--- Log opened Thu Mar 10 00:00:32 2016
20160310 00:03:33< Nobun> committed
20160310 00:06:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 00:13:25< Nobun> the userdoc and sourcedoc will be updated later
20160310 00:14:03< Nobun> a little question, celticminstrel: perhaps we must consider escaping \ with \\ ?
20160310 00:14:10< celticminstrel> Not sure.
20160310 00:14:31< celticminstrel> If it's followed by n or ", probably, but in the general case, I don't know.
20160310 00:14:31< Nobun> I didn't think about it, but since the \ is the escape character also in pot
20160310 00:14:59< Nobun> the problem is actually on WML strings
20160310 00:15:11< Nobun> since \ is used as escape character in Lua strings
20160310 00:15:32< Nobun> so on lua string you have to write \\ to print \
20160310 00:15:41< Nobun> but on WML string, not
20160310 00:15:55< Nobun> on pot you had to write \\, I assume
20160310 00:16:00< celticminstrel> In WML, a \ is literal.
20160310 00:16:06< celticminstrel> I don't know how pot works.
20160310 00:16:38< Nobun> since \n is for newline, I assume it uses a kind of c-like syntax
20160310 00:16:54< celticminstrel> This problem can affect bracketed Lua strings, too.
20160310 00:17:41< Nobun> so, on bracketed strings \ is not considered as an escape?
20160310 00:18:05< Nobun> \n is not a newline in bracketed lua string?
20160310 00:19:48< celticminstrel> Right, the backslash has no special meaning at all in bracketed strings.
20160310 00:19:55< celticminstrel> If you write \ you get \
20160310 00:20:00< celticminstrel> If you write \\ you get \\
20160310 00:20:04< celticminstrel> And if you write \n you get \n
20160310 00:23:44-!- atarocch [~atarocch@31.159.149.240] has quit [Remote host closed the connection]
20160310 00:34:36< Nobun> ok... so I will add a new fix
20160310 00:35:01< Nobun> on lua bracketed strings and on wml strings \ must be replaced by in \\ in po
20160310 00:35:36< Nobun> I assume that \ is a escape character in pot files since it is used to escape quotes (\") and to print newlines (\n)
20160310 00:35:58< celticminstrel> It's a good assumption, but I suggest you do an Internet search to confirm it.
20160310 00:41:21< Nobun> confirmed
20160310 00:41:33< celticminstrel> :)
20160310 00:41:39< Nobun> Another two escape sequences, usually of much lower frequency than the double quote and the newline, are the tabulator \t and the backslash itself \\ (because single backslash always starts an escape sequence)
20160310 00:42:15< celticminstrel> Hm. Does po let you put a literal tab in the string, or do tabs need to be replaced with \t?
20160310 00:42:21< Nobun> so now I have to figure how to make this change without introducing a bug
20160310 00:43:12< Nobun> I think that litteral tab is converted in spaces or ignored... neverer tryed
20160310 00:44:08< celticminstrel> \t is also valid in Lua quoted strings with the same meaning.
20160310 00:44:28< Nobun> yep... since it is the c-like escape codes
20160310 00:44:37< celticminstrel> Mmhm.
20160310 00:44:49< Nobun> \n, \r, \t, \", \', \\ are exactly the same in C
20160310 00:45:32< Nobun> so I can be add a rule that adds a \ when \ is to be considered a litteral \
20160310 00:45:35< celticminstrel> Huh, Lua understands \r too. o.o
20160310 00:45:58-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has quit [Disconnected by services]
20160310 00:46:05< celticminstrel> And \b
20160310 00:46:10< Nobun> Yep... since lua is written in C, it re-used some ideas around c
20160310 00:46:14< celticminstrel> Maybe should warn about those being in translated strings.
20160310 00:46:27-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has joined #wesnoth-dev
20160310 00:46:32< celticminstrel> Though I'd hope people wouldn't do it anyway.
20160310 00:46:39< Nobun> well... this is related to lua code
20160310 00:46:46< celticminstrel> Yes.
20160310 00:46:48< Nobun> since, on WML and lua bracketed it is litteral
20160310 00:46:59< celticminstrel> Yes.
20160310 00:47:12< Nobun> so adding a \ in po when raw \ is expected should be the right solution
20160310 00:47:16< celticminstrel> (Except WML has the one escape sequence of double quotes meaning a single one.)
20160310 00:47:22< Nobun> yep
20160310 00:47:25< celticminstrel> Something like that.
20160310 00:47:29< Nobun> but it is already treated and fixed
20160310 00:47:49< Nobun> (it was a bug that loonycyborg reported, but already fixed yesterday)
20160310 00:48:33< Nobun> now I have to figure how to add a \ when litteral \ found on WML and bracketed lua strings without introducing bugs...
20160310 00:50:06< celticminstrel> Easy in WML - line.replace(r'\',r'\\')
20160310 00:50:34< celticminstrel> I guess that'll work in Lua too, actually.
20160310 00:50:50< Nobun> yep... but I must do it properly
20160310 00:50:57< Nobun> the code is more or less that
20160310 01:14:04-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20160310 01:14:18-!- ancestral [~ancestral@63.92.240.233] has quit [Client Quit]
20160310 01:25:11< gfgtdf> vultraz: the known bugs secriptions inthe release annoucement still need to be fixed.
20160310 01:28:49-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 01:31:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 01:46:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 01:47:58< Nobun> celticminstrel: the fix is harder than it appers on a first look. The problem is when a litteral \" is found becouse the fix rules conflicts and until now I cannot find a good solution
20160310 01:49:05< Nobun> replacing \ with \\ creates a unascaped " since the rule who escapes the " will work if " is not preceded by \... but it this case it is...
20160310 01:49:37-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has quit [Read error: Connection reset by peer]
20160310 02:00:55-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has joined #wesnoth-dev
20160310 02:01:31-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has quit [Client Quit]
20160310 02:05:49-!- Greg-Boggs [~greg_bogg@50.252.20.85] has joined #wesnoth-dev
20160310 02:10:17-!- Greg-Boggs [~greg_bogg@50.252.20.85] has quit [Ping timeout: 244 seconds]
20160310 02:13:02-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160310 02:16:45-!- gfgtdf_ [~chatzilla@f054153247.adsl.alicedsl.de] has joined #wesnoth-dev
20160310 02:19:37-!- gfgtdf [~chatzilla@f054060056.adsl.alicedsl.de] has quit [Ping timeout: 244 seconds]
20160310 02:19:40-!- gfgtdf_ is now known as gfgtdf
20160310 02:24:45< vultraz> gfgtdf: could you help me debug this https://gna.org/bugs/?24511 since you've worked with listboxes 
20160310 02:25:05< vultraz> I tested using the steps he provided, but I can't reproduce it
20160310 02:25:06< vultraz> but
20160310 02:25:46< vultraz> if I create a Young Ogre unit and then open the dialog, no row is selected and if I put in a filter I can select more than one row (which isn't a dialog bug)
20160310 02:25:49< vultraz> (I don't think)
20160310 02:27:00< vultraz> gfgtdf: https://www.dropbox.com/s/lrgog4fkt2xc2tp/unit_create_listbox_issue.PNG?dl=0
20160310 02:29:45< gfgtdf> what do you mean by 'not a dialog bug' ?
