--- Log opened Wed Apr 03 00:00:06 2013 --- Day changed Wed Apr 03 2013 20130403 00:00:06< anonymissimus> git reset also works I think 20130403 00:00:29< anonymissimus> or --abort if it happened during a rebase, a merge, a cherry-pick,... 20130403 00:02:25-!- Blueblaze [~Blueblaze@adsl-99-148-244-197.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev 20130403 00:03:17-!- LordNasty [~NaSTy@93-43-168-243.ip92.fastwebnet.it] has quit [Ping timeout: 248 seconds] 20130403 00:04:04-!- LordNasty [~NaSTy@93-43-168-243.ip92.fastwebnet.it] has joined #wesnoth-dev 20130403 00:13:23-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20130403 00:15:36-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has joined #wesnoth-dev 20130403 00:16:59< lipkab> mordante: Yes, it's only present on trunk/master/whatever. 20130403 00:20:26< AI0867> 23:50 < fabi> Espreon: Add is for adding file under version control. How can it be used to solve conflicts? <-- add is for adding it to the index. This can be used for adding it the first time, for committing selectively (not commit -a) and for marking a conflict as resolved 20130403 00:20:48< AI0867> the third makes more sense if you realize that a merge tries to generate a commit 20130403 00:21:11< AI0867> if it fails due to conflicts, the stuff that was successfully merged is in the index. The stuff that has conflicts is not. 20130403 00:21:24< AI0867> once you've resolved the conflicts, you add the files to the index, and commit 20130403 00:21:36< fabi> hmmm, I need to read more about git. I am lacking the vocabulary to fully understand what you try to tell me. 20130403 00:21:58< AI0867> svn has two 'trees': the working copy and the 'BASE' 20130403 00:22:06< AI0867> svn diff shows you the difference between the two 20130403 00:22:09< AI0867> git has 3 20130403 00:22:31< AI0867> the index sits between the two you have in SVN 20130403 00:24:58< fabi> okay, working copy --commit--> index --push--> repository 20130403 00:25:13< AI0867> no 20130403 00:25:47< AI0867> working tree --add--> index --commit--> repository --push--> external repositories 20130403 00:26:03< fabi> i see 20130403 00:27:54 * anonymissimus doesn't see :P 20130403 00:28:28< Espreon> anonymissimus: How wonderful! 20130403 00:28:30< shadowm> 10:53:48 shadowm: Yes, I am aware of the read only connection. 20130403 00:28:44< shadowm> Then it wasn't a good idea to suggest it in the first place. 20130403 00:28:54< irker342> anonymissimus wesnoth:master * 1.11.2-92-g834680d / projectfiles/CodeBlocks/wesnoth.cbp: cb project update 20130403 00:28:55< irker342> anonymissimus wesnoth:master * 1.11.2-93-gb369ade / projectfiles/VC9/wesnoth.vcproj: vc project update 20130403 00:29:34< shadowm> fabi: I don't see any use in making the zoom buttons turbo buttons. 20130403 00:29:43< shadowm> But okay. 20130403 00:30:20-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection] 20130403 00:30:52< fabi> shadowm: It is really a relief with the palette scrolling. (It was a click orgy). The scrolling does not profit that much but is in line with the keyboard behavior now. 20130403 00:31:09< fabi> s/scrolling/zooming 20130403 00:31:43< shadowm> Yes, all scrollbar buttons are supposed to be turbo buttons in Wesnoth and the editor palette was inconsistent in that regard. 20130403 00:33:55-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!] 20130403 00:34:56< fabi> shadowm: local user registration is to be disabled at the forum, the wiki and the redstone tool. A nice little web frontend provides registration/password change and maybe some other or administrator features. 20130403 00:35:27< shadowm> Oh no. 20130403 00:35:29< shadowm> Please no. 20130403 00:35:38< shadowm> I am not willing to reinvent wheels like that. 20130403 00:35:58< shadowm> The forums' registration form was made secure against automated form-fillers through cycholka's and my own efforts. 20130403 00:36:25< shadowm> Also, the name is redmine. 20130403 00:36:46< fabi> Ah yes, redstone is an early american carrier rocket. 20130403 00:39:18 * shadowm kicks timotei for sending an HTML message to the ML. 20130403 00:39:21< anonymissimus> mordante: btw, I can confirm that the textbox cursor position is fixed. I think I never was bugged enough by this bug to report it. :) 20130403 00:39:42< shadowm> timtoei: Whatever client you are using didn't even provide a plain text fallback. 20130403 00:39:45< shadowm> timotei 20130403 00:39:50< shadowm> timotei_ 20130403 00:40:14< fabi> shadowm: That is what I already told on #wesnoth-de. It's a major improvement to share all credentials but it won't be easy to get the change through because of project politics. 20130403 00:42:05< shadowm> fabi: The fact that you people are discriminating against me by discussing stuff in a language that is not English irks me. A lot. 20130403 00:42:26< shadowm> *passively discriminating 20130403 00:43:03< shadowm> How about you figure out how to get Redmine to use phpBB's users table alone instead? 20130403 00:43:36< shadowm> I know that doing it the other way around (either the forums or the wiki) would take a potentially painful and/or destructive conversion job (especially in MW's case). 20130403 00:44:14< fabi> MW? 20130403 00:44:17< AI0867> mediawiki 20130403 00:44:47< shadowm> Yes, if you are going to discuss site architecture changes you'd better know the names and acronyms of the software we use atm. :( 20130403 00:45:06< Gambit> Well that's a bit unfair. 20130403 00:45:40< shadowm> If someone can tell me about Redmine and whether it has some kind of auth plugin system like phpBB does I may even do it myself. 20130403 00:46:10< shadowm> But apparently the discussion is taking place in a channel using a language I don't understand? 20130403 00:46:13< fabi> shadowm: redmine, phpbb and mediwiki all have ldap auth plugins available. 20130403 00:46:39< shadowm> Gambit: Please provide your opinion now. 20130403 00:46:44< fabi> shadowm: That is why I discuss it here with you in english. You have not missed anything. 20130403 00:46:50< shadowm> The opinion you just provided me in my channel. 20130403 00:48:31< Gambit> Rolling your own security system is a cardinal sin. Creating some new global Wesnoth login system is dangerous. 20130403 00:48:38< Gambit> Integrating with other systems is awesome though. 20130403 00:49:52< Gambit> (Existing systems that have presumably been thoroughly thought out and tested by people who know more about that than any of us because it's their speciality.) 20130403 00:50:03< shadowm> The thing with MW is that to my knowledge it's designed to make operations as simple as _renaming a user account_ impossible. 