|
redleader posted:jq is like "what if xpath but bad" This is like saying "what if black hole but blacker"
|
#
?
Nov 16, 2022 02:33
|
|
|
|
| # ? Apr 24, 2024 03:34 |
|
|
Twerk from Home posted:You ever experience the thrill of handling Norway in yaml? the other day I accidentally passed the special missing value NA_character_ to a 3rd party R package we pay good money for (lmao) and it straight up replied that it couldn't find any data about namibia
|
|
#
?
Nov 16, 2022 02:54
|
|
|
MrMoo posted:Like why would one think it is a good idea to reply to that conversation in public? just letting my ceo incorrectly poo poo talk all my work in public
|
|
#
?
Nov 16, 2022 07:23
|
|
|
Private Speech posted:thinking about it sure it's super dumb depending on what their policies are like on company secrets and social media use and such, but maybe that's how lax it was at twitter someone said twitter had explicit rules allowing their employees to do that, which werent updated when elon took over seems like a pretty good labor case to me, but i live in a developed nation so ymmv
|
|
#
?
Nov 16, 2022 10:57
|
|
|
MrMoo posted:Like why would one think it is a good idea to reply to that conversation in public? not everyone is a bootlicker and these particular people are not gonna be hurting for work. i would not be surprised if they are financially independent. i'm sure some of their colleagues that can't stand up to elon appreciate the gesture
|
|
#
?
Nov 16, 2022 12:25
|
|
|
MrMoo posted:Like why would one think it is a good idea to reply to that conversation in public? Publicly telling horrible bosses to shut the gently caress up makes the world a better place.
|
|
#
?
Nov 16, 2022 15:06
|
|
|
DELETE CASCADE posted:jq is good in json-brain, xpath is good in xml-brain xpath is rad 
|
|
#
?
Nov 16, 2022 15:44
|
|
|
jsonpath is a thing too
|
|
#
?
Nov 16, 2022 16:36
|
|
|
MrMoo posted:jsonpath is a thing too I like how it is defined in terms of XPath and every single example has the XPath equivalent and the JSONPath version is obviously worse.
|
|
#
?
Nov 16, 2022 17:10
|
|
|
pseudorandom name posted:I like how goddamn right you do
|
|
#
?
Nov 16, 2022 17:17
|
|
|
god drat does the Elastic Search documentation suck. I've got two instances with a logger writing to each one, on one the index template has been correctly auto created and has 94 fields, on the other it only has 28 and poo poo doesn't get published properly. why is this? Who knows! API keys on both should be the same but you can't query the permissions an API key has so you just have to guess! Oh but the docs day you can but the query doesn't work! so maybe I can export the mapping from one instance to the other? Nope that fails!
|
|
#
?
Nov 17, 2022 19:02
|
|
|
Use Loki op
|
|
#
?
Nov 17, 2022 19:49
|
|
|
aren’t the auto fields are based on the first document (log entry) that has something in that field? so which logs hit the index first will determine what gets detected you’re better off specifying them if you can. get a dump of the current one that’s close and edit it and push it to a new index
|
|
#
?
Nov 17, 2022 20:00
|
|
|
Destroyenator posted:aren’t the auto fields are based on the first document (log entry) that has something in that field? so which logs hit the index first will determine what gets detected yeah that's what I thought as well, but also that the templates could get modified by later entries as long as the API key granted overwrite permissions. good point though maybe I'll just export a document from one to the other or something like that. then again it doesn't make sense that one instance has it set up correctly but not the other because both are running the same process to bootstrap a new index pattern
|
|
#
?
Nov 17, 2022 21:17
|
|
|
hmm when I query items in the index in the console they look the same and both state they use ECS 1.5.0, so this might just be kibana being a pile of poo poo
|
|
#
?
Nov 17, 2022 21:25
|
|
|
Destroyenator posted:aren’t the auto fields are based on the first document (log entry) that has something in that field? so which logs hit the index first will determine what gets detected at an old job we shoved all our logs into a single daily index. the schema was defined by the first log entry(s) to hit it that day, every day. gotta hope you didn't name your structured logging field the same as someone else!
|
|
#
?
Nov 17, 2022 21:49
|
|
|
especially bad for error logs that don't normally happen
|
|
#
?
Nov 17, 2022 21:49
|
|
|
you will always get the same looking documents out (unless you've configured some remappings or something) because it doesn't modify the documents, it just indexes them incorrectly. you should try calling this one for each index https://www.elastic.co/guide/en/elasticsearch/reference/8.5/indices-get-mapping.html to see what it has created. you can manually edit the response to clean up any datatypes that are incorrect and then use the update (put) versions of that api to create new indices with the correct mapping, and not rely on the auto mapping detection unless whatever ingestion you're doing it creating a correct mapping for you and they both match, then idk
|
|
#
?
Nov 18, 2022 09:13
|
|
|
if you add new fields later i don't know how it will react though, that approach applies more for consistent documents
|
|
#
?
