| lkcl | ghostmansd[m], you realise that the server gets hammered every time it has to construct a page with 1,000 comments? :) | 00:28 |
|---|---|---|
| lkcl | must tell you about the glenmorangie mistake i made in a pub... twice... | 00:28 |
| *** jn <jn!~quassel@user/jn/x-3390946> has quit IRC | 03:31 | |
| *** jn_ <jn_!~quassel@2a0a-a541-9be8-0-20d-b9ff-fe49-15fc.ipv6dyn.netcologne.de> has joined #libre-soc | 03:31 | |
| *** jn_ <jn_!~quassel@2a0a-a541-9be8-0-20d-b9ff-fe49-15fc.ipv6dyn.netcologne.de> has quit IRC | 03:31 | |
| *** jn_ <jn_!~quassel@user/jn/x-3390946> has joined #libre-soc | 03:31 | |
| *** lxo <lxo!~lxo@gateway/tor-sasl/lxo> has quit IRC | 06:53 | |
| *** lxo <lxo!~lxo@gateway/tor-sasl/lxo> has joined #libre-soc | 06:57 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has quit IRC | 08:53 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.57.79> has joined #libre-soc | 08:54 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.57.79> has quit IRC | 09:13 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.57.79> has joined #libre-soc | 09:14 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.57.79> has quit IRC | 09:23 | |
| *** lx0 <lx0!~lxo@gateway/tor-sasl/lxo> has joined #libre-soc | 09:45 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.56.221> has joined #libre-soc | 09:47 | |
| *** lxo <lxo!~lxo@gateway/tor-sasl/lxo> has quit IRC | 09:48 | |
| *** octavius <octavius!~octavius@78-143-205-55.customer.gigaclear.net> has joined #libre-soc | 11:08 | |
| ghostmansd[m] | lkcl, Glenmorangie usually isn't that bad, I tried it several times :-) | 11:17 |
| ghostmansd[m] | I wouldn't say it's in my top though | 11:18 |
| ghostmansd[m] | My favorite is Laphroaig, then go Talisker and Lagavulin | 11:19 |
| ghostmansd[m] | If we talk about blended, I like Chivas | 11:19 |
| ghostmansd[m] | But I'm by no means an expert. I even doubt I'd pass blind test. | 11:20 |
| ghostmansd[m] | Except for Laphroaig perhaps, its taste is quite unique. | 11:20 |
| ghostmansd[m] | Oh, I forgot about Macallan. Really simple to drink, nothing specific, but sweet and kinda like... I don't know, butterscotch? | 11:22 |
| *** octavius <octavius!~octavius@78-143-205-55.customer.gigaclear.net> has quit IRC | 11:38 | |
| lkcl | ohh it wasn't the taste that was the problem, it was that in 1994 it was GBP 10 for a single shot. | 11:39 |
| ghostmansd[m] | Hey, just buy a bottle :-) | 12:08 |
| ghostmansd[m] | In Russia drinks are way overpriced in restaurants too | 12:09 |
| ghostmansd[m] | Sometimes one shot can reach 1/5 of the bottle, or even more | 12:09 |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.56.221> has quit IRC | 12:25 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has joined #libre-soc | 12:25 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has quit IRC | 12:39 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has joined #libre-soc | 12:45 | |
| *** libredev <libredev!libredev@ircforever.org> has quit IRC | 14:52 | |
| *** libredev <libredev!libredev@libredev.ircforever.org> has joined #libre-soc | 15:00 | |
| sadoon[m] | https://gcc.gnu.org/conduct.html | 16:29 |
| sadoon[m] | So it's not as bad as I expected. Seems reasonable compared to other CoCs I've seen recently | 16:29 |
| lkcl | ahh it's sort-of based on the "kind communications" | 17:50 |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has quit IRC | 18:29 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.160.22> has joined #libre-soc | 18:29 | |
| ghostmansd[m] | Anytime I see such stuff, I'm always wondering what's the definition of "kind" :-) | 18:35 |
