|
DELETE CASCADE posted:snowflake and bigquery also pretty good, but in different ways good ways to burn a lot of your employers cash
|
# ? Jul 23, 2022 05:22 |
|
|
# ? Apr 28, 2024 03:29 |
|
nah it's cheap as gently caress bro. if you don't actually have "big data" then the pricing is lmao compared to a live relational db
|
# ? Jul 23, 2022 05:54 |
|
necc0 here really likes mongodb but i dont think ive ever spent the time sitting down designing mongo schema, and that feels.. i dunno, "antithetical" somehow to the way it's advertised.
|
# ? Jul 23, 2022 06:55 |
|
MrQueasy posted:Example != Canonical it literally is in this case. other similar things have other names
|
# ? Jul 23, 2022 07:40 |
|
pokeyman posted:it literally is in this case. other similar things have other names Perhaps I would agree if vanilla markdown via the perl file had ever caught on. However, the pl file has no tests, no examples, no clarifications on why some of the bugs are the way they are. Nowadays, the web mostly use github-markdown, stackoverflow-markdown, or one of the children of php-markdown. It's more of a toy example rather than an exhaustive implementation of the original spec that will never be updated.
|
# ? Jul 23, 2022 08:06 |
|
markdown not only doesn’t have a canonical implementation, it doesn’t even have a canonical spec I am baffled that anyone thinks this is a good thing
|
# ? Jul 23, 2022 10:49 |
|
it means you can nih up your own implementation and nobody's allowed to tell you that you did it wrong
|
# ? Jul 23, 2022 10:55 |
|
I "inherited" (aka last owner got made redundant) an ancient wiki platform at work that uses regexes and xslt to convert markdown into html inside an asp classic wrapper
|
# ? Jul 23, 2022 11:54 |
|
why dont you hire nice peeps of your preferred gender with whips and nipple clamps like a normal person
|
# ? Jul 23, 2022 13:03 |
|
bob dobbs is dead posted:why dont you hire nice peeps of your preferred gender with whips and nipple clamps like a normal person Hmm that would probably meet some diversity goals
|
# ? Jul 23, 2022 13:51 |
|
what is so hard about "the spec is whatever Markdown.pl outputs" I was gonna say the best thing about markdown (and the only good thing I'll say about john gruber) is a refusal to tinker with it. but the other good part is how mad that decision makes people maybe the other other good part is its superficial simplicity turning into nightmarish complexity, but seemingly only after you commit to reimplementing it. good lesson about nih and why proper specs exist
|
# ? Jul 23, 2022 15:37 |
|
GitHub markdown is really good for quickly crapping out docs that you know drat well your stupid PO is never going to read but insists on you writing, and having them look professional enough that they won't send them back maybe use mermaid to make some quick block diagram svgs to make it even more impressive looking during the first and only glance they will ever get
|
# ? Jul 23, 2022 16:12 |
|
pokeyman posted:maybe the other other good part is its superficial simplicity turning into nightmarish complexity, but seemingly only after you commit to reimplementing it. good lesson about nih and why proper specs exist
|
# ? Jul 23, 2022 16:22 |
|
yaml is an interesting case because there has always been a proper spec, the spec was just already nightmarishly complex, and instead of dealing with it many people just chose not to implement it
|
# ? Jul 23, 2022 17:49 |
|
CommonMark has a spec AND tests. It was nice to stumble on when I was looking for a Markdown library to use.
|
# ? Jul 23, 2022 19:50 |
|
Powerful Two-Hander posted:regexes and xslt to convert markdown into html i just how do you come up with that in the first place
|
# ? Jul 23, 2022 23:54 |
|
redleader posted:
it was a rip of something called openwiki circa 2005 I think. All the pages are stored as text columns in an SQL database with some hierarchy system I can't be bothered to work out. I'm in the process of killing it by locking off entire sections and sticking horrible 'this will be removed banners into every page via xslt injection
|
# ? Jul 24, 2022 00:02 |
|
To be fair, regexes and stuff is basically how the original Markdown Perl script worked.
|
# ? Jul 24, 2022 00:12 |
|
to think that a perl script is the reference implementation of something as widespread as markdown
|
# ? Jul 24, 2022 06:13 |
|
I wrote perl in order to add spamassassin modules and even I cannot completely understand what I wrote unless I write copious amounts of comments
|
# ? Jul 24, 2022 06:15 |
|
sb hermit posted:I wrote perl in order to add spamassassin modules and even I cannot completely understand what I wrote unless I write copious amounts of comments the perl interpreter is a utility that translates line noise into book stores
|
# ? Jul 24, 2022 06:22 |
|
love to work at a place where people don't have a solid grasp of the difference between "configuration" and "string constants"
|
# ? Jul 24, 2022 23:32 |
|
Sapozhnik posted:love to work at a place where people don't have a solid grasp of the difference between "configuration" and "string constants" lol one of my tasks next week is I gotta write up the architecture principal for config management across the division
|
# ? Jul 24, 2022 23:42 |
|
configuration is all the stuff that differs between deployment environments not "let's make this string configurable at runtime just in case"
|
# ? Jul 24, 2022 23:44 |
|
Sapozhnik posted:configuration is all the stuff that differs between deployment environments oh sorry I read that as people using string constants as configuration which I mean...does happen.
