|
Presumably you use a query parameter to act on the resource being filtered and the query parameter indicates both that filtering is happening and the resource being used to achieve that filter
|
#
?
May 21, 2019 20:38
|
|
|
|
| # ? Feb 14, 2025 08:47 |
|
|
HoboMan posted:ok, again i have not had to design poo poo before. your /foo/bar resource would return a link to /wangs?of=/foo/bar or something like that.
|
|
#
?
May 21, 2019 20:59
|
|
|
URLs are not ids any more than pointers are ids
|
|
#
?
May 21, 2019 21:12
|
|
|
used the object id then, whatever. point is the parent resource contains links to what you need.
|
|
#
?
May 21, 2019 21:21
|
|
|
Sapozhnik posted:URLs are not ids any more than pointers are ids RFC 3986 would beg to disagree. quote:A Uniform Resource Identifier (URI) provides a simple and extensible means for identifying a resource. quote:The term "Uniform Resource Locator" (URL) refers to the subset of URIs that, in addition to identifying a resource, provide a means of locating the resource by describing its primary access mechanism (e.g., its network "location").
|
|
#
?
May 21, 2019 21:38
|
|
|
HoboMan posted:ok, again i have not had to design poo poo before. https://yourdumbservice.com/the-first-resource?the-second-resource=1
|
|
#
?
May 21, 2019 22:21
|
|
|
necrotic posted:your /foo/bar resource would return a link to /wangs?of=/foo/bar or something like that. wtf if you're using dotnet then you can bind variables out of the url so you can have foo\123\butt\456 and in your controller get fooid as 123 and buttid as 456 you just gotta decorate your controller base route and actions with a binder like foo\{int:fooid} \butt\{int:buttid} edit: I'm not really sure if that's what you mean but you can also use redirects to actions to hit an endpoint, do something, get the value returned and redirect to another that performs another action off that returned value
|
|
#
?
May 21, 2019 23:50
|
|
|
Powerful Two-Hander posted:wtf I think the debate is over how RESTful that API is for the usecase. Maybe OP should clarify the use case a little?
|
|
#
?
May 22, 2019 00:01
|
|
|
ComradeCosmobot posted:I think the debate is over how RESTful that API is for the usecase. Maybe OP should clarify the use case a little? yeah he said restful not hierarchical
|
|
#
?
May 22, 2019 00:55
|
|
|
HoboMan posted:ok, again i have not had to design poo poo before. Assuming you're just trying to GET all of the resources that "contain", for lack of a more generic word, some other resource. In other words, searching for all resources that have some property that matches another resource. I'd probably use one of two options depending upon what your actual use-case is. First would be a query param on the route for the resource: /api/butts?kissed_by=[goon_id] My alternative would be swapping the resources and use a nested route of the single resource by which you're trying to search: /api/goons/:goonId/kissed_butts I'd only use the latter if there's an obvious hierarchical relationship between resources. Like, if Bars only exist as children of Foos, then the latter would work (/foos/:fooId/bars). In basically any other case, like if Foos and Bars might be associated but could be coexistent without requiring the other, then I'd go with a query parameter. The latter is also probably not strict REST but the internet is full of awful blog posts about how to do REST correctly and all of them have different opinions so I have no loving idea.
|
|
#
?
May 22, 2019 03:55
|
|
|
sounds like you need graphql OP. or possibly SOAP
|
|
#
?
May 22, 2019 08:23
|
|
|
um pretty sure it's time to create a dsl to handle this for maximum flexibility, unless you can make do with OData (aka the Best Way to REST).
|
|
#
?
May 22, 2019 09:56
|
|
|
I'm still not sure what the OP is trying to do. Is the filtering between any two arbitrary services? What even is the use case?
|
|
#
?
May 22, 2019 10:04
|
|
|
it's good to see that even if people don't understand the use case or even its most basic requirements they won't hesitate to suggest solutions that may or may not address the problem developers.txt
|
|
#
?
May 22, 2019 10:09
|
|
|
OP definitely needs SOAP. SOAP makes everything better.
|
|
#
?
May 22, 2019 10:19
|
|
|
Sagacity posted:it's good to see that even if people don't understand the use case or even its most basic requirements they won't hesitate to suggest solutions that may or may not address the problem
|
|
#
?
May 22, 2019 12:16
|
|
|
yeah it's, as pseudorandom suggested, get all butts kissed by a given goon
|
|
#
?
May 22, 2019 13:30
|
|
|
we just had our architect copy and paste a random blog post* about RESTfulness into our confluence and announce 'start doing this now' becoming standards document #9801293 that has no relation to our codebase whatsoever and probably never will lol *it had to the gall to quote roy fielding's dissertation just for a random pull quote about 'URL == resource' yet the rest of the article ignores any important parts of rest and just talks about underscores vs hyphens and poo poo
|
|
#
?
May 22, 2019 13:37
|
|
|
sounds like every other "restful design doc" ive ever seen
|
|
#
?
May 22, 2019 16:07
|
|
|
does anybody actually use self-documenting features of REST apis? github's api has a million extra annotations on every message and they seem pretty pointless to me
|
|
#
?