| ghostmansd[m] | I mean, something which is kind for one person might be not that much for another | 18:36 |
| ghostmansd[m] | But at least that's not about _just_ being diverse and other rather political stuff | 18:37 |
| ghostmansd[m] | In the recent years, this kinda became the primary motto. It's virtually impossible to find an open source developer who concentrates on the code, not political principles. | 18:38 |
| ghostmansd[m] | I mean, I certainly understand parts of this movement. But this is not the main part of what open source is. | 18:39 |
| markos_ | ghostmansd[m], *I* concentrate on the code, I couldn't care less about political correctness, in fact I'm actually of the opinion that the earlier times of the internet with the endless flamewars were actually more honest, people were not afraid to speak up, and guess what, kick/bans were a thing then too | 18:50 |
| markos_ | people did cross the line then too and usually the communities were quick to respond | 18:50 |
| markos_ | in my 24y of Debian I remember quite a few cases where people were kicked out of the project because of that | 18:51 |
| ghostmansd[m] | markos, I cannot agree more. Everything you said, I totally join this. | 18:51 |
| markos_ | unfortunately nowadays, everyone is too trigger-happy to call for political correctness and as a result the majority stay silent -though they actually disagree | 18:52 |
| markos_ | people are actually afraid to speak their opinion right now | 18:52 |
| ghostmansd[m] | Well, now we have main/master wars. Like other issues are solved. | 18:52 |
| markos_ | this one is the most ridiculous of the wars | 18:52 |
| ghostmansd[m] | In some "open" source projects, you now cannot even say a single damn word without fear to hurt someone. | 18:53 |
| markos_ | "master" naming was never about slavery, master in old english means the person who is ready to marry, and in other cases a teacher of an art | 18:53 |
| markos_ | hence, Bachelor -> Master -> Mister (married, no not PhD :) | 18:53 |
| markos_ | just because some people used master/slave naming, doesn't mean it was about that | 18:54 |
| markos_ | anyway | 18:55 |
| markos_ | quite a few people have unsubscribed from the usual discussion mailing lists and are only interested in the strictly technical discussions | 18:56 |
| markos_ | which is a good thing in a way | 18:56 |
| markos_ | but the result is that everything non-technical is decided by other more vocal people | 18:56 |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.160.22> has quit IRC | 19:03 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has joined #libre-soc | 19:03 | |
| *** yambo <yambo!~yambo@069-145-110-003.biz.spectrum.com> has quit IRC | 19:25 | |
| *** yambo <yambo!~yambo@069-145-110-003.biz.spectrum.com> has joined #libre-soc | 19:38 | |
| *** ghostmansd <ghostmansd!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has joined #libre-soc | 19:46 | |
| ghostmansd | checking differences between setup.py, setup.cfg, pyproject.toml... this is insane | 20:01 |
| ghostmansd | why would Python developers do it to themselves? | 20:01 |
| ghostmansd | s/it/this/ | 20:01 |
| ghostmansd | Damn. Is anyone experienced in this crap? | 20:17 |
| ghostmansd | What I want to do is install mdis module from https://git.libre-soc.org/?p=mdis.git;a=summary in openpower-isa setup.py. | 20:17 |
| ghostmansd | I tried following the way cprop does it, but failed. The best URL I tried is "git+https://git.libre-soc.org/git/mdis.git#egg=mdis", but it creates UNKNOWN-0.0.0-py3.7.egg. | 20:19 |
