IRC logs for #openttd on OFTC at 2022-06-19
⏴ go to previous day
02:06:28 *** Wormnest has quit IRC (Quit: Leaving)
02:39:28 *** debdog has quit IRC (Ping timeout: 480 seconds)
02:56:10 *** wallabra has quit IRC (Remote host closed the connection)
02:58:16 *** wallabra has joined #openttd
03:12:55 *** Smedles has joined #openttd
05:33:09 *** jpw has quit IRC (Remote host closed the connection)
05:45:42 *** jpw has quit IRC (Remote host closed the connection)
06:12:39 *** Etua has quit IRC (Quit: Etua)
06:35:51 *** smoke_fumus has joined #openttd
06:35:53 *** smoke_fumus has quit IRC ()
06:53:26 *** andythenorth has joined #openttd
07:12:18 *** Etua has quit IRC (Quit: Etua)
07:44:13 *** HerzogDeXtEr has joined #openttd
09:11:58 *** andythenorth has quit IRC (Quit: andythenorth)
09:13:43 *** sla_ro|master has joined #openttd
10:53:17 *** Joel has quit IRC (Ping timeout: 480 seconds)
11:03:33 *** D-HUND is now known as debdog
13:14:15 *** Joel has quit IRC (Ping timeout: 480 seconds)
13:33:58 <Samu> im editing choochoo, tweaking some stuff
13:34:19 <TrueBrain> I am on the toilet, pooping
13:37:24 <Samu> comparing choochoo vs my tweaked choochoo
13:37:33 <Samu> i'm still on the first network
13:37:35 <TrueBrain> I am not comparing my poop, sorry
13:38:13 <Samu> the original only was able to make 2 trains, then switched to two new networks
14:09:02 *** Joel has quit IRC (Ping timeout: 480 seconds)
14:09:03 *** andythenorth has joined #openttd
14:25:32 *** Joel__ has quit IRC (Ping timeout: 480 seconds)
14:35:57 <Samu> still on the first network, that's ~25% of the map, and still going
14:37:46 <Samu> original choochoo on the left
14:38:02 <Samu> tweaked chocho on the right
14:40:34 <Samu> on the left, choochoo failed at expanding the 3 first networks until the 4th time, where it was sticking with it, until it crashed
14:40:55 <Samu> on the right, that's still the first network, fantastic
14:50:32 <Xaroth> I'm more of a peanutbutter person.
14:52:02 <Samu> im testing with breakdowns set to normal
14:53:11 <Samu> seems that the older part of the network is running dry with nearly no trains, the way the ai works, it sells the 10% least profitable and clones the 10% most profitable
14:53:45 <Samu> old trains with breakdowns are suscept to being the least profitable, makes sense
14:54:46 <Samu> problem is, the ai doesn't build newer engines running on the old parts
14:55:27 <Samu> so much room for improvements, I wish the real author would look at these
15:01:13 <andythenorth> if not os.path.exists(generated_files_path):
15:01:13 <andythenorth> os.mkdir(generated_files_path)
15:01:43 <andythenorth> that exists in a python file; with make -j 2, occasionally they collide ^^
15:01:44 <Samu> covers about 50% of the map
15:01:57 <andythenorth> I could try / except and pass
15:02:10 <andythenorth> I could have the makefile make the dir
15:30:03 *** jpw has quit IRC (Ping timeout: 480 seconds)
15:34:10 <glx> typical concurrency issue
15:35:44 <glx> you could skip the if and put mkdir in a try except, I think it's the only way here
15:38:09 <TrueBrain> Nah, what the stack overflow says is the right thing to do .. Well, including the skipping of the if
15:38:27 <TrueBrain> Using try/except here is unneeded complex :)
15:39:02 <glx> ah yes, didn't click the link :)
15:39:06 <TrueBrain> mkdir -p basically, but in Python ;)
15:45:04 *** Wormnest has joined #openttd
15:50:52 <andythenorth> give or take the bugs, multi-grf compile now works
15:58:08 <andythenorth> but $(HTML_DOCS) is a pre-requisite, so I don't understand :P
16:01:41 <andythenorth> works without -j arg to make :P
16:28:43 *** Smedles has joined #openttd
16:33:28 <andythenorth> maybe I need to add copy_docs_to_grf_farm to .PHONY list?