20160310 02:31:20< vultraz> i think there might be some bad listbox behavior here too
20160310 02:31:35< gfgtdf> vultraz: hmm ok
20160310 02:31:59< gfgtdf> vultraz: hmm what you said doesnt rproduce for me
20160310 02:32:44< vultraz> for one, when you set the rows shown get_item_count() still returns the max number, not the displayed number
20160310 02:33:10< gfgtdf> vultraz: when i repoen that dialog is still shows the young orge as selected
20160310 02:33:23< vultraz> gfgtdf: yes, now type something into the filter
20160310 02:33:31< vultraz> one letter is fine, then try to select rows
20160310 02:35:09< gfgtdf> vultraz: hmm ok isee
20160310 02:35:20< gfgtdf> could eprouce i mean
20160310 02:38:44< gfgtdf> vultraz: i also get a lot of errors 'Mouse down on non focused widget and mouse not captured, we missed events.' in stderr, maybe tehy are related
20160310 02:39:26< gfgtdf> vultraz: hmm maybe not
20160310 02:39:32< vultraz> i always get those
20160310 02:39:33< vultraz> idk why
20160310 02:40:23-!- midzer [~quassel@p5B312EF2.dip0.t-ipconnect.de] has quit [Quit: No Ping reply in 180 seconds.]
20160310 02:40:23< gfgtdf> vultraz: hmm last time i checked stderr hey werent there
20160310 02:40:50< celticminstrel> Nobun: Is that really a problem? In WML, if you write \" it means the string ends with a backslash; if you wanted a literal \" you would write \""
20160310 02:41:30-!- midzer [~quassel@p5B312EF2.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160310 02:41:52< Nobun> it is a problem on lua bracketed string
20160310 02:42:02< Nobun> in the WML string not
20160310 02:42:36< celticminstrel> If you write \" in a bracketed Lua string, you get \", so it should be converted to \\\".
20160310 02:42:45-!- ancestral [~ancestral@71-220-49-102.mpls.qwest.net] has joined #wesnoth-dev
20160310 02:42:48< Nobun> unluckly not
20160310 02:42:50< Nobun> this becouse
20160310 02:42:57< gfgtdf> vultraz: well from oberving the bahviour i'd assume that listbox internals usually select the next visible item afert teh previously selected item if the prviously selected item becomes invisible
20160310 02:42:59< Nobun> I have a rule that replace \ with \\
20160310 02:43:13< gfgtdf> vultraz: and that this fails if there is no such item
20160310 02:43:20< Nobun> but the rule for " search for " not preceded by \
20160310 02:43:28< Nobun> so the " will be unescaped
20160310 02:44:02< celticminstrel> Just replace \ with \\ and then " with \"?
20160310 02:44:10< Nobun> for this reason, on bracketed lua strings, all \ symbols before " will be erased at the moment, waiting to find a better solution
20160310 02:44:14< celticminstrel> I don't see the problem here...
20160310 02:44:32< celticminstrel> Huh?
20160310 02:44:38< Nobun> the reason is
20160310 02:45:05< Nobun> that \" is considered as "already escaped "... don't escape it 
20160310 02:45:11< celticminstrel> Why?
20160310 02:45:27< gfgtdf> vultraz: it's alread late here but i could try a quick untested githubwebpage commit if you want.
20160310 02:45:32< Nobun> since the luapending function works for any kind of functions
20160310 02:45:34< vultraz> sure
20160310 02:45:53-!- ancestral [~ancestral@71-220-49-102.mpls.qwest.net] has quit [Remote host closed the connection]
20160310 02:46:02< celticminstrel> So you're saying you use the same function for quoted and for bracketed strings?
20160310 02:46:02< Nobun> I don't want to escape, for example an already escaped \" coming from lua quoted string
20160310 02:46:07< celticminstrel> (But different for WML strings.)
20160310 02:46:12< Nobun> yep
20160310 02:46:20< Nobun> PendingWmlString
20160310 02:46:21< celticminstrel> But you can ask what kind of string you have, can't you?
20160310 02:46:30< Nobun> and PendingLuaString behave very differently
20160310 02:46:38< celticminstrel> Does the PendingLuaString not remember whether it was quoted or bracketed?
20160310 02:46:46< Nobun> yep
20160310 02:47:15< Nobun> it has luatype variable (1 means quoted, 2 means ' and 3 means bracketed)
20160310 02:47:50< Nobun> 'luatype1'. 'luatype2' and 'luatype3' to be absolutely correct
20160310 02:47:52< celticminstrel> So you can in fact have different escaping behaviour for each type.
20160310 02:48:09< Nobun> yep
20160310 02:48:22< Nobun> I have to think how to do properlu
20160310 02:48:48< celticminstrel> Well, for bracketed strings specifically, just replacing all \ and all " with the escaped form should work.
20160310 02:48:55< Nobun> (this night I will not go bed to assist to my girlfriend if needed, so why I am already here so late :P)
20160310 02:48:57-!- irker037 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160310 02:48:57< irker037> wesnoth: gfgtdf wesnoth:master 1667e8680cdd / src/gui/widgets/generator.cpp: attempt to fix bug in gui2 listbox filters. https://github.com/wesnoth/wesnoth/commit/1667e8680cdd9ee1e8c6b270fe6064caf70b2f27
20160310 02:48:59< celticminstrel> Obviously, you need to do something a bit different with quoted strings.
20160310 02:49:19< gfgtdf> vultraz: ^ feel free to revert if it doesnt improve the listboxes
20160310 02:50:29-!- gfgtdf [~chatzilla@f054153247.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 44.0.2/20160210153822]]
20160310 02:52:17< Nobun> perhaps now I understood how to do
20160310 02:52:50< Nobun> I'm tired (but I cannot go bed) so I'm starting to have problems on thinking :P
20160310 02:53:03< Nobun> perhaps I found a way... now testing
20160310 02:53:10< celticminstrel> In formula, "not" has a lower precedence than "where"... I'm wondering if this could cause weird things...
20160310 02:55:45< Nobun> celticminstrel: ok now it works perfectly, it seems
20160310 02:59:57< Nobun> celticminstrel: https://transfer.sh/nMHeM/test.tar.gz
20160310 03:00:23< Nobun> this is test results (fake test.cfg and fake test.lua produced the test.po that seems correct)
20160310 03:00:35< Nobun> I will add the commit tomorrow
20160310 03:01:14< Nobun> let me know if you have some ideas if I missed something about escape codes
20160310 03:01:19< Nobun> good night
20160310 03:01:45-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has quit [Quit: Salve a tutti]
20160310 03:02:13-!- Greg-Boggs [~greg_bogg@50.252.20.85] has joined #wesnoth-dev
20160310 03:04:25-!- Greg-Boggs [~greg_bogg@50.252.20.85] has quit [Remote host closed the connection]
20160310 03:06:49< celticminstrel> Hm. I suspect the relative precedence of not and where may be irrelevant.
