Log for #openttdcoop.devzone on 8th December 2013:
Times are UTC Toggle Colours
01:06:09  *** Supercheese has joined #openttdcoop.devzone
05:04:41  <Brot6> xUSSR Set.xls XWowanX @
09:49:33  *** ODM has joined #openttdcoop.devzone
09:57:45  *** Alberth has joined #openttdcoop.devzone
09:58:23  *** ODM has quit IRC
10:46:03  *** ODM has joined #openttdcoop.devzone
11:06:04  *** oskari89 has joined #openttdcoop.devzone
11:08:45  *** frosch123 has joined #openttdcoop.devzone
11:08:55  <Alberth> moin
11:11:09  <frosch123> hai
11:11:26  <frosch123> eints rejected xussr upload yesterday evening
11:11:37  <frosch123> i didn't find an error message yet though
11:14:16  <frosch123> hmm, do i need to adjust rights.dat btw?
11:16:48  <frosch123> hmm, it worked sometimes though
11:17:03  <Alberth> I didn't add the subdir to the @protect line, so I don't think you need to
11:17:23  <frosch123> i noticed that devzone was very slow yesterday, something took high cpu load, maybe it just ran into some timeout
11:17:50  <frosch123> let's see whether it just works now agani :p
11:18:38  <Alberth> I hit the filesize limit of 100k
11:19:25  <frosch123> 1M here
11:19:51  <frosch123> but anyway, the uploiad worked 5 minutes before it failed
11:20:05  <frosch123> so, it rather looks like some devzone issue with busy cpus or somethign
11:20:12  <planetmaker> hm
11:20:55  <frosch123> yup, it just worked now
11:21:03  <frosch123> yesterday it said "broken pipe" during the upload
11:21:36  <frosch123> and the ssh connection had like 0.5s latency or so
11:21:38  <Alberth> it works here too
11:26:33  <planetmaker> I don't see anything obvious...
11:26:52  <frosch123> well, now everything is fine again :)
11:27:14  <Brot6> Bug #6632 (Rejected): Exception in repo2eints XeintsX @
11:27:14  <Brot6> Bug #6632 (Rejected): Exception in repo2eints XfroschX @
11:27:31  <frosch123> eints log files do not contain anything suspicious, so i think the isse was caused by some other vm
11:27:59  <frosch123> 21:10 it worked
11:28:11  <frosch123> 21:20 and 21:30 it failed
11:28:32  <frosch123> (both utc)
11:30:56  <Brot6> Bug #6636 (New): Exception in repo2eints XeintsX @
11:31:10  <frosch123> wtf
11:32:02  <DevZone> Project xussrset - Trains from Russia build #118-push: SUCCESS in 2 min 34 sec:
11:32:12  <frosch123> now it failed again :p
11:32:18  <frosch123> same non-descriptive error
11:32:23  <frosch123> while it worked 5 minutes before
11:32:29  <Alberth> :(
11:34:02  <Alberth> broken pipe  usually means that where-ever you want to sent it to, doesn't exist (any more) or has closed its input
11:34:25  <frosch123> yeah, but i am sending to localhost :p
11:35:19  <Brot6> Revision 1221:3153e84b1cca: Code Review #6631: Пересмотрен код сборки ДР1А XGeorgeX @
11:35:55  <frosch123> at least eints has some cpu load now
11:36:48  <Alberth> cat foo | more    will also report broken pipe if you quit 'more' on a long input file
11:37:01  <Alberth> yeah, 1195 strings :p
11:46:27  *** gelignite has joined #openttdcoop.devzone
12:18:16  <Alberth> bleh, these string modifications really suck :(
12:31:10  *** Supercheese has quit IRC
12:31:46  *** Supercheese has joined #openttdcoop.devzone
12:41:08  <frosch123> where the heck is the logging output in
12:41:47  <frosch123>     MEMFILE_MAX = 102400 <- oh, found a number in
12:47:05  <Alberth> :o
12:53:55  <frosch123> should we just increase it and hope?
12:56:43  <Alberth> if it's bytes, the xussr lang file wouldn't fit in it
12:56:46  <planetmaker> let's try :)
12:56:56  <Alberth> so it makes you wonder what the unit is of it
12:57:33  <Alberth> the experiment is easy enough :)
12:58:36  <frosch123> assign it the same langfile limit as in the config?
12:59:18  <planetmaker> hm, maybe?