16:38:25 *** Smedles has joined #openttd
16:41:36 <andythenorth> ok I absolutely misunderstand something about make :P
16:41:54 <andythenorth> '1 last thing' :P
16:43:59 <andythenorth> ok I think the problem is bundle_tar: clean tar
16:43:59 <andythenorth> release: bundle_tar copy_docs_to_grf_farm
16:44:39 <andythenorth> with -j 2, the deps for copy_docs_to_grf_farm are satisfied, but meanwhile bundle_tar is causing them to be cleaned
16:46:31 <LordAro> don't mix clean & build deps
16:48:31 <andythenorth> for release only, there was some reason to clean first
16:48:54 <andythenorth> I think it's related to git tags not causing deps to need rebuilt
16:49:00 <andythenorth> but git tags being needed in the nml
16:49:13 <andythenorth> I wondered if I could use .NOTPARALLEL on some recipes, but seems not
16:57:29 <andythenorth> meh, so maybe I need to check if git tag is outdated
16:57:35 *** glx has quit IRC (Read error: Connection reset by peer)
17:10:34 <andythenorth> yeah if I build, then tag, it doesn't regenerate, as expected
17:10:44 <andythenorth> so the grf will have wrong version
17:12:34 <nielsm> if you're okay with potentially undefined behaviour, or rather relying on implementation details, you could maybe make a target depending on a file inside the .git dir
17:13:10 <nielsm> that would also make it annoying to build outside of a git working tree
17:13:59 <glx> a target to create a version file used by other targets
17:14:18 <nielsm> yeah something like that
17:15:17 <glx> and this file can be included in source package without git
17:15:46 <nielsm> like having a 'version_info' file with the tag name in it, a 'version_info: check_version' target, and the check_version target is then marked .PHONY and will regenerate the version_info file
17:17:44 <andythenorth> I need to make it only pick up tags
17:17:51 <andythenorth> otherwise everything will rebuild on every commit
17:19:17 <nielsm> in my idea, the check_version target would only rewrite/touch the version_info file if the version had logically changed
17:20:12 <nielsm> like if it was just on a branch it could fill the file with branch:main' or something and check if the content would change before writing to the file
17:20:43 <andythenorth> I wonder if there's a flag or built in target to force recompiling all prerequisites
17:20:52 <andythenorth> rather than just relying on clean
17:36:23 <andythenorth> nah not what I need
17:39:04 <TrueBrain> sounds what you need is what nielsm is telling you .. PHONY target that always executed, write a new version_info in like version_info.new .. compare the files, if changed, rename, if not, remove. That is how OpenTTD used to do it :)
17:40:41 <Samu> i've been searching for AIs using search in github site
17:40:53 <Samu> i find too many tutorial copies
17:43:48 <LordAro> TrueBrain: targets that always rebuild make me sad
17:44:01 <TrueBrain> the target doesn't "rebuild" :)
17:44:04 <TrueBrain> it is just always triggered
17:44:08 <TrueBrain> like all PHONY targets :P
17:44:21 <TrueBrain> or you consider "all" also "rebuilding" :P
17:44:39 <peter1138> (It's a heap of shit, they've gone for "luxury item" pricing, worse than say Apple...)
17:44:39 <LordAro> nothing better than seeing "nothing to do" message
17:45:00 <andythenorth> looks like I can just dump the output of my git script to a file
17:45:29 <andythenorth> then add it as a pre-requisite for everything
17:45:36 <TrueBrain> LordAro: we can all dream :P
17:50:22 <andythenorth> frigging makefile tabs
17:50:42 <TrueBrain> we all know that pain :P
17:51:21 <andythenorth> ach, now I have to test in shell if a file exists :P
18:01:25 <TrueBrain> if [ -f <filename> ]; then ... fi
18:10:09 *** virtualrandomnumber has joined #openttd
18:10:11 <andythenorth> it's probably wrong :P
18:11:18 *** virtualrandomnumber has quit IRC ()
18:21:37 <andythenorth> ugh this is wrong
18:22:25 *** gelignite has joined #openttd
18:24:13 <andythenorth> might be dummy spit time
18:25:55 <peter1138> That's it, remove all your NewGRFs?
18:31:50 <andythenorth> no just write a shell script for release
18:31:57 <andythenorth> instead of using make
18:32:49 <andythenorth> or have make release call a shell script
18:33:16 <andythenorth> that then calls make clean (without -j) and make real_release (with -j)
18:33:27 <andythenorth> imagine the horror
18:45:22 <DorpsGek> - Update: Translations from eints (by translators)
18:46:55 <andythenorth> it seems to work
18:47:17 <andythenorth> lego brick solutions
19:13:39 *** Flygon has quit IRC (Remote host closed the connection)
19:30:39 *** ChanServ sets mode: +v tokai
19:37:27 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
19:51:32 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
20:08:05 *** gelignite has quit IRC (Quit: Stay safe!)
20:19:00 *** Wormnest has joined #openttd
20:41:02 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:10:30 *** sla_ro|master has quit IRC ()
21:15:39 *** sla_ro|master has joined #openttd
21:18:40 *** andythenorth has quit IRC (Quit: andythenorth)
21:57:10 *** Samu has quit IRC (Quit: Leaving)
22:00:28 *** sla_ro|master has quit IRC ()
22:10:15 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
22:25:54 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:53:00 *** Joel has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵
