|
pram posted:*generates code in a useless language please don't talk about elm like that
|
# ? Aug 1, 2017 22:41 |
|
|
# ? May 25, 2024 20:47 |
|
lancemantis posted:idk plenty of ORMs will generate a bunch of (garbage) objects from schema descriptions jOOQ owns because it just generates types and it does it's best ot just map your stuff so that it can be used like a real (typesafe) query, instead of trying to make everything a loving object. code:
|
# ? Aug 1, 2017 22:41 |
|
NihilCredo posted:big difference between generating types from tables and generating types from queries orms can generate types from queries too. mybatis does it better tho where it just maps statements to queries and input and output to types.
|
# ? Aug 1, 2017 22:43 |
|
Jooq seems tempting but I have an aversion to cutesy chaining tricks like those
|
# ? Aug 1, 2017 22:47 |
|
Shaggar posted:orms are the worst. they are bad. you know what else is bad? stored procedures.
|
# ? Aug 1, 2017 22:51 |
|
Star War Sex Parrot posted:I don’t know what any of this means i don't either but they're all very happy about it!!
|
# ? Aug 1, 2017 22:51 |
|
Sapozhnik posted:Jooq seems tempting but I have an aversion to cutesy chaining tricks like those idk i use it at work and it owns. handles all the little things like using fully qualified names (for example "schema"."table" in postgres). plus its nice to know what poo poo will break at compile time when your schema changes during development
|
# ? Aug 1, 2017 22:51 |
|
also it lets you define bindings so columns that are "timestamp with time zone" automatically become ZonedDateTime and "timestamps" are exposed as LocalDateTime. you can define custom type bindings for anything, based on the type or even based on identifier regex.
|
# ? Aug 1, 2017 22:53 |
akadajet posted:they are bad. my friend's former job used psql procedures for etl
|
|
# ? Aug 1, 2017 22:57 |
re: sql tool chat. dbeaver felt a bit fucky, so i switched to sql workbench (the lovely looking java thing), and i couldn't be happier (i mean, unless job would get me a datagrip licence)
|
|
# ? Aug 1, 2017 22:58 |
|
i use pgcli
|
# ? Aug 1, 2017 23:08 |
|
Pgcli is alright for basic poo poo. Hell if I'm writing a big query with that though
|
# ? Aug 1, 2017 23:23 |
|
Right, that's when I break out vi
|
# ? Aug 1, 2017 23:28 |
|
CRIP EATIN BREAD posted:i use pgcli I'm the unhashed passwords
|
# ? Aug 1, 2017 23:33 |
CRIP EATIN BREAD posted:i use pgcli Sapozhnik posted:Right, that's when I break out vi I'm younger than 40
|
|
# ? Aug 1, 2017 23:41 |
|
Pls don't vishame
|
# ? Aug 1, 2017 23:51 |
|
we did it https://twitter.com/LuigiThirty/status/892499134447587337
|
# ? Aug 1, 2017 23:57 |
|
CRIP EATIN BREAD posted:jOOQ owns because it just generates types and it does it's best ot just map your stuff so that it can be used like a real (typesafe) query, instead of trying to make everything a loving object. But how do I do SQL injections now??
|
# ? Aug 2, 2017 00:00 |
|
akadajet posted:they are bad. lol no. stored procs are the best. not using stored procs is another sign of a bad developer along with use of p-langs or love of javascript.
|
# ? Aug 2, 2017 00:16 |
|
cinci zoo sniper posted:I'm younger than 40 dunno about pgcli, but psql will happily use whatever is in $EDITOR (or $VISUAL i can't remember which) when you use \e. it's good.
|
# ? Aug 2, 2017 00:31 |
http://www.sql-workbench.net/ sql workbench link now that im at computer btwjony neuemonic posted:dunno about pgcli, but psql will happily use whatever is in $EDITOR (or $VISUAL i can't remember which) when you use \e. it's good.
|
|
# ? Aug 2, 2017 05:32 |
also whoever mentioned the r plugin for idea, holy gently caress it's absolute garbage and 100% does not work on windows, with any version and the dev hasn't checked in for weeks, good loving job!!
|
|
# ? Aug 2, 2017 08:12 |
|
cinci zoo sniper posted:http://www.sql-workbench.net/ sql workbench link now that im at computer btw dbeaver is the best psql tool i've found so far, will be checking out sqlworkbench now, thanks small annoyance: why do java apps do that dumb "please google and download jdbc database drivers on your own"? i've run into that behaviour a few times now. why can't they just include them with the app, they're like 1MB each or less.
