|
HoboMan posted:more importantly can i replace youre posts with a gay baby? Gay! Hahahahaha!
|
# ? Oct 25, 2017 21:25 |
|
|
# ? Jun 1, 2024 19:57 |
|
oh my loving christ TimeZoneInfo has a memory leak?
|
# ? Oct 25, 2017 22:41 |
|
HoboMan posted:oh my loving christ TimeZoneInfo has a memory leak? it opens a file handle every time you call the Local property so I'd say it's very likely
|
# ? Oct 25, 2017 23:03 |
|
CRIP EATIN BREAD posted:it opens a file handle every time you call the Local property so I'd say it's very likely what if i never access that property? i know i can defeat the memory leak error i'm getting by just making a wrapper, but how do i know if what i am doing is safe? e: well it's an error about potentially causing a memory leak because the class is tagged as such HoboMan fucked around with this message at 23:37 on Oct 25, 2017 |
# ? Oct 25, 2017 23:23 |
|
HoboMan posted:more importantly can i replace youre posts with a gay baby? this was not the joke
|
# ? Oct 26, 2017 00:21 |
|
gonadic io posted:I am currently writing angular 1 I'm sorry....
|
# ? Oct 26, 2017 00:40 |
|
HoboMan posted:what if i never access that property? i was joking but its sad that it's an acceptable answer
|
# ? Oct 26, 2017 13:23 |
|
CRIP EATIN BREAD posted:i was joking but its sad that it's an acceptable answer i have confirmed the way TimeZoneInfo looks up time zones is accessing a file (with bonus magic string matching) this is sad, but it's the only game in town i mean there is noda time, but it looks like it's got a lot of concepts it wants you to learn and the dev is like "yeah my framework is hard to use, but it's like that ~by design~ to force devs to really think about what they are doing" HoboMan fucked around with this message at 14:53 on Oct 26, 2017 |
# ? Oct 26, 2017 14:50 |
|
HoboMan posted:i mean there is noda time, but it looks like it's got a lot of concepts it wants you to learn and the dev is like "yeah my framework is hard to use, but it's like that ~by design~ to force devs to really think about what they are doing" this looks like its a .net port of joda-time, which is a very good date/time implementation. if you cant figure it out, this is the thread for you i guess
|
# ? Oct 26, 2017 16:07 |
|
Date/time libs are hard and the joda family is great
|
# ? Oct 26, 2017 16:26 |
|
CRIP EATIN BREAD posted:this is the thread for you i guess i think this is well established the first thing i read about noda time was a SO comment by the author stating the the library is hard to use on purpose so i figured it wasn't worth learning
|
# ? Oct 26, 2017 17:22 |
|
i haven’t used it a ton but iirc noda / joda time is hard to use but also hard to use incorrectly, which is probably better than easy to blow your whole foot off with.
|
# ? Oct 26, 2017 17:34 |
|
LocalDate date = LocalDate.of(2017, 10, 26); LocalTime time = LocalTime.of(11, 30); ZoneId zone = ZoneId.of("America/Detroit"); ZonedDateTime dt = ZonedDateTime.of(date, time, zone); Duration d = Duration.between(dt, ZonedDateTime.now()); d.getSeconds(); guys this code is so incredibly cryptic i have no idea what its doing!
|
# ? Oct 26, 2017 18:24 |
|
CRIP EATIN BREAD posted:LocalDate date = LocalDate.of(2017, 10, 26);
|
# ? Oct 26, 2017 18:29 |
|
today I "solved" a hanging transaction error caused by a gently caress knows how many deep nest of SQL transactions by just putting 5 lines of "commit tran" on the bottom of a script and calling it a day bug closed: "works" for me edit: o gently caress more like Sapozhnik posted:BORN TO ROLLBACK except "Commit em all 2017" Powerful Two-Hander fucked around with this message at 18:42 on Oct 26, 2017 |
# ? Oct 26, 2017 18:40 |
|
WHILE @@TRANCOUNT > 0 COMMIT;
|
# ? Oct 26, 2017 19:59 |
|
Powerful Two-Hander posted:today I "solved" a hanging transaction error caused by a gently caress knows how many deep nest of SQL transactions by just putting 5 lines of "commit tran" on the bottom of a script and calling it a day i fixed this same problem my own terrible way which is to edit every script to be SQL code:
|
# ? Oct 26, 2017 20:41 |
|
Powerful Two-Hander posted:today I "solved" a hanging transaction error caused by a gently caress knows how many deep nest of SQL transactions by just putting 5 lines of "commit tran" on the bottom of a script and calling it a day this is truly terrible, congratulations
|
# ? Oct 26, 2017 20:44 |
|
having read tef's litany against message queues a few months ago i'm finishing up a new revision of our system that has a bunch of services communicating via localhost http instead of rabbitmq there's actual backpressure now and some (though not all) failures propagate all the way to the system's edge and i can now inject events into the middle of the system with curl to test things out
|
# ? Oct 26, 2017 22:44 |
|
|
# ? Oct 26, 2017 23:07 |
|
Sapozhnik posted:having read tef's litany against message queues a few months ago i'm finishing up a new revision of our system that has a bunch of services communicating via localhost http instead of rabbitmq if you want to go the whole experience, reify your service objects and return hypermedia
|
# ? Oct 26, 2017 23:30 |
|
tef posted:reify your service objects and return hypermedia i prefer
|
# ? Oct 26, 2017 23:47 |
|
tef posted:if you want to go the whole experience, reify your service objects and return hypermedia i'm not literally amazon so our rest api is strictly an organizational boundary. there are multiple http services inside the perimeter, some exposed to the outside world, some not, all talk to a single monolithic postgres database. outgrowing that single is strictly in the "nice problem to have" category right now. not really my field of expertise at any rate so i just hope i didn't screw up the design too badly.