20130403 00:50:32< fabi> shadowm: Well, you are uptodate now. We discussed that sharing credentials would be nice and that it is doable with ldap (the defacto standard nowadays on such stuff). And I ended the discussion with the statement that it won't happen because someone won't like changes. 20130403 00:50:33< shadowm> So I assume with that little bit of information I know that MW would have to be destructively converted at the expense of edit history. 20130403 00:51:11< shadowm> fabi: I'd prefer if everyone registered using the forums' facilities. 20130403 00:51:27< shadowm> That is subject to all our intricate anti-spammer verification systems. 20130403 00:51:43< shadowm> And yes, they are intricate, this is not hyperbole. 20130403 00:51:43< fabi> shadowm: That is no surprise to me. I would have asked if you are ill if you told me that you liked the idea. 20130403 00:51:55< shadowm> What? 20130403 00:52:11< fabi> I was sure that you deny any change to the current system. 20130403 00:52:22< shadowm> My English parser broke around 'ill'. 20130403 00:52:46< Gambit> This is looking much closer to a fight than a discussion right now. 20130403 00:52:46< fabi> ill == sick 20130403 00:52:59< shadowm> 'sick' doesn't make any semantic sense in context to me either. 20130403 00:53:25< shadowm> I don't know what you are trying to say. My physical health has been very poor since 2005, but I'm not terminal or anything. 20130403 00:53:42< fabi> Maybe the statement was too German specific. We ask someone if he got a disease if he is doing something unexpected. 20130403 00:53:54< shadowm> Oh. 20130403 00:54:34< shadowm> So you asked me knowing that I would refuse? 20130403 00:54:53< shadowm> But seriously, 19:45:38 If someone can tell me about Redmine and whether it has some kind of auth plugin system like phpBB does I may even do it myself. 20130403 00:55:20< fabi> Well, all our tools do have ldap plugins. 20130403 00:55:32< shadowm> That is, I'd be okay with redmine<-phpBB instead of redmine<-LDAP->phpBB. 20130403 00:57:12< fabi> Well, that is reinventing the wheel, I doubt that there is a plugin for getting auth through the phpbb database (sql?) backend. 20130403 00:57:42< shadowm> Hopefully their auth plugin architecture is generic enough. 20130403 00:58:23< shadowm> That'd mean I'd only need to read phpBB's own native auth system and translate it to redmine's arch. 20130403 00:58:30< fabi> This would mean we have to write a plugin for redmine and one for wikimedia. 20130403 00:58:37< shadowm> MediaWiki. 20130403 00:58:48< Gambit> fabi: So basically, the forum registration process is already secure and hardened against spammers, but switching to LDAP would require us to recreate all of that and we would almost certainly do it wrong. 20130403 00:58:50< shadowm> MediaWiki is the software, Wikimedia is the foundation. 20130403 00:59:04< anonymissimus> is there some git wiki page we have ? other than this one http://wiki.wesnoth.org/GIT-SVN 20130403 00:59:17< shadowm> But I won't invest any time on MW for the aforementioned reasons (I did do the research back in the day when I wanted to rename ShikadiLord to Shadowmaster). 20130403 00:59:23< anonymissimus> that seems to no longer fit (but parts of it still do) 20130403 00:59:40< Gambit> (Switching to LDAP would also require wesnothd changes) 20130403 00:59:50-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: And once again, in an explosion of smoke, the ninja disappeared into the night] 20130403 00:59:57< shadowm> Gambit: Ah yeah, I completely forgot about wesnothd. 20130403 01:00:00< Gambit> Don't know if either of you mentioned that yet. 20130403 01:00:05< fabi> Gambit: There are ready to use ldap frontends out there, I suspect one will fit our needs. 20130403 01:00:10< shadowm> It's wired directly into phpBB's database bypassing the auth arch system. 20130403 01:00:24< Gambit> The front end is a very small portion of the problem. 20130403 01:00:43< fabi> Please tell me more. 20130403 01:02:02-!- exciton [chuck-the-@89.208.169.104] has quit [Read error: Connection reset by peer] 20130403 01:02:16-!- exciton [chuck-the-@89.208.169.104] has joined #wesnoth-dev 20130403 01:02:20-!- Blueblaze [~Blueblaze@adsl-99-148-244-197.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze] 20130403 01:06:10< Octalot> Git history rewrites: I was wrong yesterday when I said that Git was intelligent about downloading after rewriting history. Changing a commit message from 2007 forces all existing clones to do a 1.3 gig download, even if nothing else changes. 20130403 01:07:08< fabi> Gambit: Please tell me more. 20130403 01:07:32< Gambit> fabi: About what? 20130403 01:07:49< fabi> "The front end is a very small portion of the problem" 20130403 01:08:07< Gambit> As compared to the backend... 20130403 01:08:16< fabi> The backend would be? 20130403 01:08:23< Gambit> The prettyness of the login form isn't really a big deal. 20130403 01:08:34< fabi> indeed 20130403 01:08:41< Gambit> fabi: The LDAP server itself and all the code that gets phpBB and wesnothd to communicate with it 20130403 01:09:06< fabi> But we don't talk about login forms. That won't change. Only registering/password change is an ldap issue. 20130403 01:09:28< Gambit> I dunno. You brought up frontends. 20130403 01:09:29< fabi> Gambit: I can setup ldap, done it more than once at work. 20130403 01:09:50< fabi> phpbb will be ready after installing the plugin. 20130403 01:09:56-!- _8680_ [~8680@2002:44e1:f952:0:e56b:9a5e:4955:e35d] has quit [Ping timeout: 240 seconds] 20130403 01:10:09< fabi> wesnothd needs to call the ldap credentials, that is a one liner. 20130403 01:10:14-!- _8680_ [~8680@2002:44e1:f952:0:76de:2bff:fed4:2766] has joined #wesnoth-dev 20130403 01:10:28< Gambit> fabi: Does ldap replace the entire users table, or just usernames, hashed passwords, and salts? 20130403 01:10:52< Gambit> And does it manage the hashing as well? 20130403 01:11:15< fabi> Yes, ldap supports all sorts of passwords. 20130403 01:12:04< shadowm> 20:10:07 wesnothd needs to call the ldap credentials, that is a one liner. 20130403 01:12:08< shadowm> False. 20130403 01:12:20< shadowm> Have you even read the FUH class implementation? 20130403 01:12:54< fabi> I never heard about FUH. 20130403 01:13:00< shadowm> forum_user_handler 20130403 01:13:21< fabi> Well, it would be ldapUH and I am sure that I can code it. 20130403 01:14:00< shadowm> It's always nice to see that kind of thing happen before making a drastic decision like switching the live forums to LDAP and losing the ability to register users through phpBB in the process. 