo> just because it was rewarding to be able to use it and have fun with it
20:54:48 <clokep> Fun is important! :-D
20:55:14 <flo> I'm not sure if the author of the website would approve it
20:55:24 <clokep> Anyway I've got to go, I don't think I've ever said thanks for making such a great program, and thanks for putting up with me learning everything haha.
20:55:36 <clokep> flo: I see. Silly software licenses?
20:55:58 <flo> no, the code I took for Eliza is free
20:56:22 <flo> I was rather thinking that the author of Omegle earns money by having google ads displayed after each conversation on the webpage
20:56:30 <flo> and the instantbird plugin won't display them
20:56:42 <clokep> Ohhhh, yeah I was wondering how you got around the CAPTCHA actually?
20:57:19 <flo> when is it required?
20:57:32 <clokep> I don't know, Wikipedia said itw as? ;) I've never used Omegle.
20:57:49 <flo> it may be for using the webcam or something like that
20:58:04 <flo> or some feature I don't use
20:58:14 <clokep> It scares me that people use these websites. :(
20:58:17 <flo> I've reverse engineered the protocol using Firebug
20:58:40 <clokep> Ah, awesome.
20:58:40 <flo> the protocol used is wonderfully simple. Perfect for a code example :)
20:59:55 <clokep> Its great when people don't make these ridiculous convoluted things, isn't it?
21:00:05 <flo> ;)
21:00:13 <clokep> Anyway, I'm off. Goodnight.
21:00:18 <-- clokep has quit (Quit: http://www.mibbit.com ajax IRC Client)
21:00:18 <flo> good evening :)
21:35:52 * Mic wonders how does this regexppal.com page works.
21:36:46 <Mic> There's a field for the regexp and one for the data to test but no button to start
21:38:20 <flo> Mic: when something is matched it's highlighted
21:39:51 <Mic> There's only highlighting on the regexp pattern
21:40:14 <Mic> Should it match it automatically when entering something in the data textarea?
21:41:00 <Mic> It is a URL pattern, isn't it? ;)
21:42:17 <Mic> These what I'd like to test: http://pastebin.instantbird.com/303
21:42:25 <Mic> Some of them should match, hopefully
21:46:49 <flo> what's the regexp you are using?
21:47:23 <Mic> I'm using the one that clokep posted
21:48:04 <Mic> Search for "22:13:18 - clokep: Uhhh, so I came up with this Mic"
21:48:22 <Mic> well, the timestampt might be different ofcourse
21:49:36 <flo> his regexp is: (\w+:\/\/(\w+(:\w+)?@)?|www\.)\w+\.\w+(\.\w+)?(:\d+)?\S*|[^\B\.\d](?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)
21:50:06 <Mic> d'oh
21:50:11 <flo> the link was cut by the IRC server so the end of his regexp is missing
21:50:15 <Mic> It was cut
21:50:18 <flo> the complete regexp was in the pastebin
21:50:22 <Mic> Thanks :)
21:50:25 <flo> np
21:50:50 <flo> obviously, an url shortener could have helped him in this case ;)
21:51:36 <Mic> Looks good on my test urls, just one missed
21:52:55 <Mic> I wonder why it is matching with the protocol but not without it
21:54:33 <flo> the one that doesn't match for me has a non-ASCII characted in it
21:54:37 <Mic> clokep, clokep_work: try the adresses from this pastebin item: (\w+:\/\/(\w+(:\w+)?@)?|www\.)\w+\.\w+(\.\w+)?(:\d+)?\S*|[^\B\.\d](?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)
21:54:56 <Mic> argh
21:55:11 <Mic> http://pastebin.instantbird.com/303
21:56:13 <Mic> It's an adress that is completely valid to enter into your browser
21:56:43 <Mic> That it is converted into a punycode adress internally is something different in my opinion
21:57:13 <flo> there are few bugs around that
21:58:22 <Mic> Still I wonder why it is matched with the protocol specified
21:58:33 <Mic> (it's the next in list)
22:03:57 <-- DGMurdockIII has quit (Quit: get satisfied! • :: www.unitedservers.de ««« (Gamers.IRC) »»» gamersirc.net ::)
22:05:35 <Mic> AH!
22:05:46 <Mic> It's misinterpreted as part of the path maybe
22:13:50 <Mic> Good night
22:14:03 <flo> good night
22:32:14 <-- skeledrew has quit (Ping timeout)
22:32:17 --> skeledrew has joined #instantbird
22:43:49 <-- skeledrew has quit (Ping timeout)
22:45:19 <-- Mic has quit (Quit: Instantbird 0.3a1pre)