|
# ? Oct 27, 2017 00:01 |
|
rjmccall posted:this is truly terrible, congratulations i might actually properly fix it tomorrow but in *theory* the root cause of it shouldn't occur because the team that set up the data that causes the error should have unfucked themselves by then they wont though so maybe ill just remove those lines from the deploy script and go "idk it was fine when i tested it on a test db" edit: i should add this is a one off script not a sproc, im not a total monster....but whoever wrote the underlying sproc it calls that leaves hanging transactions on failure is Powerful Two-Hander fucked around with this message at 00:03 on Oct 27, 2017 |
# ? Oct 27, 2017 00:01 |
|
Sapozhnik posted:having read tef's litany against message queues a few months ago i'm finishing up a new revision of our system that has a bunch of services communicating via localhost http instead of rabbitmq broad question for you and or tef. I have an environment that does long running render jobs, using what I think is a pretty normal queue setup. jobs tend to come in big batches that we don’t have the hardware to handle all at once, so we queue them up in a message queue, and workers grab jobs when they are free. how/why would I migrate this to a setup like yours?
|
# ? Oct 27, 2017 00:03 |
|
a witch posted:broad question for you and or tef. depends? what do you do when renders fail? is retrying worth it?
|
# ? Oct 27, 2017 00:09 |
|
what the gently caress are those classes of frameworks for interacting with databases called Entity Frameworks is a really heavy example of one this is bugging me
|
# ? Oct 27, 2017 00:24 |
|
ORM's?
|
# ? Oct 27, 2017 00:25 |
|
that's the fucker thanks, god that was bugging me (e) is there a local favorite of those for Oracle DBs and C#/.NET, we've got a small app that uses EF and it is so bad to work \with and causing problems so I want to try something else
|
# ? Oct 27, 2017 00:27 |
|
|
# ? Oct 27, 2017 00:27 |
|
tef posted:depends? what do you do when renders fail? is retrying worth it? retrying is worth it within a limited window, depending on external circumstances. basically a batch job finds items of interest, and will queue up a render job for each item that we don’t already have a successful render for. If an item is failed, but still interesting, it will get retried.
|
# ? Oct 27, 2017 00:29 |
|
oh my god so much poo poo to get across to an airgapped network i hate oracle i hate everything
|
# ? Oct 27, 2017 00:32 |
|
Ciaphas posted:that's the fucker Dapper is really loving good for basic stuff like "map results of this proc to a list of objects" and also allows you to do crazy poo poo like splitting to create sub classes, also it handles parameters really nicely for you as well edit: I'm pretty sure I learned about it from a previous version of this thread and it is seriously good. also entity framework blows, stored procs 4 lyfe
|
# ? Oct 27, 2017 00:38 |
|
a witch posted:retrying is worth it within a limited window, depending on external circumstances. so, like, you're already using some database to track state, and use a queue as a smart load balancer i think i explicitly talk about this being a pretty normal thing in the long winded rant the whole 'use http' jerkfest is more 'just do plain old rpc and don't use a message broker as a go-between' for short lived requests that need responses
|
# ? Oct 27, 2017 00:42 |
|
tef posted:so, like, you're already using some database to track state, and use a queue as a smart load balancer oh yeah. all state lives in a database, the messages are just IDs. I missed your initial post that’s why I asked thanks for the response. are there good resources for this sort of stuff?
|
# ? Oct 27, 2017 00:45 |
|
NihilCredo posted:i prefer please make a smiley out of this tia
|
# ? Oct 27, 2017 01:21 |
|
Powerful Two-Hander posted:Dapper is really loving good for basic stuff like "map results of this proc to a list of objects" and also allows you to do crazy poo poo like splitting to create sub classes, also it handles parameters really nicely for you as well idk about stored procs, i'm not a db guy, but being able to put in sql and get out objects is all i want and it sounds like dapper fits the bill so thanks
|
# ? Oct 27, 2017 01:24 |
|
dapper owns, but ef can be fine too if you're not inheriting an existing db and don't mind colouring inside the lines.
|
# ? Oct 27, 2017 02:08 |
|
ef is so bad
|
# ? Oct 27, 2017 02:09 |
|
|
# ? Jun 1, 2024 19:57 |
|
jony neuemonic posted:dapper owns, but ef can be fine too if you're not inheriting an existing db and don't mind colouring inside the lines. that's exactly why EF is poo poo-tier in this app, our DBs date back to the early 90s and are about as corroded as that implies, yes when ef worked, it worked great. when it didn't, cower in fear, mortals
|
# ? Oct 27, 2017 02:09 |