|
Visions of Valerie posted:
what's your issue with this? you can indent `- absolutely` however many times you want, and it is equivalent. the only weird part of this snippet is that the mapping value must be indented by >1 blank space character (normally >0 would be fine), but this is likely an artifact of you nesting a block mapping which is distinct from a block sequence and thus has different rules for indentation.
|
# ? Mar 31, 2024 21:08 |
|
|
# ? Jun 7, 2024 02:41 |
|
12 rats tied together posted:what's your issue with this? you can indent `- absolutely` however many times you want, and it is equivalent. I think semantically relevant whitespace is always a disaster waiting to happen
|
# ? Mar 31, 2024 22:53 |
|
i am pretty ok with significant whitespace. especially having newlines replace semicolons as a kind of default, which was always a thing (e.g. cpp) and has gotten more popular. but i am not really a fan of it in yaml. but, i am tbh pretty relaxed overall. as far as i am concerned if things are xml, json, yaml, or ini derivatives like toml, i am largely fine with it. csv gets a special "ok" status too not as a standard thing but as basically simple enough that you'll figure out the system-specific limits as you go. more bespoke than that needs very deep motivation though.
|
# ? Mar 31, 2024 23:08 |
|
significant whitespace is great for files that (1) fit in one screen, and (2) will be read and written by hand if (2) isn't true, just emit json instead of yaml if (1) isn't true, either split it into smaller files or figure out how not to do (2)
|
# ? Apr 1, 2024 06:54 |
|
akadajet posted:and while we’re at it stop inventing new programming languages, ui frameworks and orms. we have enough exactly, we already have the perfect language in the Lisp-1 known as prefix Dylan (which is just Scheme rebuilt atop a CLOS-style object system), just use that with an adapted CLIM and ORM
|
# ? Apr 1, 2024 08:53 |
|
Visions of Valerie posted:I think semantically relevant whitespace is always a disaster waiting to happen yeah its extremely bad
|
# ? Apr 1, 2024 14:29 |
|
Cybernetic Vermin posted:having newlines replace semicolons as a kind of default this is the worst idea of all, probably
|
# ? Apr 1, 2024 14:29 |
|
Shaggar posted:this is the worst idea of all, probably at least its one thing, unlike erlang where theres a bunch of different ways to terminate a line depending on context that could be trivially inferred
|
# ? Apr 1, 2024 15:06 |
|
if you can't read a line break as a substitute closing brace or bracket you can just use one of the yaml flow styles which require them. unfortunately examples in the wild are rare, and many tools will not emit them, because most people dont have this problem
|
# ? Apr 1, 2024 15:09 |
|
most people who use yaml are brain damaged
|
# ? Apr 1, 2024 15:14 |
|
https://ruudvanasseldonk.com/2023/01/11/the-yaml-document-from-hell I don’t think there’s a good config language tbh but if there is yaml doesn’t seem to be it
|
# ? Apr 1, 2024 15:24 |
|
I mean we’re all adults here, we can all agree that the right (semi) human readable data language to use is the one that’s already in use in the code base, or that meshes the best with whatever languages it’s used by, and that they’re all fundamentally fungible, and that nothing beats binary TLVs
|
# ? Apr 1, 2024 15:26 |
|
Surprise T Rex posted:https://ruudvanasseldonk.com/2023/01/11/the-yaml-document-from-hell this article is based on the yaml 1.1 spec which is from 2005.
|
# ? Apr 1, 2024 15:55 |
|
pyyaml also ships with a schema + loader that loads the entire document without issuePython code:
|
# ? Apr 1, 2024 16:01 |
|
Surprise T Rex posted:https://ruudvanasseldonk.com/2023/01/11/the-yaml-document-from-hell there are atleast 2 good config languages:
|
# ? Apr 1, 2024 16:04 |
|
leper khan posted:at least its one thing, unlike erlang where theres a bunch of different ways to terminate a line depending on context that could be trivially inferred This is a relic of an old line of language design that used both comma, semicolons, and full stops as terminators, depending on what you were terminating (with the full stops usually for top level declarations like functions). I think Erlang is the most popular language that still works like that. It should be preserved as a living monument, like monarchy.