20160310 03:07:50-!- Greg-Boggs [~greg_bogg@50.252.20.85] has joined #wesnoth-dev
20160310 03:09:55-!- Greg-Boggs [~greg_bogg@50.252.20.85] has quit [Client Quit]
20160310 03:19:02-!- travis-ci [~travis-ci@ec2-54-158-178-75.compute-1.amazonaws.com] has joined #wesnoth-dev
20160310 03:19:03< travis-ci> wesnoth/wesnoth#8815 (master - 1667e86 : gfgtdf): The build is still failing.
20160310 03:19:03< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/114963902
20160310 03:19:03-!- travis-ci [~travis-ci@ec2-54-158-178-75.compute-1.amazonaws.com] has left #wesnoth-dev []
20160310 03:19:27< celticminstrel> Stiiiiiiil failing...
20160310 03:19:31< celticminstrel> ^l
20160310 03:21:30< vultraz> BLAGH
20160310 03:21:34< vultraz> UNIT TESTS
20160310 03:21:49< celticminstrel> BLAGH to you too.
20160310 03:22:34< irker037> wesnoth: Charles Dang wesnoth:master 8b22341e19af / src/tests/gui/test_gui2.cpp: Attempt to fix tests, again https://github.com/wesnoth/wesnoth/commit/8b22341e19af440a6b4a310e38e6c151246f8268
20160310 03:22:42< vultraz> no I'm BLAGHing at travis
20160310 03:41:31< celticminstrel> I have one or two new things to push to PR624, but I think I'll wait until the build is fixed on master.
20160310 03:41:50< celticminstrel> PR618 is currently on hold.
20160310 03:42:13< celticminstrel> And PR615 is awaiting mattsc.
20160310 03:52:38-!- travis-ci [~travis-ci@ec2-54-158-178-75.compute-1.amazonaws.com] has joined #wesnoth-dev
20160310 03:52:39< travis-ci> wesnoth/wesnoth#8816 (master - 8b22341 : Charles Dang): The build is still failing.
20160310 03:52:39< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/114968381
20160310 03:52:39-!- travis-ci [~travis-ci@ec2-54-158-178-75.compute-1.amazonaws.com] has left #wesnoth-dev []
20160310 03:52:54< celticminstrel> Nope!
20160310 03:57:15< celticminstrel> Hmm. List indexing has even lower precedence than not, because it's not part of the operator system... trying to think if this is okay...
20160310 04:00:45< celticminstrel> In C++ it has almost the highest precedence.
20160310 04:09:56-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-176-151.zoominternet.net] has joined #wesnoth-dev
20160310 04:17:17-!- PrototypeX29A [~peter@180.159.194.21] has joined #wesnoth-dev
20160310 04:17:24< PrototypeX29A> hi
20160310 04:18:47-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-176-151.zoominternet.net] has quit []
20160310 04:19:21< celticminstrel> Hi.
20160310 04:21:51-!- PrototypeX29A [~peter@180.159.194.21] has quit [Ping timeout: 250 seconds]
20160310 04:22:07-!- PrototypeX29A [~peter@180.159.194.21] has joined #wesnoth-dev
20160310 04:24:41< celticminstrel> I might have swapped "high" and "low" in the above.
20160310 04:25:03< celticminstrel> Or not.
20160310 04:25:40< celticminstrel> Okay yeah, I said it correctly.
20160310 04:25:52< celticminstrel> I think I've now fixed it.
20160310 04:26:33< celticminstrel> I also fixed exponentiation being left-associative. It was surprisingly easy, even.
20160310 04:26:56< celticminstrel> Fortunately, it's the only operator that really should be right-associative.
20160310 04:32:41< celticminstrel> Function call still has lowest precedence, but... I think that's not a problem since functions are only at toplevel scope.
20160310 04:33:45< celticminstrel> If "x.f(y)" were possible, it would be a problem.
20160310 04:36:26< celticminstrel> Oh wow.
20160310 04:36:40< celticminstrel> The dot operator doesn't work as you would expect, coming from other programming languages.
20160310 04:36:57< celticminstrel> Need to update the wiki page with this new information.
20160310 04:40:42< celticminstrel> Incidentally, "x.f(y)" is impossible, but it doesn't do what it looks like.
20160310 04:44:17-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160310 04:55:10< celticminstrel> So basically, the left operand of the dot operator in WFL is an "object", or what the code callse formula_callable.
20160310 04:55:18< celticminstrel> The right operand is a formula.
20160310 04:55:39< celticminstrel> The right operand is evaluated with the contents of the left operand brought into scope.
20160310 04:56:10< celticminstrel> I suppose it's comparable to "with" in JavaScript.
20160310 04:56:48< celticminstrel> "my_leader.a" returns null. "my_leader.a where a = 5" returns 5, even though there is no a in my_leader, because the where clause scope defines a.
20160310 05:00:03< celticminstrel> "my_leader . (hitpoints < max_hitpoints / 2)" returns 1 if your leader is at less than half their max hitpoints.
20160310 05:02:43< celticminstrel> It's left associative, but I'm not sure if you could construct a case where that matters.
20160310 06:04:37< vultraz> &@(*&*#*(&@$*(# travis
20160310 06:05:53< irker037> wesnoth: Charles Dang wesnoth:master 07302e273048 / src/tests/gui/test_gui2.cpp: More attempts to fix tests https://github.com/wesnoth/wesnoth/commit/07302e273048f3dd0ab95a55d2e293838a2cd116
20160310 06:20:29-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20160310 06:23:13< vultraz> gfgtdf: thanks that fixes it
20160310 06:35:29-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has joined #wesnoth-dev
20160310 06:36:43-!- travis-ci [~travis-ci@ec2-54-158-178-75.compute-1.amazonaws.com] has joined #wesnoth-dev
20160310 06:36:44< travis-ci> wesnoth/wesnoth#8817 (master - 07302e2 : Charles Dang): The build is still failing.
20160310 06:36:44< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/114985083
20160310 06:36:44-!- travis-ci [~travis-ci@ec2-54-158-178-75.compute-1.amazonaws.com] has left #wesnoth-dev []
20160310 06:39:08< vultraz> OH FKLSDJFIOJWIOJFIOJIO
20160310 06:43:42-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160310 06:43:42< vultraz> screw travis 
20160310 06:54:52-!- Kwandulin [~Miranda@p200300760F7D73B4612B9DB5C1B2FD47.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160310 07:26:40-!- Crendgrim_ [~crend@wesnoth/forum-moderator/crendgrim] has joined #wesnoth-dev
20160310 07:27:51-!- Crendgrim [~crend@wesnoth/forum-moderator/crendgrim] has quit [Ping timeout: 276 seconds]
20160310 07:29:26-!- Prototyp1X29A [~peter@58.34.17.72] has joined #wesnoth-dev
20160310 07:32:43-!- PrototypeX29A [~peter@180.159.194.21] has quit [Ping timeout: 244 seconds]
20160310 07:33:20< wedge009> vultraz: Sad to see you're getting ticked off by travis. I think you've been making some more gui2 stuff, right? I put the missing files in PR #625.