20130403 01:14:47< shadowm> (And by 'switching' I mean "importing the users table into LDAP and switching the phpBB auth plugin to LDAP".) 20130403 01:15:17< fabi> Well, nobody wants to work on the running system. I would do a copy of the forum for setup and testing. If the result is fine we can think about doing the stuff to the real forum. 20130403 01:15:20< Gambit> According to RFC 4519, LDAP servers are to store passwords in plaintext for interoperability. 20130403 01:15:30< Gambit> Which means phpbb would still be responsible for hashing. 20130403 01:15:37< Gambit> And redmine would then need to be modified as well. 20130403 01:17:27< fabi> Those plugins claim to work out of the box without any coding involved. 20130403 01:17:56< Gambit> Given that wesnoth.org has been pwned on at least one occasion in recent memory, plaintext password storage is absolutely not an option. 20130403 01:18:18< Gambit> Not sure if I'm allowed to get into the specifics of that incident here though. 20130403 01:19:47< Gambit> fabi: Right, so you install the phpBB LDAP plugin, and it stores passwords in LDAP in the hashed and salted format that phpBB expects. 20130403 01:19:57< Gambit> fabi: However, Redmine is not expecting that particular format. 20130403 01:20:00< Gambit> And still has to be modified. 20130403 01:20:21< Gambit> You can't get out of this without rewriting large portions of one of the platforms' password management schemes. 20130403 01:20:33< Gambit> One of them will have to be patched by us. 20130403 01:21:10< Gambit> The question is: is it Redmine or phpBB? 20130403 01:21:29< Gambit> And when presented with that question, the answer is "Ditch LDAP all together, and teach Redmine to talk to MySQL instead." 20130403 01:21:47< Gambit> Why introduce a third chef into the kitchen as it were? 20130403 01:22:39< fabi> Because there are ready to use plugins for all tools. We do not have to patch any of them. ldap can serve them all at once without code changes. 20130403 01:22:51< AI0867> with plain text passwords? 20130403 01:22:59< fabi> no 20130403 01:23:01< Gambit> Yeah it can if we give up all security. 20130403 01:23:14< Gambit> And that is not an option. 20130403 01:23:32< fabi> AI0867: Gambit claimed that it will only work with plain text passwords. I don't share that opinion. 20130403 01:24:45< Gambit> I just explained the entire process. It's not an opinion. phpBB expects passwords to be salted and hashed in a particular manner. The LDAP plugin you refer to doesn't change that. 20130403 01:24:52< Gambit> It just stores the salts and hashes in LDAP instead of MySQL 20130403 01:25:10< Gambit> Actually just the usernames and hashes. I bet the salts are still in MySQL 20130403 01:25:33< Gambit> Or it stores the passwords in plaintext. A massive nono. 20130403 01:26:41< Gambit> There's only two possible answers. It either encrypts the passwords or it doesn't. 20130403 01:27:18< Gambit> If it doesn't, we can't use it. 20130403 01:27:32< Gambit> If it does, you still have to teach Redmine to use the encrypted passwords. 20130403 01:28:16< Gambit> And so if you have to screw with Redmine regardless, it's optimal to just cut LDAP out of the equation. 20130403 01:29:55< Gambit> I dunno how familiar you are with password storage principles, but "evil" is a legitimately suitable adjective for anyone who opts for plaintext storage of their users' passwords. 20130403 01:29:55-!- anonymissimus [~chatzilla@HSI-KBW-078-042-163-136.hsi3.kabel-badenwuerttemberg.de] has quit [Quit: ChatZilla 0.9.90 [Firefox 11.0/20120312181643]] 20130403 01:30:14< Gambit> And please do not think that wesnoth.org will not be broken into some time again in the future. 20130403 01:30:17< Gambit> Sony thought that. 20130403 01:30:19< Gambit> Yahoo thought that. 20130403 01:30:21< Gambit> Myspace thought that. 20130403 01:30:26< Gambit> LinkedIn thought that. 20130403 01:30:38< Gambit> And they were all wrong. 20130403 01:30:40< fabi> Gambit: Please stop to claim that I ever proposed to store passwords in plain text. You came up with this thing. 20130403 01:31:12< Gambit> fabi: Right. Okay. BUT, who is handling the encryption? 20130403 01:31:17< Gambit> phpBB or redmine? 20130403 01:31:38< Gambit> or do plan to modify LDAP? 20130403 01:31:50< AI0867> Gambit: hashing, not encryption 20130403 01:32:09< Gambit> That, yes. 20130403 01:32:43< Gambit> fabi: Because whoever's scheme you go with, you still have to teach all the others to use. 20130403 01:33:54< Gambit> I believe phpbb's protection against cross-site request forgery is also reliant on the user's password salt. 20130403 01:34:11< Gambit> It uses the salt to generate the anti-CSRF tokens. 20130403 01:34:50< fabi> Gambit: We use mysql as database backend? 20130403 01:35:06< Gambit> Yes. 20130403 01:35:27-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has quit [Quit: Konversation terminated!] 20130403 01:36:29-!- Octalot [~noct@host86-148-68-117.range86-148.btcentralplus.com] has quit [] 20130403 01:37:49< fabi> Gambit: I am right now setting up a phpbb with ldap and later redmine. That will solve all questions. 20130403 01:43:06< Gambit> fabi: Okay. Can I get a link to the LDAP phpbb plugin you are considering? 20130403 01:43:23< fabi> Gambit: Is there more than one? 20130403 01:43:40< fabi> Gambit: http://sourceforge.net/projects/ldapauthmod/?source=dlp 20130403 01:44:08< fabi> Gambit: Okay, phpbb is already working. 20130403 01:44:17< shadowm> phpBB has a built-in LDAP auth plugin. 20130403 01:45:39< shadowm> The thing you linked isn't for phpBB 3.0.x. 20130403 01:45:58< fabi> shadowm: Right, "/usr/share/phpbb3/www/includes/auth/auth_ldap.php" it is already installed. 20130403 01:46:12< shadowm> That's the built-in LDAP auth plugin. 20130403 01:46:37< Gambit> okay let me find the source for it then 20130403 01:46:58< Gambit> fabi's link was the one I just looked at and they do indeed only perform stripslashes() on the password before comparing it to the stored value 20130403 01:47:12< Gambit> hopefully the official one is better :P 20130403 01:48:18< shadowm> Gambit: https://github.com/shikadilord/weldyn/ 20130403 01:48:28< shadowm> Add includes/auth/auth_ldap.php to that URL. 20130403 01:49:32< Gambit> the official plugin does phpBB's usual hashing on the password 20130403 01:49:40< Gambit> https://github.com/phpbb/phpbb3/blob/develop/phpBB/includes/auth/auth_ldap.php#L232 20130403 01:50:12< shadowm> Be careful about using phpbb/phhpbb3. 20130403 01:50:14< shadowm> You might be on the 3.1 dev branch. 