| ghostmansd | Apparently pyproject.toml and setup.py miss something, but what? | 20:19 |
| programmerjake | copy the code I used for cached_property... (don't forget the two different spots, one is used by pip the other by setup.py install/develop | 20:19 |
| ghostmansd | I'm really drowning in information about pyproject.toml, setup.cfg and setup.py | 20:20 |
| ghostmansd | Yes I have two different spots. | 20:20 |
| ghostmansd | But when I try to use a fixed version, I get this: | 20:20 |
| programmerjake | does setup.py ignore pyproject.toml? | 20:20 |
| ghostmansd | It's just setup() call | 20:22 |
| programmerjake | cached-property uses everything in setup.py, try copying that | 20:22 |
| programmerjake | instead of in pyproject.toml | 20:22 |
| programmerjake | https://git.libre-soc.org/?p=cached-property.git;a=blob;f=setup.py;h=8fc7658d245284e820692affe15e06a806ea3b95;hb=HEAD | 20:22 |
| ghostmansd | I had an impression they switched to pyproject.toml, and setup.py is obsolete | 20:23 |
| ghostmansd | What's strange, when I do this: `pip3 install git+https://git.libre-soc.org/git/mdis.git#egg=mdis` | 20:23 |
| ghostmansd | Everything works | 20:23 |
| ghostmansd | It just doesn't work with openpower-isa setup.py | 20:23 |
| programmerjake | it is, but since luke doesn't like pip we have to use setup.py | 20:23 |
| ghostmansd | 1 sec, let me post what I have so far | 20:23 |
| programmerjake | it may work in setup,cfg too | 20:24 |
| ghostmansd | diff for openpower-isa: https://pastebin.com/HKj7aqND | 20:25 |
| programmerjake | you forgot to specify the git branch/tag | 20:28 |
| programmerjake | where cached_property has @1.5.2 you need @master | 20:28 |
| ghostmansd | why pip doesn't need it? | 20:29 |
| ghostmansd | I haven't used it since pip is fine w/o it | 20:29 |
| programmerjake | cuz pip isn't setuptools? idk | 20:30 |
| ghostmansd | adding @master doesn't fix UNKNOWN egg | 20:31 |
| ghostmansd | OK this seems I need to repeat setup.py from cprop | 20:32 |
| programmerjake | i guess pip calls setup.py to get the list of deps, then pip installs them itself, then runs setup.py install | 20:32 |
| ghostmansd | so much copy-paste | 20:32 |
| programmerjake | yeah, or setup.cfg since setuptools actually reads that | 20:32 |
| ghostmansd | ah well, even worse | 20:33 |
| ghostmansd | error: The 'mdis@ git+https://git.libre-soc.org/git/mdis.git@master#egg=mdis' distribution was not found and is required by libresoc-openpower-is | 20:33 |
| ghostmansd | error: The 'mdis@ git+https://git.libre-soc.org/git/mdis.git@master#egg=mdis-0.5.1' distribution was not found and is required by libresoc-openpower-isa | 20:33 |
| programmerjake | setup.cfg is probably easiest since the format is nearly identical | 20:33 |
| ghostmansd | damn this is fricking madness | 20:33 |
| programmerjake | yeah...cargo is waay easier | 20:34 |
| programmerjake | maybe you do actually need the mdis version at the end? not just #egg=mdis | 20:36 |
| ghostmansd | I tried it | 20:36 |
| ghostmansd | same result | 20:37 |
| ghostmansd | again, pip works with #egg, without it, with #egg and specific version | 20:37 |
| programmerjake | hmm, lemme try locally... | 20:37 |
| ghostmansd | fuck it, let me just import toml module which parses pyporject.toml and passes it to setup.py | 20:39 |
| ghostmansd | (I mean setup function there) | 20:39 |
| ghostmansd | with open("pyproject.toml", "r", encoding="UTF-8") as stream: | 20:41 |
| ghostmansd | pyproject = toml.load(stream) | 20:42 |
| ghostmansd | setup(**pyproject["project"]) | 20:42 |
| ghostmansd | on the first glance this works :-) | 20:42 |