20160310 07:33:27< irker037> wesnoth: Wedge009 wesnoth:master 206bed161e50 / projectfiles/VC9/wesnoth.vcproj: Add missing files https://github.com/wesnoth/wesnoth/commit/206bed161e50d4e41757bf48e02be27e438893af
20160310 07:33:29< irker037> wesnoth: Andreas wesnoth:master 6128aef97ae0 / projectfiles/VC9/wesnoth.vcproj: Merge pull request #625 from Wedge009/patch-1 https://github.com/wesnoth/wesnoth/commit/6128aef97ae063e1509e9980363716bd71c772fa
20160310 07:33:37< vultraz> hah!
20160310 07:33:40< vultraz> dat timing
20160310 07:33:44< wedge009> O.O
20160310 07:33:49< wedge009> I'm amazed.
20160310 07:34:03 * Aginor reads emails on occasion
20160310 07:37:41< vultraz> I just wish I could figure out why travis is failing
20160310 07:38:07 * Aginor goes to look
20160310 07:38:53< vultraz> it's a bunch of gui2 stuff but I have absolutely no idea what's causing it
20160310 07:41:18< Aginor> vultraz: maybe this is related: 20160310 06:33:45 error validation: Tag [unit_preview_pane] may not be used in [column]
20160310 07:41:45< vultraz> that's a lack of updated schema... but it's possible
20160310 07:41:54< Aginor> warnings like these are usually done for a good reason
20160310 07:42:09< Aginor> if you violate a spec/schema, don't expect things to work
20160310 07:45:26< vultraz> I'll see if I can figure out the correct schema syntax
20160310 07:47:33< Aginor> sounds like a good plan :)
20160310 08:04:26-!- travis-ci [~travis-ci@ec2-54-163-65-175.compute-1.amazonaws.com] has joined #wesnoth-dev
20160310 08:04:27< travis-ci> wesnoth/wesnoth#8819 (master - 6128aef : Andreas): The build is still failing.
20160310 08:04:27< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/114994528
20160310 08:04:27-!- travis-ci [~travis-ci@ec2-54-163-65-175.compute-1.amazonaws.com] has left #wesnoth-dev []
20160310 08:05:51< wedge009> Aginor: Looks like the journey map isn't 100% fixed: https://gna.org/bugs/index.php?24510
20160310 08:05:57< wedge009> I think this is related, isn't it?
20160310 08:07:57< Aginor> that looks related, yes
20160310 08:08:07< Aginor> I'll give it attention tomorrow or over the weekend
20160310 08:08:21< Aginor> I'm planning to fix a number of lingering minor bugs
20160310 08:08:28< Aginor> and probably rip out SDL1
20160310 08:21:49< wedge009> Kill it! Kill it with fire!
20160310 08:22:00< wedge009> Um, I mean yes, very good. Carry on.
20160310 08:24:45< Aginor> I shall
20160310 08:36:13-!- Lohengramm [sid1929@gateway/web/irccloud.com/x-isprwspinenskqqz] has quit [Ping timeout: 250 seconds]
20160310 08:36:46-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160310 08:39:19-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 252 seconds]
20160310 08:39:19-!- wedge010 is now known as wedge009
20160310 08:39:48-!- Kwandulin [~Miranda@p200300760F7D73B4612B9DB5C1B2FD47.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160310 08:45:47-!- Lohengramm [sid1929@gateway/web/irccloud.com/x-lryweagybqobssfo] has joined #wesnoth-dev
20160310 08:52:52-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 260 seconds]
20160310 09:00:23-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20160310 09:23:02< vultraz> Aginor: once sdl1 is gone I'll start on the flip() refactor
20160310 09:23:06< vultraz> (with your help, of course)
20160310 09:26:37-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 252 seconds]
20160310 09:28:06< Aginor> I've got strong opinions on the matter ;)
20160310 09:29:16-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit []
20160310 09:30:31< vultraz> Aginor: oh?
20160310 09:30:35< vultraz> what do you mean
20160310 09:30:48< Aginor> this is also an important step towards ogl
20160310 09:30:52< vultraz> (disclaimer: I'll probably just push code around wherever you tell me to)
20160310 09:31:14< Aginor> ok
20160310 09:40:30-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20160310 09:43:03-!- Appleman1234 [~Appleman1@KD106161141205.au-net.ne.jp] has quit [Ping timeout: 264 seconds]
20160310 09:48:19< vultraz> Aginor: the biggest problem is I can't figure out what you mean by "draw cycle"
20160310 09:49:47< vultraz> is it like, the drawing that occurs per-frame?
20160310 09:50:01< vultraz> ie, n flip() calls per second where n = fps?
20160310 09:50:17< vultraz> or is it a clump of draw operations that must be done, then flip() called?
20160310 09:50:46< vultraz> or to put it better, only at the end of a cycle where the back framebuffer is modified?
20160310 10:09:46-!- mjs-de [~mjs-de@x4db6b340.dyn.telefonica.de] has joined #wesnoth-dev
20160310 10:33:52-!- irker037 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160310 10:39:39-!- Appleman1234 [~Appleman1@KD106161151150.au-net.ne.jp] has joined #wesnoth-dev
20160310 10:43:49-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160310 10:45:20-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 250 seconds]
20160310 10:45:21-!- wedge010 is now known as wedge009
20160310 10:51:22-!- irker852 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160310 10:51:22< irker852> wesnoth: Charles Dang wesnoth:master 49102677a975 / images/icons/icon-hotkeys.png: Improved Hotkeys icon for prefs https://github.com/wesnoth/wesnoth/commit/49102677a97547590c927ccf6b9d7b5b0f62116d
20160310 10:54:33< wedge009> Was it just changed to a lady holding a key?
20160310 10:55:33< vultraz> it was changed to a better sprite
20160310 10:55:50< vultraz> That's the "new" rogue baseframe we've had sitting around for years
20160310 10:56:00< vultraz> can't be committed until Jetrel finishes the animations
20160310 10:57:41< vultraz> about a year ago I did the defense animations... but he's put off the rest until after the Desert elves
20160310 10:57:45< vultraz> so, expect them in maybe a year
20160310 10:58:43< vultraz> or more
20160310 10:59:01< vultraz> depends
20160310 11:07:15< irker852> wesnoth: Charles Dang wesnoth:master 481bad8539b1 / src/gui/widgets/unit_preview_pane.cpp: tunit_preview_pane: removed newlines between attacks to save some space https://github.com/wesnoth/wesnoth/commit/481bad8539b16f1f5e1e8c81e58a8e78b04532dd
20160310 11:30:24-!- travis-ci [~travis-ci@ec2-54-163-65-175.compute-1.amazonaws.com] has joined #wesnoth-dev
20160310 11:30:25< travis-ci> wesnoth/wesnoth#8820 (master - 4910267 : Charles Dang): The build has errored.