May 22, 2019 16:24
|
|
|
animist posted:does anybody actually use self-documenting features of REST apis? github's api has a million extra annotations on every message and they seem pretty pointless to me sounds about as useful as other forms of "self documenting"
|
|
#
?
May 22, 2019 16:27
|
|
|
animist posted:does anybody actually use self-documenting features of REST apis? github's api has a million extra annotations on every message and they seem pretty pointless to me We produce RAML for our which our customers do seem to use, well they complain when we break it anyway. I made a toy app that uses the RAML to create a visualisation of all the data in the system which is quite cool.
|
|
#
?
May 22, 2019 16:30
|
|
|
animist posted:does anybody actually use self-documenting features of REST apis? github's api has a million extra annotations on every message and they seem pretty pointless to me I spent quite a long tine trying to generate swagger from http4s routes, or generating them from a swagger spec file, never really made any progress though.
|
|
#
?
May 22, 2019 16:54
|
|
|
animist posted:does anybody actually use self-documenting features of REST apis? github's api has a million extra annotations on every message and they seem pretty pointless to me I've used it to scrape data off apis backing website frontends
|
|
#
?
May 22, 2019 17:40
|
|
|
swagger generation in .net is super easy. just add the swashbuckle package
|
|
#
?
May 22, 2019 17:42
|
|
|
swashbuckle is fuckin fantastic and the swagger api docs i had on a tool at lastjob were probably my group's most widely used documentation and definitely the most used per hour spent writing
|
|
#
?
May 22, 2019 17:49
|
|
|
idk why anyone would bother with something other than .net for web stuff
|
|
#
?
May 22, 2019 17:52
|
|
|
animist posted:does anybody actually use self-documenting features of REST apis? github's api has a million extra annotations on every message and they seem pretty pointless to me we're merging with a different company in the same sector and their cto was very excited to tell me that they've started doing the full HATEOAS thing with their new microservice architecture. july is gonna be a fun month
|
|
#
?
May 22, 2019 18:03
|
|
|
yeah swagger works out of the box with e.g. play, just not http4s
|
|
#
?
May 22, 2019 18:24
|
|
|
florida lan posted:OP definitely needs SOAP. SOAP makes everything better. just lol at the idea of goons using SOAP
|
|
#
?
May 22, 2019 21:05
|
|
|
i think i'm getting gaslit by my dev tools literally cant tell what is real anymore
|
|
#
?
May 22, 2019 21:43
|
|
|
HoboMan posted:i think i'm getting gaslit by my dev tools I wonder if there's a way to have preprocessor macros that change behavior depending on the time of day or calendar day the build is occurring builds that only work m-f 9-5 with a 30 minute break for lunch builds that produce different output on April 1 and/or April 20
|
|
#
?
May 22, 2019 21:53
|
|
|
make can read system environment variables and it can change the values of preprocessor macros with command line arguments to gcc only step remaining is some job running in the background to update the environment variables based on current datetime
|
|
#
?
May 22, 2019 22:52
|
|
|
HoboMan posted:i think i'm getting gaslit by my dev tools we’re terrible programmers, not dev tools
|
|
#
?
May 22, 2019 23:13
|
|
|
Progressive JPEG posted:I wonder if there's a way to have preprocessor macros that change behavior depending on the time of day or calendar day the build is occurring RPATDO_LAMD posted:make can read system environment variables and it can change the values of preprocessor macros with command line arguments to gcc gcc posted:__DATE__
|
|
#
?
May 22, 2019 23:49
|
|
|
i was hoping this wasn't a real domain and now i am contemplating buying it
|
|
#
?
May 23, 2019 00:23
|
|
|
rest is dumb and bad asking here rather than the PL thread, but what are git submodules and why are they so bad?
|
|
#
?
May 23, 2019 01:24
|
|
|
redleader posted:rest is dumb and bad git submodules are where one git repo includes a pointer to specific commit in another git repo, that gets checked out inside the main one as a directory in theory they’re a nice way of breaking up a giant repo into logical chunks in a manageable way in practice they’re a tire fire and massive pain in the rear end to keep up to date. imo this is mainly a tooling problem, but git devs would do diy dentistry than make a sensible UI decision so it’ll never get any better
|
|
#
?
May 23, 2019 01:30
|
|
|
what if your git repo could incorporate by reference other git repos? and what if each of those git repos could do the same thing?! and what if none of it worked reliably, you constantly fought with stupid recursive git commands, and you came to wish that you'd just written and checked-in a five line shell script containing some git commands instead?
|
|
#
?
May 23, 2019 01:30
|
|
|
|
| # ? Feb 14, 2025 08:47 |
|
|
there are multiple layers of badness to it. git does not handle submodules well at all. from top to bottom it is way more of a pita than it should be. SVN linked repos were easier to work with ffs. people also largely use them for lazy, ad-hoc dependency management in a way that's awful, fragile and stupid. Arcsech posted:in theory they’re a nice way of breaking up a giant repo into logical chunks in a manageable way yes, this is effectively ad hoc dependency management and you're basically 100% always better off doing it whatever the "right way" is for your lang/environment
|
|
#
?
May 23, 2019 01:36
|
|