|
# ? Apr 1, 2024 18:32 |
|
Shaggar posted:there are atleast 2 good config languages: plists are just xml so once again apple is good
|
# ? Apr 1, 2024 18:36 |
|
DELETE CASCADE posted:plists are just xml so once again apple is good ok but when someone says "xml is fine you just need to use a better IDE" i think of the plist interface in xcode and wonder what the hell an actually good XML interface would look like because it sure as poo poo isnt that
|
# ? Apr 1, 2024 19:37 |
|
Shaggar posted:most people who use yaml are brain damaged never by choice. but you're right
|
# ? Apr 1, 2024 19:45 |
|
abraham linksys posted:ok but when someone says "xml is fine you just need to use a better IDE" i think of the plist interface in xcode and wonder what the hell an actually good XML interface would look like because it sure as poo poo isnt that idk I like the gambling high from guessing what’s gonna happen when I press the return key in a text field
|
# ? Apr 1, 2024 20:29 |
|
abraham linksys posted:ok but when someone says "xml is fine you just need to use a better IDE" i think of the plist interface in xcode and wonder what the hell an actually good XML interface would look like because it sure as poo poo isnt that
|
# ? Apr 1, 2024 20:47 |
|
what if you took your xml and when presented it stripped out the end tags and replaced first one with something like 'foo:' and then for nested fields you just indent to make it clear what is part of which tag. you could do attributes as like 'foo(a: "blah", b: 123, c: "hi"):'. if you have a series of things you could prefix the entries with a dash or something
|
# ? Apr 1, 2024 21:44 |
|
Sweeper posted:what if you took your xml and when presented it stripped out the end tags and replaced first one with something like 'foo:' and then for nested fields you just indent to make it clear what is part of which tag. you could do attributes as like 'foo(a: "blah", b: 123, c: "hi"):'. if you have a series of things you could prefix the entries with a dash or something yeah so the thing is: that sucks
|
# ? Apr 1, 2024 22:06 |
|
Sweeper posted:what if you took your xml and when presented it stripped out the end tags and replaced first one with something like 'foo:' and then for nested fields you just indent to make it clear what is part of which tag. you could do attributes as like 'foo(a: "blah", b: 123, c: "hi"):'. if you have a series of things you could prefix the entries with a dash or something why would you do that? just use it as is
|
# ? Apr 1, 2024 22:26 |
|
things must change
|
# ? Apr 1, 2024 22:52 |
|
sounds like you’ve got an xyml problem
|
# ? Apr 1, 2024 23:38 |
|
Sagacity posted:xcode isn't a good anything so im not surprised it doesn't work for xml either when steve jobs said that apple thinks different he was talking about plists
|
# ? Apr 2, 2024 01:14 |
|
Apple should license the IDE work out to Jetbrains and have them make Xcode.
|
# ? Apr 2, 2024 05:07 |
|
swing and a miss
|
# ? Apr 2, 2024 05:15 |
|
simply do not configure your software. just write your software. need to save user preferences? we have a place for that. it’s called hkcu
|
# ? Apr 2, 2024 05:19 |
|
Bloody posted:simply do not configure your software. just write your software. need to save user preferences? we have a place for that. it’s called hkcu
|
# ? Apr 2, 2024 05:29 |
|
need to save user preferences? no, you don't if their preferences were really that strong they could patch the program and recompile it
|
# ? Apr 2, 2024 06:06 |
|
that implies that you provide source code which suggests it may be open source which is as we all know “bad” see also rotors manifesto . therefore use the registry
|
# ? Apr 2, 2024 06:08 |
|
Internet Janitor posted:need to save user preferences?
|
# ? Apr 2, 2024 06:09 |
|
Internet Janitor posted:need to save user preferences?
|
# ? Apr 2, 2024 06:30 |
|
Internet Janitor posted:need to save user preferences?
|
# ? Apr 2, 2024 07:29 |
|
Internet Janitor posted:need to save user preferences? Bloody posted:that implies that you provide source code which suggests it may be open source which is as we all know “bad” see also rotors manifesto . therefore use the registry if they care enough they can binary patch it. no need for recompiling, providing source, or using the registry.
|
# ? Apr 2, 2024 11:58 |
|
Do not cater to users who do not have the entire build infrastructure on their system (which is exactly a C89 compiler and nothing more).
|
# ? Apr 2, 2024 12:02 |
|
I see you also use suckless applications
|
# ? Apr 2, 2024 13:19 |
|
|
# ? Jun 7, 2024 02:41 |
|
You beat me to ithttps://st.suckless.org posted:Configuration https://dwm.suckless.org posted:- dwm is customized through editing its source code, which makes it extremely fast and secure - it does not process any input data which isn't known at compile time, except window titles and status text read from the root window's name. You don't have to learn Lua/sh/ruby or some weird configuration file format (like X resource files), beside C, to customize it for your needs: you only have to learn C (at least in order to edit the header file). The elite of... people who willingly choose C as their config file format.
|
# ? Apr 2, 2024 13:32 |