20160310 11:30:25< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/115029236
20160310 11:30:25-!- travis-ci [~travis-ci@ec2-54-163-65-175.compute-1.amazonaws.com] has left #wesnoth-dev []
20160310 11:35:18-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160310 11:39:28< irker852> wesnoth: Charles Dang wesnoth:master cb0ebc01d408 / data/gui/default/widget/unit_preview_pane.cfg: tunit_preview_pane: removed invalid keys in toplevel grid https://github.com/wesnoth/wesnoth/commit/cb0ebc01d4087839e048c1992f1fe9518e598c82
20160310 11:39:31< irker852> wesnoth: Charles Dang wesnoth:master 0584118c7e47 / data/gui/schema.cfg: Updated GUI2 schema for unit preview pane https://github.com/wesnoth/wesnoth/commit/0584118c7e4753a68990ba24d90afdbb9727d483
20160310 11:49:21< vultraz> I'd really like to make it so you can just click on the sprite to get to the help page
20160310 11:54:44-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160310 11:56:51-!- travis-ci [~travis-ci@ec2-54-163-65-175.compute-1.amazonaws.com] has joined #wesnoth-dev
20160310 11:56:52< travis-ci> wesnoth/wesnoth#8821 (master - 481bad8 : Charles Dang): The build is still failing.
20160310 11:56:52< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/115032313
20160310 11:56:52-!- travis-ci [~travis-ci@ec2-54-163-65-175.compute-1.amazonaws.com] has left #wesnoth-dev []
20160310 12:02:03-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160310 12:14:34-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160310 12:15:07< vultraz> zookeeper: proposal to rename "always hits" to "true strike". it's a dota term, but it would go with "first strike"
20160310 12:15:35< zookeeper> what "always hits"?
20160310 12:17:05< vultraz> hm
20160310 12:17:09< vultraz> maybe it was an addon thing
20160310 12:17:15< vultraz> it was in help
20160310 12:17:17< vultraz> now it's not
20160310 12:17:55< zookeeper> indeed
20160310 12:18:31< vultraz> I thought addon abilities didn't show up in the main menu help, though
20160310 12:19:37-!- travis-ci [~travis-ci@ec2-54-163-65-175.compute-1.amazonaws.com] has joined #wesnoth-dev
20160310 12:19:38< travis-ci> wesnoth/wesnoth#8822 (master - 0584118 : Charles Dang): The build passed.
20160310 12:19:38< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/115038082
20160310 12:19:38-!- travis-ci [~travis-ci@ec2-54-163-65-175.compute-1.amazonaws.com] has left #wesnoth-dev []
20160310 12:20:39< vultraz> IT PASSED
20160310 12:20:42< vultraz> IT PASSED
20160310 12:20:44< vultraz> THE BUILD PASSED
20160310 12:43:18-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has joined #wesnoth-dev
20160310 12:55:31-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit []
20160310 13:04:28-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160310 13:05:12< irker852> wesnoth: Charles Dang wesnoth:master 7c239974da5b / / (4 files in 2 dirs): GUI2: added 'help' button definition https://github.com/wesnoth/wesnoth/commit/7c239974da5b7e0839bae7d48e4d644b5a31f22e
20160310 13:05:15< irker852> wesnoth: Charles Dang wesnoth:master 475e87c07677 / data/gui/default/window/unit_recruit.cfg: tunit_recruit: used help button definition for help button https://github.com/wesnoth/wesnoth/commit/475e87c076772ecebe5e00d5ec017471428bd415
20160310 13:05:18< irker852> wesnoth: Charles Dang wesnoth:master 60c7680ba40a / data/gui/default/widget/unit_preview_pane.cfg: tunit_preview_pane: moved unit name under profile button https://github.com/wesnoth/wesnoth/commit/60c7680ba40aefdb5c35cefec2e63d4c9098468d
20160310 13:24:48-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160310 13:30:06< vultraz> zookeeper: opinion: bring name/image back to Recruit details section or not
20160310 13:32:31< zookeeper> bring
20160310 13:35:13< irker852> wesnoth: Charles Dang wesnoth:master dc2a4d05c6ed / data/gui/default/window/unit_create.cfg: tunit_create: make sure listbox fills horizontal space https://github.com/wesnoth/wesnoth/commit/dc2a4d05c6ed988f056018e12c6c69ca28f68079
20160310 13:35:17< irker852> wesnoth: Charles Dang wesnoth:master 7bb10ec2d61f / data/gui/default/window/unit_recruit.cfg: tunit_recruit: use full unit preview pane https://github.com/wesnoth/wesnoth/commit/7bb10ec2d61f330ba8d94926b4e44d3a836edb08
20160310 13:42:18-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has quit [Quit: Leaving]
20160310 14:00:52-!- horrowind [~Icedove@2a02:810a:8380:834:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160310 14:31:33< vultraz> gfgtdf: what does the alpha= key do in units/unit types?
20160310 14:40:50< zookeeper> for ambush, if i had to guess
20160310 14:42:06< zookeeper> it's quite inherently ungreppable, so can't say for sure.
20160310 14:56:52< irker852> wesnoth: Charles Dang wesnoth:master c6aa92d0283c / / (3 files in 2 dirs): taddon_list: initial prototype for new full-window GUI2 addons manager https://github.com/wesnoth/wesnoth/commit/c6aa92d0283c32fe8f2fc78c8104fc2e8af4460c
20160310 14:57:08< vultraz> NOTE ^ that has *not* been deployed yet
20160310 14:57:14< vultraz> you need --new-widgets to look at it
20160310 14:58:53< vultraz> however, I did manage to get the layout working properly
20160310 14:58:57< vultraz> so it just needs functionality
20160310 15:08:09< vultraz> shadowm: do you deem the announcement ready?
20160310 15:25:21-!- horrowind [~Icedove@2a02:810a:8380:834:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160310 15:41:09-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160310 15:43:32-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 244 seconds]
20160310 15:53:37-!- Kwandulin [~Miranda@p200300760F7D734B206067F27FCA773C.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160310 16:11:34-!- Prototyp1X29A [~peter@58.34.17.72] has quit [Quit: Lost terminal]
20160310 16:23:47-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20160310 16:23:53-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20160310 16:25:30< celticminstrel> So, is Travis fixed yet?
20160310 16:42:45-!- Rhonda [~rhonda@anguilla.noreply.org] has quit [Quit: Going down the drain....]
20160310 16:43:01-!- Rhonda [~rhonda@anguilla.noreply.org] has joined #wesnoth-dev
20160310 16:44:10-!- Rh0nda [~rhonda@anguilla.noreply.org] has joined #wesnoth-dev
20160310 16:44:10-!- Rh0nda [~rhonda@anguilla.noreply.org] has quit [Client Quit]
20160310 16:54:32< celticminstrel> vultraz: "Tag [unit_preview_pane] may not be used in [column]" and "Tag [unit_preview_pane_definition] may not be used in [gui]"
20160310 16:54:40< celticminstrel> Did you fix those?