12:59:36  <frosch123> well, just changed the number locally for testing
13:02:21  <frosch123> let's see whether it will happen again :p
13:02:45  <Brot6> Bug #6636 (Closed): Exception in repo2eints XeintsX @
13:02:45  <Brot6> Bug #6636 (Closed): Exception in repo2eints XfroschX @
13:05:07  <frosch123> body = BytesIO() if maxread < self.MEMFILE_MAX else TemporaryFile(mode='w+b') <- maybe the TemporaryFile fails or so
13:09:38  *** ODM has quit IRC
13:10:30  *** tycoondemon has quit IRC
13:10:44  *** tycoondemon has joined #openttdcoop.devzone
13:13:35  *** ODM has joined #openttdcoop.devzone
13:22:16  *** ODM has quit IRC
13:22:35  *** ODM has joined #openttdcoop.devzone
14:29:51  <Brot6> Code Review #6637 (Assigned): Запретить недодизельпоезда XGeorgeX @
14:31:34  <Alberth> George:  STR_CAN_ATTACH_NO_AFTER_MIDDLE_UNIT  has a typo in the english string "tot" instead of "not"
14:33:57  <George> Alberth: thanks fixed
15:25:25  <DevZone> Project Canadian Railway Set build #1-push: SUCCESS in 26 sec:
15:29:32  <Brot6> Revision 55:2797b138ddd7: Codechange: Use templates for CanadianNational graphics (templated by S... XplanetmakerX @
15:29:32  <Brot6> Revision 54:b7cf80c4bd1d: Change: [Makefile] Update to make-nml 0.4.5+ XplanetmakerX @
16:25:02  <George> What is the right way to get the year vehicle at position was build?
16:25:41  <George> checking var C4 would be enough?
16:26:44  <George> or it has value 00 for 1920 and does not allow to check years before 1920 and after 2175?
16:48:45  <DevZone> Project Iron Horse build #498-push: SUCCESS in 1 min 13 sec:
16:50:04  <DevZone> Project xussrset - Trains from Russia build #119-push: SUCCESS in 2 min 44 sec:
16:53:43  <DevZone> Project FIRS Industry Replacement Set build #54-push: SUCCESS in 4 min 57 sec:
17:02:37  <Brot6> Revision 3868:98b664d20524: Update: Translations XeintsX @
17:04:33  <Brot6> Revision 463:a9e05065993f: Update: Translations XeintsX @
17:10:41  <Brot6> Revision 24:a22a79db9759: Update: Translations XeintsX @
17:14:01  <Brot6> Revision 1222:27977680e442: Update: Translations XeintsX @
17:14:08  *** yorick has joined #openttdcoop.devzone
17:18:40  <DevZone> Project NML - NewGRF Meta Language build #168-nightlies: SUCCESS in 1 min 16 sec:
18:06:12  <DevZone> Project finnishtrams build #50-nightlies: SUCCESS in 48 sec:
18:21:42  <DevZone> Project Japanese Buildings build #43-nightlies: SUCCESS in 17 sec:
18:32:22  <DevZone> Project Finnish Rail Infrastructure - Rails build #156-nightlies: SUCCESS in 6 min 58 sec:
18:47:55  <DevZone> Project Iron Horse build #499-nightlies: SUCCESS in 1 min 26 sec:
18:47:56  <Alberth> planetmaker: may I request a rebuild of the eints documentation? I added GS stuff there last thursday
18:52:00  <planetmaker> sure
18:53:22  <DevZone> Project eints build #11-nightlies: SUCCESS in 31 sec:
18:53:40  <Alberth> thanks
18:54:36  <planetmaker> can you see and start a build now even yourself?
18:55:04  <planetmaker> though... it's supposed to build daily...
18:55:47  <Alberth> LATEST is messed up
18:56:05  <Alberth> it points to 5008
18:56:06  <planetmaker> oh?
18:56:09  <planetmaker> ah
18:56:33  <Alberth> but 5089 also seems (re)build
18:56:53  <planetmaker> 5089 should be the one which was actually built, not 'LATEST'.
18:59:48  <planetmaker> should be fixed now, too. And hopefully update next time correctly as well
19:00:01  <planetmaker> thanks for pointing that out :)
19:01:35  * Alberth saves the url
19:01:56  <planetmaker> it's a beautiful documentation
19:02:11  <Alberth> blame the python people :p
19:03:45  <Alberth> I think it's important to explain these things, unfortunately, it's aging fast :)
19:04:26  <planetmaker> I like your way to explain cases :)
19:11:07  <Alberth>
19:11:46  <Alberth> it does create the problem of what to show when displaying the base language though
19:12:32  <Alberth> right now it displays the untranslated strings, which may thus be different from the strings in translate context
19:22:13  <planetmaker> I don't quite understand... isn't the base language defined unambiguously?