20130403 01:50:46< shadowm> IIRC develop-olympus is the 3.0 dev branch. 20130403 01:50:50< Gambit> er, wait a second 20130403 01:51:13< shadowm> Really, why don't you use my repository instead of peering into the future? 20130403 01:51:27< shadowm> https://github.com/shikadilord/weldyn/blob/phpBB-3.0.11/includes/auth/auth_ldap.php 20130403 01:52:12-!- mattsc [~mattsc@BeaverNet-166.caltech.edu] has quit [Quit: bye] 20130403 01:52:27< Gambit> Okay so... the phpbb LDAP plugin gives the password to LDAP in plaintext. 20130403 01:52:56< Gambit> But after the user logs in, the plugin hashes the password for the sake of the rest of phpBB which is expecting it to be hashed. 20130403 01:53:12< Gambit> The passwords are still being stored and passed around in plaintext. 20130403 01:53:14< Gambit> Eww. 20130403 01:56:07< Gambit> No. 20130403 01:56:19< Gambit> I shouldn't be narrating my deciphering of this file in real time :P 20130403 01:56:28< fabi> :-) 20130403 01:56:28< Gambit> I'll be back when I've audited the whole thing. 20130403 01:56:57< fabi> Gambit: Thank you for doing this. You have surely more knowledge about this kind of stuff. 20130403 01:58:16< fabi> Gambit: It is just that I never had problems to put hashed passwords in ldap for any kind of client I used. Thus I just assumed that nowadays there is no one going to still store plain text passwords in ldap. 20130403 01:59:03< AI0867> sure, but everyone needs to agree in which way they do the hashing 20130403 01:59:08< Gambit> ^ 20130403 01:59:27< AI0867> phpbb salts its hashes, and stores the salt separately from the hashed password 20130403 02:00:19< AI0867> if you pass just the hashed password to LDAP, nobody else can use it without getting the salt out of the database first 20130403 02:02:37< fabi> AI0867: you just store the same password hashed differently and use the one that fits the client. 20130403 02:02:57< AI0867> um 20130403 02:03:11< AI0867> do you mean LDAP gets to store 3 different hashes for each user? 20130403 02:03:32< AI0867> and has to know the specifics of the salting and hashing for each connected application? 20130403 02:03:32< fabi> depends on which hash types are supported by each client. 20130403 02:04:23-!- ancestral [~ancestral@mobile-198-228-235-003.mycingular.net] has joined #wesnoth-dev 20130403 02:04:35-!- ancestral [~ancestral@mobile-198-228-235-003.mycingular.net] has quit [Client Quit] 20130403 02:05:42< fabi> AI0867: Yes, is that a problem? 20130403 02:11:58-!- ancestral [~ancestral@75-168-48-55.mpls.qwest.net] has joined #wesnoth-dev 20130403 02:13:08< AI0867> well, as has been pointed out, if we have to do interoperability code for everything anyway, we might as well make it all interoperate with phpBB 20130403 02:13:20< shadowm> fabi: The radio buttons you committed are LordBob's? 20130403 02:13:29< shadowm> You haven't replied to the thread yet. 20130403 02:13:31< AI0867> I'm not familiar with any of the them though 20130403 02:13:39< fabi> shadowm: Yes, I will do so soon. 20130403 02:15:37< shadowm> fabi: Okay, one problem with the palette turbobuttons is that the palette is not refreshed while they are active. 20130403 02:15:59< fabi> shadowm: Oh, then I forgot to push my commits :-) 20130403 02:16:16< shadowm> If I hold the 'down' button down, it only takes about a second for the palette to refresh, already scrolled all the way to the last row. 20130403 02:16:23< shadowm> Okay. 20130403 02:16:55< fabi> Ah yes, I couldn't resolve that merge conflict. Still a git novice. 20130403 02:16:59< shadowm> In fact, it looks like the button is way too sensitive compared to real GUI1 scrollbars. 20130403 02:17:29< fabi> What does that mean? 20130403 02:17:55< shadowm> Never mind, real GUI1 scrollbars also have hyper-sensitive turbobuttons. 20130403 02:18:25< shadowm> They are basically useless compared to dragging the scroll marker around or using a mouse with a scrollwheel. 20130403 02:18:48< shadowm> GUI2 scrollbars are on the opposite side of the spectrum. 20130403 02:18:50< fabi> Please tell me. I can do some magic to them to make them more useful. 20130403 02:18:56< fabi> They scroll to fast, right? 20130403 02:19:03< shadowm> Yes. 20130403 02:19:30< shadowm> This is evident both with the editor palette, and the Add-ons Manager dialog. 20130403 02:20:20< shadowm> Incidentally, since Oxygen lets me do this, I have disabled scrollbar buttons here on Qt 4 and Gtk+ applications to save space. 20130403 02:20:40< shadowm> But yeah, not all of us have a mouse with scrollwheel apparently? 20130403 02:20:57-!- {V} [~V@139-79-ftth.on.nl] has quit [Read error: Connection reset by peer] 20130403 02:21:13< shadowm> I remember being told that a certain GUI2 person is one of those people stuck with a non-functional mouse from the past century. 20130403 02:21:33-!- {V} [~V@139-79-ftth.on.nl] has joined #wesnoth-dev 20130403 02:22:23< Gambit> Okay so I have discovered something amusing. 20130403 02:22:24< shadowm> Looking at the behavior on Windows, it seems like GUI1 scrollbar buttons should scroll about 50% slower, and GUI2 scrollbar buttons about 50% faster. 20130403 02:22:36< Gambit> phpbb salts are derived from the password itself 20130403 02:22:43< Gambit> they are then appended to the password 20130403 02:22:48< Gambit> The resulting string is hashed 20130403 02:22:54< Gambit> The salt is not stored anywhere 20130403 02:23:25< Gambit> The CSRF stuff I was remembering is a separate value entirely known as user_form_salt 20130403 02:25:52-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev 20130403 02:25:58< Gambit> Also if the usual salting and hashing fails, phpbb_hash() just returns md5($password) hahahaha 20130403 02:27:52< AI0867> derived from the password? 20130403 02:27:58< AI0867> then it's just a more complicated hash 20130403 02:28:08< AI0867> which would allow you to still build rainbow tables 20130403 02:28:08< Gambit> ikr? 20130403 02:29:20< Gambit> Yeah I'm actually pretty disappointed now that I've gone through this. 20130403 02:31:01-!- shadowm_desktop2 [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev 20130403 02:31:33-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 248 seconds] 20130403 02:33:04< Gambit> AI0867: Their hope is that the salt pads the password length enough that rainbow tables are not economical I guess. 20130403 02:33:13< Gambit> But that's a terrible thing to rely on. 20130403 02:33:22< AI0867> um 20130403 02:33:30< AI0867> rainbow tables care about *input* 20130403 02:33:47< AI0867> so you do that on the password itself, not the intermediate 20130403 02:33:53< AI0867> this is not going to help at all 20130403 02:34:09< AI0867> except perhaps increase the computation time of the hash 20130403 02:34:35< Gambit> Oh. I was thinking of generic md5 rainbow tables. 