| programmerjake | imho you really should use setup.cfg instead of that mess | 20:42 |
| ghostmansd | I'll support this one later, right now I just want to move on | 20:47 |
| ghostmansd | If this format can replace both setup.py and pyproject.toml, I'm fine with that | 20:47 |
| ghostmansd | I just had an impression they advocate pyproject.toml, though | 20:47 |
| ghostmansd | Fuck, before today I thought that C build systems suck | 20:47 |
| ghostmansd | Now I see they're not even close | 20:47 |
| programmerjake | they do, but because they don't expect anyone to install using setup.py install instead of pip | 20:48 |
| ghostmansd | it seems we always choose unorthodox solutions, eh? | 20:49 |
| programmerjake | well, if you were doing this the same way as some C build systems you'd just have instructions in a readme somewhere to manually git clone and build/install all deps | 20:49 |
| ghostmansd | found yet another beast which "solves" everything: https://python-poetry.org/ | 20:49 |
| ghostmansd | yes, and with the way it's handled in Python I have tons of instructions which contradict to each other and advocate different options :-) | 20:50 |
| ghostmansd | careful, done a force push in mdis | 20:51 |
| ghostmansd | should break nothing since it's not used yet, but that was a mandatory disclaimer | 20:51 |
| programmerjake | i got git dependencies to work: | 21:05 |
| programmerjake | prj1/setup.py: | 21:06 |
| programmerjake | from setuptools import setup | 21:06 |
| programmerjake | setup(name="prj1", version='0.1.0') | 21:06 |
| programmerjake | prj2/setup.py: | 21:06 |
| programmerjake | from setuptools import setup | 21:06 |
| programmerjake | prj1="git+ssh://localhost:10022/data/data/com.termux/files/home/prj1@master#egg=prj1-0.1.0" | 21:06 |
| programmerjake | setup( | 21:06 |
| programmerjake | name="prj2", | 21:07 |
| programmerjake | version='0.1.0', | 21:07 |
| programmerjake | install_requires=["prj1@"+prj1], | 21:07 |
| programmerjake | dependency_links=[prj1], | 21:07 |
| programmerjake | ) | 21:07 |
| programmerjake | modify git path as appropriate... | 21:07 |
| ghostmansd | Jacob, want a fun story? | 21:08 |
| ghostmansd | Here's an excerpt from setup.py logs | 21:08 |
| ghostmansd | Writing /tmp/easy_install-zety6evm/mdis.git/setup.cfg | 21:08 |
| ghostmansd | This is damn insane, really | 21:08 |
| programmerjake | well, it's python, what did you expect :) | 21:10 |
| ghostmansd | some hours since I met Python build system, and I'm already in the haters camp | 21:11 |
| programmerjake | makes me wish we had decided to use rust...cargo does a much better job of dependency management afaict | 21:11 |
| programmerjake | i get the impression git dependencies are barely supported by setuptools | 21:13 |
| ghostmansd | then there'd likely be only several people to develop it :-) | 21:13 |
| programmerjake | another reason not to parse toml and give that to setup(): toml isn't guaranteed to exist, tomllib was only added to python stdlib in python 3.11 | 21:22 |
| ghostmansd | I don't use tomllib | 21:28 |
| ghostmansd | but I agree this is disgusting hack | 21:30 |
| programmerjake | yes, but toml (which you do use) is an external dependency that setup.py install can't install for you if needed | 21:30 |
| programmerjake | since you try to import it before running setup() | 21:30 |
| ghostmansd | Jacob, I have no objections if you want to switch it to setup.cfg now :-) | 21:31 |
| programmerjake | k, i may do that later | 21:31 |
| ghostmansd | for now I simply pick something that works, I really want to complete this task one day | 21:32 |