Nov 18, 2022 09:15
|
|
|
thx I'll try that. the indexes are pushed from Serilog behich uses the elastic common schema, so mappings etc are supposed to be loaded from the xpack or whatever because they conform to the logs-* pattern, and the API key has cluster privileges to push templates (and Serilog is configured to update/overwrite mappings) and this worked fine on one instance honestly though the docs *suck* stuff just randomly working or not between versions or things you'd expect to be there just not, or being named in such a weird way you can't find them it's worse than goddamn technet
|
|
#
?
Nov 18, 2022 09:58
|
|
|
Destroyenator posted:you will always get the same looking documents out (unless you've configured some remappings or something) because it doesn't modify the documents, it just indexes them incorrectly. you should try calling this one for each index https://www.elastic.co/guide/en/elasticsearch/reference/8.5/indices-get-mapping.html to see what it has created. you can manually edit the response to clean up any datatypes that are incorrect and then use the update (put) versions of that api to create new indices with the correct mapping, and not rely on the auto mapping detection just be warned that if you’re trying to dump a bunch of different docs into one soup, then there IS a field limit
|
|
#
?
Nov 18, 2022 16:23
|
|
|
I've never used it but I've been curious about cuelang instead of json or yaml is it any good?
|
|
#
?
Nov 21, 2022 02:50
|
|
|
it's for use on computers, so no
|
|
#
?
Nov 21, 2022 02:55
|
|
|
Dijkstracula posted:awk owns, we should have an awk appreciation thread this
|
|
#
?
Nov 21, 2022 03:06
|
|
|
Achmed Jones posted:it's for use on computers, so no gottem
|
|
#
?
Nov 21, 2022 03:27
|
|
|
titaniumone posted:I've never used it but I've been curious about cuelang instead of json or yaml ive inherited a lot of structured configuration that we have a burgeoning json-schema commitment to, and this looks extremely better than that, so i will try it out and let you know in a week or so
|
|
#
?
Nov 21, 2022 06:26
|
|
|
early paper from Amazon, interesting read: https://www.allthingsdistributed.com/2022/11/amazon-1998-distributed-computing-manifesto.html
|
|
#
?
Nov 21, 2022 15:56
|
|
|
oh my god OH MY GOD this loving elastic logging problem was nothing to do with the index templates at all! It was that I'd moved a goddamn statement close in the logger init so the first log target would get added, log a message saying "I'm initialised", thenthe second one was getting added and overwriting the whole config instead of adding to the source list of the existing one so the logger looked like it started but was actually being shut down on whatever the first defined target was. it was only when I thought to flip the log target config I realised what was happening gently caress!
|
|
#
?
Nov 23, 2022 12:52
|
|
|
Powerful Two-Hander posted:oh my god lol computers are a hell gently caress
|
|
#
?
Nov 23, 2022 13:59
|
|
|
you hate to see it
|
|
#
?
Nov 23, 2022 16:53
|
|
|
Powerful Two-Hander posted:gently caress! hell, 
|
|
#
?
Nov 23, 2022 18:22
|
|
|
Powerful Two-Hander posted:oh my god 
|
|
#
?
Nov 23, 2022 18:23
|
|
|
owned aside from that, anyone knows if there's any good books for c++20? i learned c++11 by going through "the c++ programming language" by stroustrop when the updated c++11 version came out and that was pretty comprehensive, but there doesn't seem to be a new edition for c++20. i don't think i'm gonna find anything as comprehensive, but anything that goes into the language in some depth would be welcome
|
|
#
?
Nov 24, 2022 18:59
|
|
|
The C++ thread in SH/SC would know better, I'm sure, but I imagine for the major features like stripped-down concepts ("template constaints") and the new memory model, going to the published papers is likely a better resource. I know the "Concepts Lite" paper is fairly old but I'm fairly sure it's only now getting integrated into the actual standard, for instance. As for the memory model, I'm not actually sure what they settled on for c++20/23 but going to Viktor Vafeiadis' and Derek Dreyer's work in the space might be a good intro.
|
|
#
?
Nov 24, 2022 19:23
|
|
|
Deep Dish Fuckfest posted:owned The new tour of C++ book is probably the best place to start https://www.stroustrup.com/tour3.html
|
|
#
?
Nov 24, 2022 19:27
|
|
|
Deep Dish Fuckfest posted:owned I read the c++17 version, but Marc Gregoire’s Professional C++ was good and I think there’s an updated edition out now.
|
|
#
?
Nov 25, 2022 01:54
|
|
|
someone in the sh/sc c++ thread suggested this http://cppstd20.com/ and it seems more or less exactly what i was looking for. there's a c++17 version too and i think i'll start there since they really seem to cover only the new stuff and nothing else in each of the books
|
|
#
?
Nov 25, 2022 04:24
|
|
|
nice, that looks good, ty for reporting back 
|
|
#
?
Nov 25, 2022 05:13
|
|
|
as a reminder: c++ is so complex that there’s an entire book on how the = operator works (cppmove.com) Really let that sink in
|
|
#
?
Nov 25, 2022 17:07
|
|
|
|
| # ? Apr 24, 2024 03:34 |
|
|
yeah but if you use it for long enough you literally go insane and it starts making a lot of sense
|
|
#
?
Nov 25, 2022 17:21
|
|