20130403 02:35:05< AI0867> if this is universal, you can be sure someone has phpbb-specific rainbow tables lying around 20130403 02:35:20< AI0867> it's not like they'd be hard to generate, just somewhat time-consuming 20130403 02:38:54-!- shadowm_desktop2 [ignacio@wesnoth/developer/shadowmaster] has quit [Read error: Connection reset by peer] 20130403 02:39:09-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev 20130403 02:41:21< Gambit> Actually... I'm a little bit confused now. 20130403 02:41:42< Gambit> I think once again I should stop narrating in real time. 20130403 02:44:21-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 248 seconds] 20130403 02:44:28< shadowm> And Gambit gives us the play-by-play. 20130403 02:44:42< Gambit> When will I learn to stop doing that? :P 20130403 02:45:19-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev 20130403 02:47:12< Gambit> Okay I have it. 20130403 02:47:19< Gambit> The salt is random. 20130403 02:47:33< Gambit> The salt and the password are combined and md5()ed. 20130403 02:47:43< Gambit> Then the salt and hash are combined and stored. 20130403 02:47:58< Gambit> As one string. 20130403 02:48:09< Gambit> So it is immune to rainbow tables. 20130403 02:48:24< shadowm> The Grickit has the ball! 20130403 02:48:48< shadowm> He's now running audience-wards with it! 20130403 02:50:03< AI0867> okay, so it's a sane system 20130403 02:50:04< AI0867> yay 20130403 02:50:16< AI0867> it's even all stuck together, so you might store it in LDAP 20130403 02:50:25< AI0867> though not with the default plugin 20130403 02:53:15< Gambit> And now for the unfortunate news, auth_ldap.php does not use that system 20130403 02:54:11< Gambit> I followed the stack up from login_ldap($username,$password) and confirmed that $password reaches it as plaintext 20130403 02:54:31< Gambit> auth_db.php's login method uses phpbb_check_hash() 20130403 02:57:33< fabi> Gambit: Thus the password is stored hashed in ldap but the request is plain text? 20130403 02:57:50< Gambit> login_db($username,$password) gets a row from the database for $username and then calls phpbb_check_hash($password,$row['hashed_password']) 20130403 02:57:52< AI0867> other way around 20130403 02:58:04< AI0867> phpbb uses ldap to store a plaintext password 20130403 02:58:14< AI0867> and then hashes whatever it gets from ldap for internal use 20130403 02:58:31< Gambit> login_ldap($username,$password) gets a row from ldap matching $username and then checks that $password matches $row['password'] 20130403 02:58:52< Gambit> (some of that was pseudocode) 20130403 02:59:05< Gambit> AI0867: yes 20130403 02:59:36< Gambit> the only place login_ldap() uses phpbb's hashing system is right before one of the possible return values 20130403 02:59:59< Gambit> which is then sent to a more generic part of phpbb's auth code 20130403 03:00:11< Gambit> user_add() to be specific 20130403 03:02:05< fabi> I think it is like this: The password is stored hashed in ldap. The ldap request sends a plain text password to the client. Then it gets encrypted again by the plugin, right? 20130403 03:02:13-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has quit [Ping timeout: 245 seconds] 20130403 03:02:56-!- naeg [~naeg@170-18-182-46.NbIServ.com] has quit [Read error: Operation timed out] 20130403 03:03:02< Gambit> Not unless some hashing is done by the ldap server. 20130403 03:03:05-!- exciton [chuck-the-@89.208.169.104] has quit [Ping timeout: 252 seconds] 20130403 03:03:10< Gambit> There is no phpbb code here that does that. 20130403 03:03:29< Gambit> And *if* the ldap server is doing something, it is reversable. 20130403 03:03:59< Gambit> login_ldap() gets a plaintext password in and compares it to the value it gets from the ldap server. 20130403 03:04:29-!- naeg [~naeg@170-18-182-46.NbIServ.com] has joined #wesnoth-dev 20130403 03:04:36< Gambit> login_ldap() is here: https://github.com/shikadilord/weldyn/blob/eb04f161147aac8729350f4e2baab1213a1b0fc7/includes/auth/auth_ldap.php#L101 20130403 03:04:51-!- exciton [chuck-the-@89.208.169.104] has joined #wesnoth-dev 20130403 03:04:58< Gambit> called from here: https://github.com/shikadilord/weldyn/blob/eb04f161147aac8729350f4e2baab1213a1b0fc7/includes/auth.php#L916 20130403 03:05:54< Gambit> and note that the equivilent login_db compares a hash of the password it receives here: https://github.com/shikadilord/weldyn/blob/eb04f161147aac8729350f4e2baab1213a1b0fc7/includes/auth/auth_db.php#L213 20130403 03:08:09< Gambit> s/phpbb code here/php code here/ 20130403 03:10:41< Gambit> fabi: do you have an ldap client? 20130403 03:11:06< Gambit> The simplest test would have been to log in to your ldap server and see whether or not you can get out your forum password. 20130403 03:12:21< fabi> Gambit: phpldapadmin 20130403 03:13:35< Gambit> So you have your forum all set up and created yourself an account? 20130403 03:15:15< fabi> Gambit: Note that the forum can't create ldap accounts. 20130403 03:15:36-!- ToBeFree [~tobefree@unaffiliated/tobefree] has quit [Read error: Operation timed out] 20130403 03:23:42< Gambit> ... 20130403 03:23:51< Gambit> Then what are we even doing discussing this? 20130403 03:24:05< shadowm> Is this the part where I should say "I told you so"? 20130403 03:25:34< Gambit> Okay so that's why you brought up front ends. 20130403 03:25:48< AI0867> from what I gather, half of the module needs to be thrown out/rewritten anyway, so this is hardly a big obstacle ;) 20130403 03:26:01< Gambit> We would need a global Wesnoth Account™ registration page. 20130403 03:26:30< Gambit> And then after that 20130403 03:26:35< Gambit> People have to go to phpbb's registration page 20130403 03:26:58< Gambit> And be really confused as to why they have to fill all that out when they just registered 20130403 03:27:11< Gambit> But also they have to make sure to use the exact same username and password. 20130403 03:27:19< Gambit> And then when they go to Redmine they have to register for it 20130403 03:27:24< Gambit> And once again make sure to use the same credentials. 20130403 03:28:07< Gambit> Oh. 20130403 03:28:19< Gambit> Actually the pass to user_add inside of login_ldap() makes perfect sense now. 20130403 03:28:24< Gambit> Wow. What a mess. 20130403 03:28:40< Gambit> s/inside of/directly after/ 20130403 03:28:57< Gambit> shadowm: If you're the kind of person that does that :P 20130403 03:29:19< Gambit> All we really need to know about LDAP is that it stores passwords in the clear. End of story. 