20160310 17:10:55-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160310 17:22:49< Nobun> celticminstrel: a little doubt about string cuncatenation on Lua/Wml code
20160310 17:23:04< celticminstrel> Hmm?
20160310 17:23:14< Nobun> I am assuming that all the part of the cuncatenation (+) should start with _ ... is it right?
20160310 17:23:18< Nobun> example
20160310 17:23:24< celticminstrel> No.
20160310 17:23:42< celticminstrel> You can concatenate strings that are all in different textdomains.
20160310 17:23:47< celticminstrel> I'll show you an example.
20160310 17:23:50< Nobun> ok
20160310 17:24:02< celticminstrel> Or, maybe it's simple enough to just explain.
20160310 17:24:02< Nobun> thank a lot
20160310 17:24:30< celticminstrel> If the + comes at the end of a line, the sequences of concatenated strings continues to the next line.
20160310 17:24:47< celticminstrel> Then you can have another textdomain declaration before the continuation.
20160310 17:25:06< celticminstrel> You see this in saved games a lot, not so much in normal WML.
20160310 17:26:06< Nobun> yep but what I am asking is a bit different
20160310 17:26:13< Nobun> for example
20160310 17:26:23< Nobun> _ "a" + _ "b" + _ "c"
20160310 17:26:56< Nobun> I assume (regardless of textdomain change) that each sequence on cuncatenation must be marked as translatable to be captured as translatable
20160310 17:27:04< celticminstrel> No.
20160310 17:27:21< Nobun> so... is it _ "a" + "b" + "c"?
20160310 17:27:34< celticminstrel> Something like _"some text" + "---divider---" + _"more text" is entirely legal.
20160310 17:27:44< Nobun> yep
20160310 17:27:49< celticminstrel> Each string in the concatenation can be marked or not marked.
20160310 17:28:07< Nobun> yep but
20160310 17:28:16< Nobun> how can I explain my doubt....
20160310 17:28:20< Nobun> figure you have
20160310 17:28:32< Nobun> _ "a" + "b" + _ "c"
20160310 17:28:35< celticminstrel> Ugh, XCode's (or lldb's) refusal to recognize certain symbols in Wesnoth is really starting to get annoying.
20160310 17:29:35< timotei_> http://www.theregister.co.uk/2016/03/09/microsoft_sonic_debian/ :D
20160310 17:29:48< Nobun> currently my script would record only "a" and "c" on the pot file
20160310 17:30:00< Nobun> since the athomized "b" string is not considered translatable
20160310 17:30:03< celticminstrel> That's correct.
20160310 17:30:10< Nobun> ok
20160310 17:30:47< celticminstrel> I believe this technique is used in unit descriptions - b is a string containing only newlines.
20160310 17:34:44< Nobun> loonycyborg: when  you are online I'd like to ask you more details about string cuncatenation bug you experienced... on my little test (sample WML code) seems that all works fine
20160310 17:35:55< loonycyborg> Nobun: Actually your commits already fixed bug that you reported
20160310 17:36:08< loonycyborg> that I reported
20160310 17:36:32< Nobun> ah ok... so it was only the escape quote bug
20160310 17:36:55< Nobun> I thought you experienced also a problem on string cuncatenation... perhaps I misunderstood
20160310 17:36:57< Nobun> thank :)
20160310 17:37:46< loonycyborg> maybe someone else reported that
20160310 17:38:32< loonycyborg> I need to do more proper testing, last bug I found that it failed in some test cfg file
20160310 17:38:42< loonycyborg> in code that was mixing wml and lua
20160310 17:39:40< Nobun> hmm... possible... the language switch theorically works, but wasn't deeply tested
20160310 17:40:14< celticminstrel> ...why the heck is variant::operator= implemented with memcpy?
20160310 17:40:42< celticminstrel> Okay, it's probably not unsafe since this a union of pointers and basic types, but... still...
20160310 17:46:00-!- mjs-de [~mjs-de@x4db6b340.dyn.telefonica.de] has quit [Remote host closed the connection]
20160310 17:50:38-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 17:57:48-!- irker852 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160310 18:01:53< celticminstrel> So, ilua's pretty_print appears to be flawed. :|
20160310 18:03:03-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 240 seconds]
20160310 18:03:07< celticminstrel> ...
20160310 18:04:13< celticminstrel> ...oh, I'm seeing the line wrapped. There's no way to distinguish this from the output.
20160310 18:05:38< celticminstrel> Anyway, ilua appears to not print string keys if there are any integer keys.
20160310 18:05:53< celticminstrel> Or rather, if there are any consecutive integer keys starting from 1.
20160310 18:45:23-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 18:56:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 18:58:59< Nobun> loonycyborg: before testing the mixing lua/wml code be sure you updated to the lastest commit
20160310 18:59:44< Nobun> I did a first test and it seems it works correctly (I didn't test with #textdomain change, but I tryed a sample mixed fake WML file
20160310 19:01:26-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160310 19:02:09-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 19:03:30-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 19:03:40-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 19:03:48< Nobun> this is the test I did: https://transfer.sh/pPIeu/test.tar.gz
20160310 19:04:18< Nobun> the tar.gz contins the test.cfg used for the fake test and the resulting *.po file
20160310 19:04:34< Nobun> contains *
20160310 19:05:04< Nobun> feel free to report any bug :) thank for your time :)
20160310 19:13:54-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has joined #wesnoth-dev
20160310 19:14:01< Elvish_Hunter> Hi all
20160310 19:16:12< celticminstrel> Hi.
20160310 19:20:34-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 19:21:38-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 19:21:56-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 19:21:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: Connection reset by peer]
20160310 19:31:35-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160310 19:36:45-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 19:40:10-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 19:40:12-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 244 seconds]
20160310 19:42:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160310 19:47:58< celticminstrel> Elvish_Hunter: Since you seem to do most of the Python stuff around here, might you be interested in taking a look at my WML syntax highlighter? I'm hoping it can be used on the wiki at some point. (If you're not interested, or unfamiliar with pygments, that's okay. Just asking.)
20160310 19:47:59< celticminstrel> http://celmin.pwcsite.com/wesnoth/wesnoth.py
20160310 19:48:52-!- Crendgrim_ is now known as Crendgrim
20160310 19:50:01< Elvish_Hunter> I didn't even know that there was a library called `pygments`...
20160310 19:50:21< Elvish_Hunter> I'll take a look at it, but I can't promise anything :(
20160310 19:50:37< celticminstrel> Apparently the major MeduaWiki syntax highlighting extension uses it.
20160310 19:50:54< celticminstrel> http://pygments.org
20160310 19:51:40< celticminstrel> I've also got sample output: http://celmin.pwcsite.com/wesnoth/test_syntax.html (From a slightly older revision of the file though)
20160310 20:11:31-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20160310 20:11:37-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20160310 20:16:30-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 20:18:47< celticminstrel> I just discovered a very poorly-written page about variables. https://wiki.wesnoth.org/VariablesWML/How_to_use_variables
20160310 20:27:05< Ravana_> that page is about https://forums.wesnoth.org/viewtopic.php?f=21&t=39218
20160310 20:29:54< celticminstrel> I feel like Dugi's post is wrong...