| ghostmansd | Jacob, could you, please, check, that openpower-isa@mdis setup.py works for you? | 21:34 |
| ghostmansd | I'm afraid to break something | 21:34 |
| ghostmansd | setup.py + import mdis | 21:36 |
| ghostmansd | Should be sufficient | 21:36 |
| programmerjake | yeah, sec... | 21:36 |
| programmerjake | seems to work for me when using setup.py develop | 21:39 |
| programmerjake | pip install -e . seems to work too | 21:41 |
| programmerjake | i didn't test with missing toml though | 21:42 |
| ghostmansd | thank you! | 21:45 |
| ghostmansd | missing toml should never happen: I set it as dependency for mdis | 21:45 |
| programmerjake | it should happen when installing via setup.py install, since it tries to import toml before it ever gets to dependency resolution | 21:46 |
| *** ghostmansd <ghostmansd!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has quit IRC | 21:52 | |
| lkcl | i love the MAINery of the english language that people have today. MAINing of good communication is so important. | 21:53 |
| ghostmansd[m] | lol | 21:54 |
| lkcl | ghostmansd[m], "install mdis module from https://git.libre-soc.org/?p=mdis.git;a=summary in openpower-isa setup.py." no please don't do that | 21:54 |
| ghostmansd[m] | That's why it is in branch :-) | 21:55 |
| ghostmansd[m] | I'm more than willing to know the alternative. Below are several pages of my butthurt caused by python packages. | 21:55 |
| ghostmansd[m] | So yes, anything sane is absolutely welcome. | 21:56 |
| lkcl | put it into here https://git.libre-soc.org/?p=dev-env-setup.git;a=blob;f=hdl-dev-repos;hb=HEAD | 21:56 |
| lkcl | you already have write-permission | 21:56 |
| programmerjake | ah, yeah, mdis is expected to change so should be installed with setup.py develop | 21:57 |
| ghostmansd[m] | Should I use `pip3 install mdis`, `pip3 install -e /mdis`, `python3 setup.py` (beware, that's hacky!)? | 21:57 |
| lkcl | absolutely not. | 21:57 |
| lkcl | under no circumstances use pip3 | 21:58 |
| programmerjake | cached_property won't change so a git dependency is fine | 21:58 |
| lkcl | it is an external resource and we absolutely must not depend on external resources | 21:58 |
| ghostmansd[m] | Ok, I simply chose the stuff I knew we took from git | 21:58 |
| lkcl | debian is borderline (and pragmatic) | 21:58 |
| lkcl | pypi is so unstable and easily hackable they are barely keeping it under control | 21:59 |
| ghostmansd[m] | Ok, no pip, noted | 21:59 |
| ghostmansd[m] | Is there any way I could install pyproject.toml? | 21:59 |
| ghostmansd[m] | Just look at setup.py and you'll see why I ask :-) | 21:59 |
| lkcl | not without a large budget to cover the massive conversion that would be required | 21:59 |
| ghostmansd[m] | No, I mean this project | 22:00 |
| ghostmansd[m] | Not the whole openpower-isa | 22:00 |
| lkcl | feel free if you prefer it but i honestly wouldn't bother. | 22:00 |
| ghostmansd[m] | Do you mean it's OK to keep as is, just move to dev scripts? | 22:00 |
| lkcl | if you've done it and it works don't waste time | 22:01 |
| ghostmansd[m] | There's another issue here, there's toml dependency | 22:01 |
| lkcl | urr | 22:01 |
| ghostmansd[m] | Because I started with pyproject.toml | 22:01 |
| ghostmansd[m] | But I can just move the crap into setup.py I think | 22:01 |
| programmerjake | for openpower-isa.git, pyproject.toml only has pytest config, the rest is in setup.py | 22:02 |
| lkcl | which is precisely f*****g why guido van rossum did *not* make setup.py critically depend on any external dependency | 22:02 |
| * lkcl sigh | 22:02 | |