20130403 03:29:48< shadowm> Yeah, no, we have enough with campaignd passphrases being stored in plain-text. 20130403 03:30:00< shadowm> Written wink. 20130403 03:31:03-!- ToBeFree [~tobefree@unaffiliated/tobefree] has joined #wesnoth-dev 20130403 03:39:49-!- Blueblaze [~Blueblaze@adsl-99-148-244-197.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev 20130403 03:41:52-!- davidwan [~chatzilla@60.8.123.21] has joined #wesnoth-dev 20130403 03:44:00< Gambit> So I have now spent 10 of the last 14 hours reading, writing, or thinking about php. 20130403 03:44:18< Gambit> And in a short 10 hours I will be back at work for another 8 hour stretch of reading, writing, or thinking about php. 20130403 03:44:23< Gambit> So I'm done with this conversation now. 20130403 03:44:37< Gambit> shadowm and fabi: You can go back to being at each other's throats over this 20130403 03:44:44< Gambit> But uh... now you have some facts. 20130403 03:45:24< Gambit> Also: php sucks. I hate it. :( 20130403 03:45:41< Gambit> I hear it clubs baby seals over the head. 20130403 03:46:26< shadowm> I... can go back to what now? 20130403 03:48:31< AI0867> whatever you were doing before you got Gambit to jump in and produce facts 20130403 03:48:51< shadowm> shurg 20130403 03:50:00< AI0867> 00:46 < shadowm> Gambit: Please provide your opinion now. 20130403 03:54:30-!- Guest62848 [~eli@dhip-029.rrw.residences.colby.edu] has quit [Ping timeout: 256 seconds] 20130403 04:01:10-!- ancestral [~ancestral@75-168-48-55.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.] 20130403 04:04:27-!- ancestral [~ancestral@67-6-53-117.mpls.qwest.net] has joined #wesnoth-dev 20130403 04:32:38-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy] 20130403 04:34:42-!- Ivanovic_ [~ivanovic@dtmd-4db2ad2a.pool.mediaWays.net] has joined #wesnoth-dev 20130403 04:36:49-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 248 seconds] 20130403 04:38:09-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has quit [Read error: Operation timed out] 20130403 04:38:36-!- Ivanovic_ is now known as Ivanovic 20130403 04:41:31-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection] 20130403 04:53:34-!- ancestral [~ancestral@67-6-53-117.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai] 20130403 05:11:44-!- Gambit [~gambit@wesnoth/developer/grickit] has quit [Remote host closed the connection] 20130403 05:14:29-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev 20130403 05:27:20-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 246 seconds] 20130403 05:35:46-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has joined #wesnoth-dev 20130403 05:47:45-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev 20130403 06:05:55-!- ancestral [~ancestral@67-6-53-117.mpls.qwest.net] has joined #wesnoth-dev 20130403 06:24:18-!- irker342 [~irker@ai0867.net] has quit [Quit: transmission timeout] 20130403 06:25:04-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Read error: Connection reset by peer] 20130403 06:25:53-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev 20130403 06:37:15-!- ancestral [~ancestral@67-6-53-117.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai] 20130403 06:56:54-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev 20130403 07:18:34-!- Blueblaze [~Blueblaze@adsl-99-148-244-197.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze] 20130403 07:30:27-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev 20130403 07:41:30-!- Ivanovic [~ivanovic@dtmd-4db2ad2a.pool.mediaWays.net] has quit [Changing host] 20130403 07:41:30-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev 20130403 07:42:48-!- davidwan [~chatzilla@60.8.123.21] has quit [Ping timeout: 264 seconds] 20130403 08:00:44-!- _Coffee [~david@ppp118-210-51-209.lns20.adl2.internode.on.net] has joined #wesnoth-dev 20130403 08:03:25< timotei_> shadowm: Well, I use thunderbird. And I set it to HTML specifically. 20130403 08:04:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20130403 08:04:53-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 248 seconds] 20130403 08:14:39-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving] 20130403 08:14:50< shadowm> timotei_: That's awful. 20130403 08:15:52-!- timotei_ is now known as timotei 20130403 08:15:59-!- timotei [~pi@79.119.97.34] has quit [Changing host] 20130403 08:15:59-!- timotei [~pi@wesnoth/developer/timotei] has joined #wesnoth-dev 20130403 08:16:03< timotei> shadowm: Why? 20130403 08:16:12< timotei> HTML is so much better than plain-text 20130403 08:16:15< shadowm> Because it's HTML and there is no plain text fallback. 20130403 08:17:23< shadowm> https://dl.dropbox.com/u/21371130/screenshots/crappy-html-email.png 20130403 08:17:27< timotei> So, does it make sense to make it send both in HTML and Text? 20130403 08:17:41< timotei> Ah 20130403 08:17:44< shadowm> The plain text fallback portion of the message is literally empty. 20130403 08:18:07< _Coffee>
What's so hard to  read about HTML. 
20130403 08:18:17< shadowm> Grr. 20130403 08:18:34 * shadowm spills _Coffee on the floor for the terrible joke. 20130403 08:19:00< timotei> shadowm: I've sent you an email with _both_ HTML & Test. Let me know if you can view it ok. 20130403 08:19:10< _Coffee> np, I'll be here all week tipping my hat ;) 20130403 08:19:11< timotei> The fact is, I always used HTML for my emails... 20130403 08:19:42< shadowm> timotei: Where did you send it? 20130403 08:22:44-!- davidwan [~chatzilla@61.48.184.202] has joined #wesnoth-dev 20130403 08:24:33< shadowm> Okay, the answer is: no, I cannot view it "ok" because I have no idea where you sent it and it's certainly not in my inbox. 20130403 08:24:37< shadowm> Bye. 20130403 08:26:14< timotei> shadowm: shadowm@wesnoth.org :P 20130403 08:35:55-!- davidwan [~chatzilla@61.48.184.202] has quit [Ping timeout: 240 seconds] 20130403 08:50:55-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev 20130403 08:51:25-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20130403 08:54:57-!- davidwan [~chatzilla@61.48.184.202] has joined #wesnoth-dev 20130403 09:19:16< AI0867> timotei: there are plenty of mailinglists that automatically reject emails that contain HTML, even if they do have a plain-text fallback 20130403 09:45:32-!- thunderstruck [~thunderst@cpc5-sgyl29-2-0-cust174.sgyl.cable.virginmedia.com] has joined #wesnoth-dev 20130403 09:45:33-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has left #wesnoth-dev [] 20130403 09:49:30-!