20160310 20:30:27< celticminstrel> Variable substitutions are applied right-to-left, so I think his second "can't do this" example would actually work.
20160310 20:31:08< celticminstrel> Anyway, the page does have some useful information, but like I said, it's very poorly written.
20160310 20:32:16< celticminstrel> Though at least half of that is grammar, spelling, and wiki-formatting.
20160310 20:34:50 * celticminstrel likes how TextWrangler will open gz files and automatically decompress them for me. (Though after trying to save Wesnoth reported the save was corrupted for some reason.)
20160310 20:36:39-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160310 20:41:30-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 276 seconds]
20160310 20:41:30-!- wedge010 is now known as wedge009
20160310 21:00:12-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 21:00:37-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 21:00:43-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 21:03:34-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 21:04:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 21:05:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 21:05:58-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 21:11:18< loonycyborg> Nobun: error: data/test/scenarios/test_lua_wml.cfg:50: expected closing tag '[/event]' but '[/foo]' found.
20160310 21:15:45< Nobun> I will take a look
20160310 21:17:59< Nobun> the test_lua_wml.cfg is the one on wesnoth 1.12 or 1.13?
20160310 21:19:38< Nobun> loonycyborg: ^
20160310 21:19:51< loonycyborg> from git master
20160310 21:20:37< Nobun> ok
20160310 21:20:46< Elvish_Hunter> Nobun, loonycyborg: that file doesn't even have a textdomain... 
20160310 21:20:47< Nobun> thank... I will investigate
20160310 21:21:13< Nobun> yep but there could be a hidden bug... I have to inspect the file a moment
20160310 21:21:47< celticminstrel> When there's no textdomain declared it's supposed to assume "wesnoth" textdomain.
20160310 21:22:07< Elvish_Hunter> Maybe you should skip files that don't have the textdomain as their first line?
20160310 21:22:34< celticminstrel> Uh, probably not.
20160310 21:22:50< Nobun> nope... as celticminstrel said, when no textdomain found the current domain value is the initialdomain one (by default "wesnoth")
20160310 21:23:50< Elvish_Hunter> In any case, the line containing the opening [foo] is the 48th:
20160310 21:24:01< Nobun> loonycyborg: ok it is not a bug on my script
20160310 21:24:14< Nobun> the error message is correct
20160310 21:24:35< Elvish_Hunter> {ASSERT ([foo]
20160310 21:24:40< Nobun> when you must use wmlxgettext with unbalanced tags on purpose
20160310 21:24:53< Nobun> (for example on macro definition or macro call)
20160310 21:25:14< Nobun> you must use the # wmlxgettext special comment
20160310 21:25:23< Nobun> like in the perl wmlxgettext version
20160310 21:25:43< celticminstrel> Nobun: Sounds like this isn't a case of unbalanced tags.
20160310 21:25:47< Nobun> this can be skipped only if the unbalanced tags are raw balanced
20160310 21:25:55< Nobun> read the file at line 50
20160310 21:26:10< Nobun> you notice that the tag [foo] is unbalanced
20160310 21:26:41< celticminstrel> Elvish_Hunter said to look at line 48.
20160310 21:26:42< Nobun> or perhaps I read bad?
20160310 21:26:49< celticminstrel> I dunno, I haven't seen the file.
20160310 21:27:13< celticminstrel> Are you saying the open tag is missing, or the close tag?
20160310 21:27:23< Nobun> nope I misreaded
20160310 21:27:27< loonycyborg> there is [foo] and  [/foo]
20160310 21:27:28< Nobun> Elvish_Hunter was right
20160310 21:27:39< Nobun> yep... I misreaded the file
20160310 21:27:45< Nobun> I understood the problem
20160310 21:27:51< Nobun> here the problem is this one
20160310 21:28:12< Nobun> the regexp rule that captures the tag will check for \s*[tag]
20160310 21:28:22< Nobun> but the assertion breaks the rule
20160310 21:28:37< Nobun> while the close tag started with \s*[/tag]
20160310 21:28:52< Nobun> this is why it sees only the close tag and not the open tag
20160310 21:29:08< Nobun> now I will try to figure how to solve this bug
20160310 21:30:08< Elvish_Hunter> How does the Perl version behave in this case?
20160310 21:30:37< Nobun> I will try
20160310 21:31:51< celticminstrel> So when you say \s*[tag] do you mean using re.match()?
20160310 21:32:01< celticminstrel> So it's closer to ^\s*[tag]
20160310 21:32:06< Nobun> yep
20160310 21:32:12< Nobun> at the start of the string
20160310 21:32:18< celticminstrel> Makes sense.
20160310 21:32:19< Nobun> at the start of the line
20160310 21:32:39< celticminstrel> But, obviously, breaks down sometimes if tags are included in a macro.
20160310 21:33:45< Nobun> yep
20160310 21:33:49< Nobun> this is what happened
20160310 21:34:38< Nobun> now I'm trying to figure how to do a good regexp rule without introduce bugs to the state sequence system
20160310 21:35:00< loonycyborg> perl wmlxgettext doesnt result in an error from that file
20160310 21:35:01< Nobun> in the meantime I will check what perl wmlxgettext does on that file
20160310 21:35:13< Nobun> thank for test, loonycyborg 
20160310 21:35:24< loonycyborg> yw
20160310 21:35:56< loonycyborg> perl wmlxgettext errors in couple other files though, but on encountering an error it just skips the rest of the file
20160310 21:50:50-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160310 22:02:37< Nobun> yep... it depends of errors
20160310 22:02:49-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has left #wesnoth-dev ["Ciao!"]
20160310 22:03:00< Nobun> if it is a tag error, also perl wmlxgettext stops with an error
20160310 22:36:58-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 22:38:50-!- louis94 [~~louis94@91.178.242.191] has joined #wesnoth-dev
20160310 22:47:15-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20160310 22:50:34< ancestral> Alright, package is being uploaded as I type
20160310 22:50:51< celticminstrel> vultraz is MIA again. :/
20160310 22:51:01< ancestral> That’s okay
20160310 22:51:06< celticminstrel> Did loonycyborg upload a Windows package yet?
20160310 22:51:23< ancestral> I see one on the server
20160310 22:51:29< ancestral> 4 days ago
20160310 22:51:53< ancestral> I’ll have to do the terminal thing to put the stuff on the server later
20160310 22:52:03< ancestral> but a package is on sf.net now
20160310 22:52:04-!- ancestral [~ancestral@63.92.240.233] has quit [Client Quit]
20160310 22:52:09< celticminstrel> Was the announcement made public yet?72/24
20160310 22:52:25< celticminstrel> Gah.
20160310 22:52:37< loonycyborg> celticminstrel: yes, I uploaded it
20160310 22:52:54< celticminstrel> It should've been by now.
20160310 22:53:44< celticminstrel> I think there's supposed to be a warning for not passing a terminal NULL to luaW_getglobal, but it doesn't show up in the XCode build.
20160310 22:57:36-!- Kwandulin [~Miranda@p200300760F7D734B206067F27FCA773C.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160310 22:58:50-!- nurupo is now known as toxcrawler
20160310 23:00:48-!- toxcrawler is now known as nurupo
20160310 23:02:05-!- Nobun [~nobun@host62-49-dynamic.12-79-r.retail.telecomitalia.it] has quit [Quit: Salve a tutti]
20160310 23:02:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 23:04:15-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160310 23:06:21-!- mjs-de [~mjs-de@x4db6b340.dyn.telefonica.de] has joined #wesnoth-dev
20160310 23:24:11-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160310 23:25:23-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 248 seconds]
20160310 23:25:31< vultraz> celticminstrel: yes
20160310 23:25:51< celticminstrel> Hi.
20160310 23:25:56< vultraz> re [03:54:38]	celticminstrel	Did you fix those?
20160310 23:26:04< celticminstrel> Yay.
20160310 23:26:27< celticminstrel> My formula PR is updated with some new stuff and a few tweaks and fixes.
20160310 23:26:48< celticminstrel> Now working on the same thing with the effects PR.
20160310 23:26:56< celticminstrel> (Updating it, that is.)
20160310 23:27:03-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160310 23:27:24< celticminstrel> The formula PR is still failing on Travis though... :|
20160310 23:27:36< vultraz> yay, now you have to deal with travis :)
20160310 23:27:55< celticminstrel> "control reaches end of non-void function"
20160310 23:27:57< celticminstrel> Uh, what. 
20160310 23:28:07< celticminstrel> My compiler issues an error for that.
20160310 23:28:22< vultraz> travis works in mysterious ways
20160310 23:28:24-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160310 23:28:25-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20160310 23:28:29< celticminstrel> Oh, wait, this is in a switch statement.
20160310 23:28:55< celticminstrel> It's switching on an enum.
20160310 23:29:16< vultraz> I wonder why switch() only works on ints/implicit ints
20160310 23:30:05< celticminstrel> Every constant is represented, and every switch case returns, but for whatever reason I removed the default case, so I guess it considers the possibility that the function will not return if the enum's value is out of range.
20160310 23:30:23< celticminstrel> Which might be very possible, considering it's set by a cast.
20160310 23:30:36< celticminstrel> vultraz: It only works on ints because it creates a static jump table.
20160310 23:31:08< vultraz> you mean it doesn't work like the WML [switch] tag and perform a == check against each value?
20160310 23:31:09< celticminstrel> Extending it to strings is fairly trivial, but ugly. (It basically amounts to doing manually what the Java compiler does for you.)
20160310 23:31:15< celticminstrel> vultraz: No, it does not.
20160310 23:32:08< celticminstrel> Imagine there is an array of labels called switch_table.
20160310 23:32:34< celticminstrel> A switch statement takes the value of its variable and uses it as an index into that array, jumping to that label.
20160310 23:33:31< vultraz> so it's like table[i] where i is the value of case: ?
20160310 23:36:15< celticminstrel> Pretty much. It's like "goto switch_table[val]" where switch_table is the array of case labels, sorted by their value, and val is the actual numeric value of the variable being switched on.
20160310 23:37:17< vultraz> ah
20160310 23:37:29< vultraz> (for some reason I don't see goto much in actual code)
20160310 23:37:31< celticminstrel> (I'm not entirely sure how default cases are handled; it might just shove that lable into the array at all locations that don't have an explicitly provided label.)
20160310 23:37:44< celticminstrel> Yeah, there are two main not-terrible uses for goto.
20160310 23:38:20< celticminstrel> One is when you're in a loop that is nested in a loop, and you want to exit both loops at the same time. You set a label just after the outer loop, and goto that label. (There are other ways to do this too, of course.)
20160310 23:39:03< celticminstrel> The other is more of a C idiom rather than C++, and amounts to function-internal exception handling - if an operation returns an error, goto a label at the bottom of the function (often, after a return statement) which cleans up and returns an error code or whatever.
20160310 23:39:35< vultraz> Why not just use an if block instead
20160310 23:40:20< celticminstrel> For the second case? Because you might have several different operations that might fail, but the cleanup work is the same each time.
20160310 23:40:35< celticminstrel> In C++ you'd use try/catch instead, generally.
20160310 23:40:36< vultraz> ah
20160310 23:41:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 23:41:21-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 23:43:30-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160310 23:44:54-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 23:45:14-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: Connection reset by peer]
20160310 23:45:28-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160310 23:46:52-!- vultraz changed the topic of #wesnoth-dev to: 1.13.4 announcing within 24 hours if shadowm doesn't get back | Wesnoth Developers Channel | >>> Want to help? Go here: http://r.wesnoth.org/t42911 (and thanks!) <<< | Logs: http://irclogs.wesnoth.org | Bug tracker: http://bugs.wesnoth.org
20160310 23:49:17< vultraz> celticminstrel: why are you doing this formula stuff, anyway
20160310 23:49:25< vultraz> do you have a specific usecase?
20160310 23:49:43< celticminstrel> Yes and no.
20160310 23:50:23< celticminstrel> There is definitely a usecase for the first commit - people writing unit filter formulas have likely been confused by the fact that "moves" is not an existing value.
20160310 23:50:29< celticminstrel> (As an example)
20160310 23:51:04< vultraz> (though I do like what you've done with "name = "('buttons/misc/orb{STATE}.png~RC(magenta>[icon])')")
20160310 23:51:15< shadowm> vultraz: Why are you asking me?
20160310 23:52:30< vultraz> shadowm: are you going to add any more screenshots? is there anything else I should mention? how do I update the frontpage?
20160310 23:52:45< shadowm> I will not add any more screenshots unless someone gives me something to wok with, I said.
20160310 23:53:05< shadowm> loonycyborg can update the frontpage for you or you can ask me to do so when I'm around (like right now).
20160310 23:53:24< shadowm> There probably is a lot more you should mention. AFAICT you just used RELEASE_NOTES.
20160310 23:53:35< shadowm> But it's not my job to tell you what anymore.
20160310 23:53:48< celticminstrel> vultraz: You should go through the changelog and see if anything interesting is there.
20160310 23:54:40< celticminstrel> vultraz: About the formula stuff, the reasoning or usecase situation (whether I have one and what it is) is different for different parts of it.
20160310 23:55:13< celticminstrel> Some things I changed because the current implementation was simply wrong (like associativity of exponentiation).
20160310 23:55:29< celticminstrel> Some things I changed/added because they were simple and seemed like they'd be convenient.
20160310 23:56:02< celticminstrel> I could be convinced to drop some of the new functions - particularly set ops and hypot() seem like they might have limited use.
20160310 23:58:40< shadowm> Also, for future reference, adding my name to the topic does not highlight me.
--- Log closed Fri Mar 11 00:00:17 2016