| ghostmansd[m] | When I started digging into python build systems, everyone recommends pyproject.toml | 22:02 |
| lkcl | that's because they're f*****g idiots who have no thought for consequences. | 22:02 |
| programmerjake | yes, because everyone but us use pip | 22:02 |
| lkcl | if "import toml" was part of the *standard python library* it would be perfectly fine | 22:03 |
| ghostmansd[m] | Folks, let's not start pip discussion | 22:03 |
| ghostmansd[m] | Yeah it isn't | 22:03 |
| programmerjake | tomllib is in 3.11 | 22:03 |
| ghostmansd[m] | So I think I should either get rid of pyproject.toml | 22:03 |
| lkcl | but making absolutely every single install depend on an *independ...*... and now you know why | 22:03 |
| ghostmansd[m] | Or perhaps there's some magic command to do the same as setup.py does but with toml? | 22:04 |
| lkcl | but under no circumstances are we going to drop everything and "convert to python 3.11" as the baseline mininum version | 22:04 |
| lkcl | yes just put it into a dictionary. | 22:04 |
| ghostmansd[m] | I know about `python3 -m build` | 22:04 |
| ghostmansd[m] | Is it standard? | 22:04 |
| lkcl | use "setup.py" that has been used for 25+ years | 22:04 |
| programmerjake | yeah, i know we're not converting to 3.11 | 22:04 |
| ghostmansd[m] | Ok, let's use goddamn dict :-D | 22:04 |
| ghostmansd[m] | Fuck so tired of this build system journey | 22:05 |
| programmerjake | yeah, it's a mess... | 22:05 |
| ghostmansd[m] | Several hours down the drain | 22:05 |
| lkcl | there is standard online documentation on how to do this. | 22:05 |
| lkcl | using setup.py | 22:05 |
| lkcl | and setuptools | 22:05 |
| lkcl | copy the style from any of the setup.py that i have done | 22:06 |
| programmerjake | sorry, i should have told you to not use a git dependency earlier | 22:06 |
| lkcl | openpower-isa, soc, nmutil, | 22:06 |
| ghostmansd[m] | I'll simply put everything from toml | 22:06 |
| lkcl | but leave out the dependencies because you don't have any | 22:06 |
| ghostmansd[m] | Into a dict | 22:06 |
| ghostmansd[m] | Nope, haven't had until this evening | 22:06 |
| lkcl | ghostmansd[m], this will do | 22:07 |
| lkcl | https://git.libre-soc.org/?p=nmutil.git;a=blob;f=setup.py;hb=HEAD | 22:07 |
| lkcl | sorry i meant a list | 22:07 |
| ghostmansd[m] | Yep | 22:07 |
| ghostmansd[m] | I meant arguments in setup | 22:07 |
| lkcl | yes that's it | 22:07 |
| ghostmansd[m] | You see, I populated them from dict :-) | 22:07 |
| ghostmansd[m] | Totally disgusting | 22:07 |
| lkcl | aawwwesome :) | 22:08 |
| lkcl | setup (**kwargs) is an awesome way to do it :) | 22:08 |
| ghostmansd[m] | Yeah, but a dependency :-( | 22:08 |
| ghostmansd[m] | So I'll put them directly now | 22:08 |
| ghostmansd[m] | You see, these kwargs came from toml | 22:09 |
| lkcl | you can still do that in setup.py - just not with a dumb toml file | 22:09 |
| lkcl | indeed. | 22:09 |
| ghostmansd[m] | Yeah | 22:09 |
| ghostmansd[m] | That's the idea | 22:09 |
| ghostmansd[m] | Ok, willdo | 22:09 |
| ghostmansd[m] | BTW, I renamed the instance argument to node | 22:09 |
| ghostmansd[m] | I guess it's the most accurate definition | 22:09 |
| lkcl | hey y'know if you feel like writing a braindead-simple toml reader in setup.py i mean feel free :) | 22:10 |
| ghostmansd[m] | Nope | 22:10 |
| lkcl | probably for the best eh? :) | 22:10 |
| ghostmansd[m] | TBH, I almost hate this task already :-D | 22:10 |
| ghostmansd[m] | And want to finish it | 22:10 |
| ghostmansd[m] | I like the code which we get | 22:10 |
| ghostmansd[m] | But so fucking tired | 22:10 |