- irker552 [~irker@ai0867.net] has joined #wesnoth-dev 20130403 09:49:30< irker552> Alexander van Gessel wesnoth:master * 1.11.2-94-ga53fe30 / images/buttons/checkbox-touched.png images/buttons/editor/brush-1-active.png images/buttons/editor/brush-1-pressed.png images/buttons/editor/brush-1.png images/buttons/editor/brush-2-active.png images/buttons/editor/brush-2-pressed.png images/buttons/editor/brush-2.png images/buttons/editor/brush-3-active.png images/buttons/editor/brush-3-pressed.png images/buttons/editor/brush-3.png i 20130403 09:49:31< irker552> Alexander van Gessel wesnoth:master * 1.11.2-96-g399ce6a / images/buttons/editor/units_button_editor-pressed.png images/buttons/editor/units_button_editor.png: Give two editor images an alpha channel 20130403 09:49:31< irker552> Alexander van Gessel wesnoth:master * 1.11.2-95-g83fee1d / utils/indexed2rgb.sh: Use grep's -q option rather than &>/dev/null 20130403 10:08:24-!- happygrue_ [~happygrue@c-76-119-97-171.hsd1.ma.comcast.net] has joined #wesnoth-dev 20130403 10:08:24-!- happygrue_ [~happygrue@c-76-119-97-171.hsd1.ma.comcast.net] has quit [Changing host] 20130403 10:08:24-!- happygrue_ [~happygrue@wesnoth/developer/wintermute] has joined #wesnoth-dev 20130403 10:09:14-!- mjs-de [~mjs-de@g224183085.adsl.alicedsl.de] has joined #wesnoth-dev 20130403 10:12:12-!- happygrue [~happygrue@wesnoth/developer/wintermute] has quit [Ping timeout: 256 seconds] 20130403 10:19:55< AI0867> shadowm: I can't get #20385 to crash for me, on 1.11.0 or HEAD 20130403 10:28:55-!- Dragos [~chatzilla@5-15-193-87.residential.rdsnet.ro] has quit [Ping timeout: 256 seconds] 20130403 10:39:29-!- siddharth [~siddharth@115.248.130.148] has joined #wesnoth-dev 20130403 10:42:51-!- siddharth [~siddharth@115.248.130.148] has quit [Remote host closed the connection] 20130403 10:43:02< shadowm> wesbot: bug 20385 20130403 10:43:03< wesbot> Bug #20385 Assigned to: Jody Northup Status: None Priority: 5 - Normal 20130403 10:43:06< wesbot> Summary: Using subnamespaces with [set_global_variable] causes invalid memory access 20130403 10:43:09< wesbot> Original submission: Using subnamespaces like "Namespace.Subnamespace" with th 20130403 10:43:12< wesbot> e [set_global_variable] WML action ( http://wiki.wesnoth.org/PersistenceWML ) ca 20130403 10:43:15< wesbot> URL: https://gna.org/bugs/?20385 20130403 10:43:18< wesbot> Attached file (1st): https://gna.org/bugs/download.php?file_id=16870 20130403 10:43:22-!- siddharth [~siddharth@115.248.130.148] has joined #wesnoth-dev 20130403 10:43:34< shadowm> AI0867: If you use valgrind you'll still see a bunch of invalid memory accesses that may or may not bite you in the rear later. 20130403 10:43:45< AI0867> right 20130403 10:43:51< shadowm> I originally didn't have an instant crash either, only a delayed crash. 20130403 10:43:55< AI0867> =/ 20130403 10:48:05< AI0867> valgrind spits out plenty of errors just by starting wesnoth... 20130403 10:48:50< shadowm> Yeah, ignore anything that has to do with strtod() calls. 20130403 10:53:31-!- siddharth [~siddharth@115.248.130.148] has quit [Remote host closed the connection] 20130403 10:53:57-!- siddharth [~siddharth@115.248.130.148] has joined #wesnoth-dev 20130403 10:54:23< siddharth> I'm new to wesmoth-dev, can someone guide me as to how to go about for solving the EasyCoding problems? 20130403 10:54:41-!- davidwan [~chatzilla@61.48.184.202] has quit [Read error: Connection reset by peer] 20130403 10:57:26-!- siddharth [~siddharth@115.248.130.148] has quit [Remote host closed the connection] 20130403 10:57:56-!- siddharth [~siddharth@115.248.130.148] has joined #wesnoth-dev 20130403 10:58:30-!- siddharth [~siddharth@115.248.130.148] has left #wesnoth-dev [] 20130403 10:59:01< thunderstruck> siddharth, have you found an interesting problem in that page yet? 20130403 11:00:18-!- siddharth950 [~siddharth@115.248.130.148] has joined #wesnoth-dev 20130403 11:00:39< siddharth950> Hi, I'm new to wesmoth-dev, can someone guide me as to how to go about for solving the EasyCoding problems? 20130403 11:04:00< _Coffee> siddharth950: some of the problems are out of date 20130403 11:05:24< thunderstruck> Indeed. For up-to-date bugs and feature requests go to bugs.wesnoth.org 20130403 11:05:38< _Coffee> I'm also newish here as a dev, but it looks like each person has there area of expertise 20130403 11:05:54< _Coffee> *their 20130403 11:06:11< siddharth950> _Coffee: Where should I start then? I have no idea of what to do, but I know C++ and AI 20130403 11:06:50< _Coffee> siddharth950: you might want to ask mattsc, who is developing new AI or crab who is a long time devloper with AI 20130403 11:07:36< _Coffee> neither are here at the moment though 20130403 11:07:47< siddharth950> okay, at what time are they generally online? 20130403 11:08:41< _Coffee> siddharth950: you could go on the forums under coders corner and get an idea of some of the new AI stuff and LUA to see if it is along what you might be able to contribute to 20130403 11:09:22< siddharth950> _Coffee: thanks a lot, I'll try that 20130403 11:10:03-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy] 20130403 11:10:58-!- siddharth950 [~siddharth@115.248.130.148] has quit [Remote host closed the connection] 20130403 11:11:25-!- siddharth950 [~siddharth@115.248.130.148] has joined #wesnoth-dev 20130403 11:11:52< AI0867> now that I'm looking in the right places, I see really evil config magic being done 20130403 11:12:55-!- davidwan [~chatzilla@111.199.94.27] has joined #wesnoth-dev 20130403 11:15:09-!- siddharth950 [~siddharth@115.248.130.148] has quit [Client Quit] 20130403 11:24:27-!- Octalot [~noct@host86-148-68-117.range86-148.btcentralplus.com] has joined #wesnoth-dev 20130403 11:31:09-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has joined #wesnoth-dev 20130403 11:32:09-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!] 20130403 11:48:40-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [] 20130403 11:55:49-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has quit [Ping timeout: 248 seconds] 20130403 11:56:57-!- EdB [~edb@89-93-184-215.hfc.dyn.abo.bbox.fr] has joined #wesnoth-dev 20130403 11:58:08-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev 20130403 12:03:57-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20130403 12:45:33-!- davidwan [~chatzilla@111.199.94.27] has quit [Ping timeout: 252 seconds] 20130403 12:53:44-!- irker552 [~irker@ai0867.net] has quit [Quit: transmission timeout] 20130403 13:05:03-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev 20130403 13:15:00-!