|
# ? Aug 2, 2017 09:02 |
|
Imma guess licensing
|
# ? Aug 2, 2017 09:14 |
|
NihilCredo posted:small annoyance: why do java apps do that dumb "please google and download jdbc database drivers on your own"? i've run into that behaviour a few times now. why can't they just include them with the app, they're like 1MB each or less. they're probably not 1mb each, the oracle connection adapters are like 600mb or some bullshit
|
# ? Aug 2, 2017 11:05 |
|
what is the actual proper way to force IE to load in edge mode? I keep seeing 'set the doctype to html' and yet IE still ignores it if the user has set the browser to put the entire intranet domain in compatibility mode for some stupid reason there are docs that say "use x-ua-compatible" then others that say it's deprecated and to use doctype. also stored procedures are good and Dapper is good entity framework is garbage
|
# ? Aug 2, 2017 13:07 |
|
MononcQc posted:yeah the key to keeping it working you need to ensure a good overall distribution. Fully randomized treaps (as they are originally intended) don't guarantee O(log n) but mathematically come pretty close to it no matter what. The question would be if there is such a metric or metric modification that could keep good overall tree balance while being able to bubble entries up (say, +1% to the weight when being picked, but -0.01% when being on the path but not selected), but also whether it would be faster still than just maintaining 3 distinct data structures for each use case. Trip report: modified my little treap structure to allow a priority to rise to the top, using 100,000 entries and all random values 0 > N > 1 for weights. I then updated random items to have a top priority (0.9999) without any jitter, giving higher chances for unbalancing in there. The following tests were run:
That stuff is pretty neat. I haven't done tests with dynamic updates (where I assume memory shuffling is likely to lose a bunch of perf), but if you have some form of statistical analysis on the frequency of access (say, words of the english language), you're likely gonna get something pretty close to an optimal tree. MononcQc fucked around with this message at 14:51 on Aug 2, 2017 |
# ? Aug 2, 2017 14:49 |
|
Powerful Two-Hander posted:what is the actual proper way to force IE to load in edge mode? I keep seeing 'set the doctype to html' and yet IE still ignores it if the user has set the browser to put the entire intranet domain in compatibility mode for some stupid reason intranet sites always load in compatibility mode by default. using the <meta http-equiv="X-UA-Compatible" content="IE=edge" /> tag should take care of that. you can also set it as an http header so you don't have to do it for every application on a web server. you can also probably change it via gpo if all of your intranet sites work fine in edge mode.
|
# ? Aug 2, 2017 15:01 |
|
Shaggar posted:intranet sites always load in compatibility mode by default. using the <meta http-equiv="X-UA-Compatible" content="IE=edge" /> tag should take care of that. you can also set it as an http header so you don't have to do it for every application on a web server. you can also probably change it via gpo if all of your intranet sites work fine in edge mode. yeah we used to have that auto injected into the headers but some fucker seems to have disabled it in other news: gently caress css and gently caress bootstrap
|
# ? Aug 2, 2017 15:25 |
|
cinci zoo sniper posted:what i like about sql workbench is that i can configure it to auto-rollback dbeaver has this too and its almost always on for me. how is auto completion in sql workbench? thats one thing i really like in dbeaver
|
# ? Aug 2, 2017 15:44 |
necrotic posted:dbeaver has this too and its almost always on for me. how is auto completion in sql workbench? thats one thing i really like in dbeaver dbeaver has auto commit, but i wasn't able to find auto rollback. nothing to complain about autocompletion in workbench
|
|
# ? Aug 2, 2017 16:10 |
|
NihilCredo posted:small annoyance: why do java apps do that dumb "please google and download jdbc database drivers on your own"? i've run into that behaviour a few times now. why can't they just include them with the app, they're like 1MB each or less. im guessing its because there's always different versions of the connectors and sometimes you want to use a different one? also different connectors work with different java/jdbc versions so there's that too.
|
# ? Aug 2, 2017 17:18 |
|
it lets you choose the jdbc bug ensemble that'll perfectly complement your project
|
# ? Aug 2, 2017 17:28 |
|
cinci zoo sniper posted:my friend's former job used psql procedures for etl what's wrong with that?
|
# ? Aug 2, 2017 18:45 |
|
psql is cool and there's poo poo you can do in it that makes things way faster than plain SQL so i dont see what the problem is.
|
# ? Aug 2, 2017 20:45 |
|
Powerful Two-Hander posted:yeah we used to have that auto injected into the headers but some fucker seems to have disabled it bootstrap is fine, less/sass are terrible.
|
# ? Aug 2, 2017 20:55 |
JewKiller 3000 posted:what's wrong with that? apparently there's some kind of ridiculously large overhead compare to using, say, pentaho. don't ask me the details tho
|
|
# ? Aug 2, 2017 20:59 |
|
idk about psql but in tsql you can do bulk direct table operations that are more efficient for large etl jobs
|
# ? Aug 2, 2017 21:01 |
|
cinci zoo sniper posted:also whoever mentioned the r plugin for idea, holy gently caress it's absolute garbage and 100% does not work on windows, with any version rstudio stamping on a human face forever then. i don't understand why it crashes when the R process segfaults.
|
# ? Aug 2, 2017 22:38 |
|
|
# ? May 25, 2024 20:47 |
|
Shaggar posted:less/sass are terrible. shaggar was wrong?
|
# ? Aug 2, 2017 23:12 |