|
# ? Jul 24, 2022 23:48 |
|
Sapozhnik posted:configuration is all the stuff that differs between deployment environments My predecessor at my current job did this all over the place. I'm guessing they had a vague idea that magic values were bad, but didn't know how one is supposed to avoid them. This resulted in a really long configuration file full of values like the name of the XML element we expect in some API responses and various other values that will break everything if they're ever changed so they don't match values in database views, vendor software, etc. And of course, the importance of those values being what they are is not documented. I keep having to add comments that say stuff like, "It doesn't matter what this value is, as long as it matches these other places (...)"
|
# ? Jul 25, 2022 01:20 |
|
Sapozhnik posted:configuration is all the stuff that differs between deployment environments i had a senior guy up in my pull request asking me why I didn't make an environment variable name configurable
|
# ? Jul 25, 2022 06:43 |
|
No Pants posted:i had a senior guy up in my pull request asking me why I didn't make an environment variable name configurable Should have made it configurable via environment variable
|
# ? Jul 25, 2022 06:44 |
|
Powerful Two-Hander posted:oh sorry I read that as people using string constants as configuration it can be a decent choice if you go full greybeard on it. quote:Customisation so it's still basically a configuration file, it just happens to be in C header format it's not great but i've definitely seen xml and yaml files that were way less readable than this: C code:
|
# ? Jul 25, 2022 08:36 |
|
i think I've realized why I don't like "reduce" a lot of the time it's used - it's often doing some combination of other functions all at once (e.g. filter and group by) in a way that is harder to understand just by looking at the top level of the code
|
# ? Jul 25, 2022 13:56 |
|
NihilCredo posted:it can be a decent choice if you go full greybeard on it. Throwing that in a header will duplicate the constants across translation units. Swapping the static for extern and splitting declaration from definition while hiding definition behind a define you set prior to inclusion in one of your TUs will condense the space used.
|
# ? Jul 25, 2022 14:38 |
|
leper khan posted:Throwing that in a header will duplicate the constants across translation units. Swapping the static for extern and splitting declaration from definition while hiding definition behind a define you set prior to inclusion in one of your TUs will condense the space used. fanciful cutting edge technology such as *check notes* "parsing text files at runtime" will also do that, and you don't even need to recompile!
|
# ? Jul 25, 2022 15:26 |
|
Zlodo posted:fanciful cutting edge technology such as *check notes* "parsing text files at runtime" will also do that, and you don't even need to recompile! dwm is like 2000 lines of c. i don't think the authors are fan of such scandalous bloat as configuration files. heck, even the "plugin system" is just a collection of diffs to apply. quote:dwm has no Lua integration, no 9P support, no shell-based configuration, no remote control, and comes without any additional tools, such as for printing the selection or warping the mouse.
|
# ? Jul 25, 2022 15:53 |
|
I just read up on StandardJS being a thing, i.e. a fairly randomly chosen set of rules to format JavaScript. The best thing is to deliberately ignore it just to wind up the authors. I love how the tooling just breaks on almost anything, completely defeating any goal of improving productivity.
|
# ? Jul 25, 2022 17:01 |
|
so, one guy just decided that they know the best way to format javascript and is now finding the best way to force their personal style on the biggest number of people possible... which is the wet dream of many egomaniacs. the writing style certainly doesn't help their case I wonder how it runs on minifiers
|
# ? Jul 25, 2022 17:43 |
|
Frankly, all I ask of devs is to stick to four space indents and otherwise follow the coding style that is already in the source file. Given that we write a lot of python, the indent thing is never a problem
|
# ? Jul 25, 2022 17:46 |
|
gofmt is the only serious innovation of go. its a good and productive innovation tho. just get peeps to shut the gently caress up about this sorta thing upfront as a lang
|
# ? Jul 25, 2022 18:03 |
|
Zlodo posted:fanciful cutting edge technology such as *check notes* "parsing text files at runtime" will also do that, and you don't even need to recompile! I'm trying to minimize the computation costs on the deployed hardware to maximize my billable hours. This suggestion is harmful to both of these goals.
|
# ? Jul 25, 2022 18:09 |
|
MrMoo posted:I just read up on StandardJS being a thing, i.e. a fairly randomly chosen set of rules to format JavaScript. The best thing is to deliberately ignore it just to wind up the authors. I love how the tooling just breaks on almost anything, completely defeating any goal of improving productivity. they say "no .eslintrc" to manage but j clicked on a random faq question and they describe how to add some config to your package.json
|
# ? Jul 25, 2022 19:03 |
|
|
# ? Apr 28, 2024 03:29 |
|
they decided to go with a bunch of dumb rear end rules like no semicolons which means that this “standard” style is just obnoxious if you’re used to idiomatic js
|
# ? Jul 25, 2022 19:49 |