- Wesbane [~Wesbane@sub227-213.elpos.net] has joined #wesnoth-dev 20130403 13:18:19-!- Wesbane [~Wesbane@sub227-213.elpos.net] has quit [Client Quit] 20130403 13:19:04-!- exciton [chuck-the-@89.208.169.104] has quit [Read error: Connection reset by peer] 20130403 13:19:24-!- exciton [chuck-the-@89.208.169.104] has joined #wesnoth-dev 20130403 13:20:37-!- happygrue_ is now known as happygrue 20130403 13:22:57-!- Octalot [~noct@host86-148-68-117.range86-148.btcentralplus.com] has quit [Remote host closed the connection] 20130403 13:26:04-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev 20130403 13:55:33-!- Guest62848 [~eli@dhip-029.rrw.residences.colby.edu] has joined #wesnoth-dev 20130403 14:10:24-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: bumbadadabum] 20130403 14:14:04-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-dev 20130403 14:19:49-!- Ayne [~Ayne@wesnoth/developer/ayne] has joined #wesnoth-dev 20130403 14:32:08-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has quit [Quit: Leaving.] 20130403 14:42:13-!- irker293 [~irker@ai0867.net] has joined #wesnoth-dev 20130403 14:42:13< irker293> Alexander van Gessel wesnoth:master * 1.11.2-97-ga3fd22e / src/persist_var.cpp: Fix some error messages in persist 20130403 14:42:14< irker293> Alexander van Gessel wesnoth:master * 1.11.2-98-g3939490 / src/persist_var.cpp: Correct another persist error message 20130403 14:52:05-!- Ayne [~Ayne@wesnoth/developer/ayne] has quit [Quit: Leaving] 20130403 14:52:06-!- thunderstruck [~thunderst@cpc5-sgyl29-2-0-cust174.sgyl.cable.virginmedia.com] has quit [Quit: Leaving] 20130403 14:53:48< irker293> Alexander van Gessel wesnoth:master test 20130403 14:54:20-!- thunderstruck [~thunderst@cpc5-sgyl29-2-0-cust174.sgyl.cable.virginmedia.com] has joined #wesnoth-dev 20130403 14:54:27< irker293> Alexander van Gessel wesnoth:master test 20130403 15:00:42-!- davidwan [~chatzilla@60.8.123.20] has joined #wesnoth-dev 20130403 15:06:42-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 276 seconds] 20130403 15:07:32< AI0867> Ivanovic: the poll has ended. I believe esr is now waiting for your signal to freeze the current repo and redo the conversion 20130403 15:10:09-!- EdB [~edb@89-93-184-215.hfc.dyn.abo.bbox.fr] has quit [Quit: Konversation terminated!] 20130403 15:14:15-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev 20130403 15:15:58-!- Yukiria [~chatzilla@124.109.10.167] has joined #wesnoth-dev 20130403 15:18:25-!- Yukiria is now known as vultraz 20130403 15:18:35-!- artisticdude [~artisticd@108.sub-70-192-196.myvzw.com] has joined #wesnoth-dev 20130403 15:19:22-!- vultraz is now known as Yukiria 20130403 15:19:45-!- Yukiria [~chatzilla@124.109.10.167] has quit [Client Quit] 20130403 15:21:27-!- Yukiria [~chatzilla@124.109.10.167] has joined #wesnoth-dev 20130403 15:22:51-!- Yukiria is now known as vultraz 20130403 15:23:00-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host] 20130403 15:23:00-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev 20130403 15:24:06< vultraz> I think I found a bug. Seems if wesnoth (1.11.2) can't close a quoted string, it will indefinitely use memory until it runs out and aborts itself 20130403 15:26:14< elias> sounds more like the bug is in the WML (missing quote) 20130403 15:27:06< vultraz> yeah, I was missing a quote. But I don't think a missing quote is supposed to cause wesnoth to freeze and cause my whole system to freeze as well from lack of memory 20130403 15:27:15-!- LordNasty [~NaSTy@93-43-168-243.ip92.fastwebnet.it] has quit [Ping timeout: 252 seconds] 20130403 15:29:31< AI0867> true 20130403 15:29:31-!- prkc [~negusnyul@4E5CCA6E.dsl.pool.telekom.hu] has joined #wesnoth-dev 20130403 15:29:38< AI0867> can you throw up a testcase somewhere? 20130403 15:29:40< elias> well, i'd certainly agree, i needed some advanced workarounds around that behavior to parse addons on units.wesnoth.org :) 20130403 15:30:12< AI0867> WML is allowed to abort wesnoth, but not to crash or hang it 20130403 15:32:01< elias> i don't think it uses infinite memory since i could never see the behavior at home with 16GB RAM 20130403 15:32:25< AI0867> I can lower the ulimit for testing 20130403 15:33:48< vultraz> It happens for me whenever I try to load a scenario with an odd number of quotes (eg, one missing) 20130403 15:38:55-!- mjs-de [~mjs-de@g224183085.adsl.alicedsl.de] has quit [Ping timeout: 258 seconds] 20130403 15:41:34< bumbadadabum> that doesn't happen to me IIRC 20130403 15:43:20-!- mjs-de [~mjs-de@g224183085.adsl.alicedsl.de] has joined #wesnoth-dev 20130403 15:43:51< zookeeper> vultraz, you sure the missing quote isn't somehow causing infinite macro recursion or something? 20130403 15:44:00< zookeeper> just a thought 20130403 15:44:01< bumbadadabum> ^ 20130403 15:44:15< bumbadadabum> This was the only thing that ever crashed wesnoth in this way 20130403 15:44:23< bumbadadabum> for me, at least 20130403 15:44:51-!- LordNasty [~NaSTy@93-43-168-243.ip92.fastwebnet.it] has joined #wesnoth-dev 20130403 15:45:20< vultraz> well, in the current case, the missing string was at the end of a macro argument. On the previous occasion, though, I think it was just at the end of a message= string 20130403 15:45:59< vultraz> (though in the previous case, I force restarted my computer before wesnoth had a chance to kill itself, since everything was frozen) 20130403 15:46:03-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has joined #wesnoth-dev 20130403 15:47:10< lipkab> fabi: http://forums.wesnoth.org/viewtopic.php?f=19&t=37290 is obsolete now, isn't it? 20130403 15:47:22< lipkab> (Just saying because it's a sticky) 20130403 15:47:45< fabi> lipkab: No, it still holds. 20130403 15:48:30< fabi> lipkab: I will need a few more tests and debuging but the unit feature will be commited soon. 20130403 15:49:07< lipkab> Okay. 20130403 15:49:54< artisticdude> fabi: For future reference, what is the proper image name for the mouseover state of a GUI button? 20130403 15:49:58< artisticdude> -touched? 20130403 15:51:04< fabi> artisticdude: I told it so. But I do not claim that the name is best choice. 20130403 15:52:03< artisticdude> It makes perfect sense to me. 20130403 15:52:15< artisticdude> I just wanted to make sure that was the standard. 20130403 15:52:54< artisticdude> I've been using 'enabled', but that didn't seem like a very accurate description of the state in question. 20130403 15:53:17< fabi> artisticdude: wait 20130403 15:53:23< zookeeper> "hover" seems like a pretty standard way of describing mouseover state 20130403 15:56:13< fabi> artisticdude: The mouseover states are called