| programmerjake | just temporarily replace setup(**args) with pprint(args) and you can copy the output into setup.py | 22:10 |
| ghostmansd[m] | Oh, nice | 22:11 |
| ghostmansd[m] | Thanks Jacob! | 22:11 |
| ghostmansd[m] | Yeah this is ass-kicking | 22:11 |
| ghostmansd[m] | Moar hacks needed! | 22:11 |
| programmerjake | you need from pprint import pprint | 22:11 |
| lkcl | it was worth getting it right, but we need to stop designing and start "using-then-rethinking" | 22:12 |
| lkcl | moving on to the "third iteration" of "plan-to-do-three-iterations-because-if-you-don't-you-will-anyway" | 22:13 |
| ghostmansd[m] | Printed arguments as json with indent=4, nice | 22:13 |
| programmerjake | well, technically pprint is a fancy repr rather than json | 22:15 |
| ghostmansd[m] | Yep, I just don't like its indents | 22:16 |
| ghostmansd[m] | Perhaps they have an option to do it better | 22:16 |
| lkcl | i really like pprint | 22:18 |
| programmerjake | well, most good editors have autoindent when you type tab, can just manually adjust the opening/closing {} | 22:18 |
| programmerjake | plus iirc pprint has an indent-depth option | 22:19 |
| lkcl | yes it does | 22:19 |
| ghostmansd[m] | I edit with ed | 22:19 |
| * lkcl snorts | 22:19 | |
| lkcl | what's wrong with ex?? | 22:20 |
| programmerjake | https://xkcd.com/378/ | 22:20 |
| ghostmansd[m] | I'm old-fashioned | 22:20 |
| ghostmansd[m] | ex is too recent | 22:20 |
| ghostmansd[m] | :-D | 22:20 |
| ghostmansd[m] | Nice cartoon | 22:20 |
| lkcl | programmerjake, lol | 22:22 |
| lkcl | C-x M-c M-butterfly haha | 22:22 |
| ghostmansd[m] | I probably already mentioned my favorite manual on emacs? | 22:24 |
| *** ghostmansd <ghostmansd!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has joined #libre-soc | 22:24 | |
| ghostmansd | http://man.cat-v.org/plan_9/1/emacs | 22:24 |
| programmerjake | lkcl: thx for helping getting 2022-08-051 up and working! | 22:25 |
| ghostmansd | BUGS: Yes. | 22:25 |
| lkcl | programmerjake, yeah took a while | 22:27 |
| ghostmansd | Behold! The simplest setup.py ever: https://git.libre-soc.org/?p=mdis.git;a=blob;f=setup.py;h=85a50a47ea2fd04339d12794340e2affdb0f5768;hb=HEAD | 22:28 |
| ghostmansd | It even seems to work with `python3 -m build`! | 22:29 |
| programmerjake | nice! | 22:30 |
| ghostmansd[m] | Added mdis to hdl-dev-repos | 22:37 |
| ghostmansd | Luke, your "hack-revert" reverted too much | 22:46 |
| ghostmansd | I mean changes in insndb | 22:46 |
| ghostmansd | Please next time don't do it this way, because like it or not, it at least _worked_. | 22:46 |
| ghostmansd | I wasted many hours for this infrastructure for it be "hack-reverted". | 22:47 |
| ghostmansd | I have no idea how to restore it. Fuck. | 22:50 |
| ghostmansd | I know there's path branch, but your hack-revert makes it close to impossible to re-apply parts of these patches. | 22:50 |
| ghostmansd | This is a complete disregard of other's work. | 22:51 |
| ghostmansd | The only good news is that _some_ of these changes are obsolete. But everything else is _bad_ news. | 22:52 |
| ghostmansd | I fixed the original issue, but you still did this. I have to restore these changes manually. | 22:55 |
| ghostmansd | OK, perhaps more of it got obsolete by mdis approach. Yet another good news. | 23:00 |
| ghostmansd | OK, I already managed to switch some commands to new interface. | 23:21 |
| ghostmansd | list, opcodes, pcode and extras work! | 23:21 |
| ghostmansd | In fact, pcode command even already uses dispatcher hook for PCode object | 23:21 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!