19:25:37  <planetmaker> command translation as in RAW_STRING -> STRING and cases?
19:25:39  <planetmaker> ah yes
19:26:46  <planetmaker> Alberth, I don't think it's a exactly a problem to show the base language untranslated
19:27:24  <planetmaker> At least OpenTTD translators also have to know that. And if the commands are just copied an appropriate error message could be given
19:27:26  <Alberth> I translate it in the above patch queue
19:27:42  <Alberth> for the string edit page
19:27:55  <planetmaker> yeah.
19:28:13  <planetmaker> actual reading gave me that conclusion, too ;)
19:28:14  <Alberth> so as translator you see {STRING}   instead of {RAW_STRING}
19:28:43  <planetmaker> ah, that's what 3 does
19:29:44  <Alberth> with one small issue, if we go beyond GS, namely that it has no case support for {RAW_STRING}, and thus some {STRING} has case support and some other has not then
19:30:26  <Brot6> Revision 1223:4ddc4c00a986: Code Review #6631: Пересмотрен код сборки всего... XGeorgeX @
19:30:26  <Brot6> Revision 1224:5eff466ccf43: Слияние XGeorgeX @
19:30:49  <DevZone> Project xussrset - Trains from Russia build #120-push: SUCCESS in 2 min 33 sec:
19:32:08  <planetmaker> hm, that might be hard to explain then
19:32:40  <Alberth> yeah, but it's not in GS, since it doesn't support cases :)
19:32:46  <DevZone> Project Iron Horse build #500-push: SUCCESS in 1 min 11 sec:
19:32:59  *** George has quit IRC
19:32:59  <Alberth> but I think we need a new string-name or so
19:34:20  <planetmaker> you mean in strgen?
19:35:21  <Alberth> that might do the trick
19:35:28  <planetmaker> maybe it would be nice if strgen simply accepted RAW_STRING in translations, too
19:35:49  <planetmaker> and on compilation just replaced it by what {STRING} does now
19:35:50  <Alberth> in eints it's going to be messy perhaps
19:36:24  <Alberth> I am not convinced {RAW_STRING} should exist
19:36:53  <Alberth> it sounds like an implementation detail of the program which is exposed to the strings
19:36:59  *** oskari89 has quit IRC
19:37:08  <Alberth> as are {STRINGn} variants
19:37:54  <planetmaker> not sure that {RAW_STRING} can be replaced by {STRING} or {STRINGx}. Maybe
19:38:12  <planetmaker> would make it look nicer for translators for sure
19:38:56  <Brot6> Revision 464:53693caeda73: Fix: metro multiple units show correct graphics for second unit (close... XandythenorthX @
19:38:56  <Brot6> Revision 465:e5efbd6d6d61: merge XandythenorthX @
19:38:56  <Brot6> Bug #6626 (Closed): Rear car not showing in Metro Consists XandythenorthX @
19:39:54  <Alberth> to me, it seems that the program must provide a string one way or another. The string itself should not dictate how the program does that
19:40:27  <planetmaker> yeah, true
19:40:41  <Alberth> but instead, we have a printf-like notation that tells the program how to provide the string
19:43:28  <Alberth> in freerct, I created a data structure to abstract from that
19:43:29  <Webster> Title: language.h - freerct - Build roller coasters, get rich - Google Project Hosting (at
19:45:02  <Alberth> which is almost the same that we have in OpenTTD, except here the data structure resolves how to derive the string
19:46:23  <planetmaker> that allows arbitrary nesting, yes?
19:56:24  <Alberth> currently it doesn't
19:57:15  <Alberth> I briefly considered it, and see two directions. One is to provide a pointer to another StringParameters for a supplied StringID
19:57:56  <Alberth> another directions is to use some of the 16 parameters for the sub-strings
19:58:25  <Alberth> that is easier, but needs a nice scheme to decide what parameter goes where
19:59:51  <Alberth> you could even encode that in the "num" parameter
20:00:29  <Alberth> in short, it looks feasible, so I stopped considering it until I actually need it :)
20:02:28  <planetmaker> I think OpenTTD uses the 2nd approach
20:04:29  <frosch123> how would you abstract STRING vs RAW_STRING?
20:04:45  <frosch123> the latter won't support cases, no matter what
20:04:55  <frosch123> ofc you could ignore cases, if they are given
20:05:22  <frosch123> anyway, imo display only STRING to the translator in all cases
20:05:48  <frosch123> and just strip the cases on saving the string
20:06:35  <frosch123> or keep RAW_STRING to express "no case", and only replace STRINGn with STRING
20:08:01  <frosch123> or we teach strgen to ignore cases for RAW_STRING
20:08:08  <frosch123> while allowing translators to enter them
20:08:28  <Alberth> will give bug reports :)
20:08:36  <frosch123> but most likely a translator wouldn't enter a case for a RAW_STRING anyway
20:08:51  <frosch123> i would claim that RAW_STRING is used where cases make no sense
20:10:00  <frosch123> hmm, RAW_STRING is mostly used for names, maybe some languages would try to apply cases to names
20:10:03  <frosch123> like latin :p
20:10:48  <Alberth> it doesn't play in GS currently
20:11:06  <Alberth> just showing {STRING} everywhere will do the trick
20:15:23  <Alberth> perhaps eventually, you need to add a translation layer from the input to eints internals, as well as a translation from eints back to the output language files
20:15:53  <Alberth> that should provide enough room to make changes
20:16:05  <planetmaker> frosch123, German also has cases for names
20:16:19  <planetmaker> frosch sagt, was froschs Meinung ist ;)
20:16:20  <Alberth> at that point it may however be useful to save the strings in parsed form :)
20:30:18  <Brot6> Code Review #6638 (New): Увеличение срока службы ДР1Б XWowanX @
20:37:31  <frosch123> <- alternative explanation of "cases" in english
20:39:37  <Alberth> persion   line 261
20:39:51  <planetmaker> even nicer, frosch123 :)
20:40:06  <frosch123> <- improved spelling
20:41:28  <Alberth> indeed, very good examples
20:44:48  *** ODM has quit IRC
20:58:16  <Brot6> Revision 360:b537a0641ddc: Doc: Different description of 'cases' in the manual (inspired my accus... XfroschX @
21:04:52  <Alberth> pushed my patch queue too
21:07:48  <frosch123> updated the server
21:08:05  <frosch123> anything left to do?
21:13:18  <Alberth> base language is still displaying the {RAW_STRING}, which is fine I think, but there should be a warning at the top of the page about it
21:13:53  <planetmaker> probably good solution
21:14:44  <Alberth> and features #6611 make sense, not as new view, but as view instead of the current translation overview page, imho
21:14:59  <Alberth>
21:15:01  <frosch123> what about the fancy javascript that copies the baselang into the translation?
21:15:21  <frosch123> resp. the placeholder text
21:15:40  <planetmaker> yes, not as new view but as modification of the existing
21:16:00  <planetmaker> which placeholder text, frosch123 ?
21:16:24  <Alberth>  is also a valid request imho, and easy to implement as a check-box to safe-guard against accidental empty translation save
21:16:36  <frosch123> if the translation entry of a case is empty, it displays the default case in grey
21:16:59  <frosch123> but, yeah, that is actually not affected by the RAW_STRING
21:17:05  <frosch123> only the copy-paste thingie
21:17:34  <Alberth> depends on where it pulls the source text from
21:20:21  <Alberth> oh, the default translated case is no problem of course, that has {STRING}
21:23:32  <Brot6> Revision 363:1148d10fbbe3: Feature: In the string edit page, display the base string with transla... XAlberthX @
21:23:32  <Brot6> Revision 361:e30d9f48cbba: Codechange: Save ParameterInfo of the positional commands instead of t... XAlberthX @
21:23:32  <Brot6> Revision 362:510c830d7d82: Feature: Translate the string commands while checking correctness. XAlberthX @
21:29:34  <Brot6> Feature Request #6639 (New): Add warning about possibly changed strings in the base language display XAlberthX @
21:30:13  <Brot6> Bug #6640 (New): Variotram color Xjuzza1X @
21:40:13  <Alberth> good night
21:40:32  *** Alberth has left #openttdcoop.devzone
21:54:24  *** frosch123 has quit IRC
22:07:43  <Brot6> Bug #6640 (New): Color variations Xjuzza1X @
22:07:43  <Brot6> Bug #6640: Color variations Xjuzza1X @
22:31:21  *** gelignite has quit IRC

Powered by YARRSTE version: svn-trunk