Is REST perfect? Probably not. But it certainly does its job, and that's why most middle-tier APIs these days, whether it be those that power web applications or mobile apps, are in REST, with the backend itself being written in a variety of languages (PHP, Python, Java, C#, etc).
I will admit that core REST doesn't support basic "querying" functionality, or even things like pagination, filtering, sorting, etc, which is why there are a set of standards, or best practices on top of REST that aim to standardize those commonly used patterns. [1] [2] [3]
Finally REST, whether by design or not, follows the KISS principle (Keep It Simple and Stupid), and that's probably why it's gained as much traction as it has over the years.
Want to get a user object? GET /user/<id>
Want to update it? POST or PUT /user/<id>
Get a list of all users? GET /user/
If you look at just the "basic" examples for GraphQL, you will understand why it's never going to replace REST in its current form.
I find that REST was 'perfect' given the status-quo of APIs at the time it came out. It allowed for massive simplification of APIs compared to its XML-based predecessors and enabled every man and his dog to make a half-decent semi-standardized API.
For better of for worse as things progressed new requirements and complicating factors like the ones you mentioned (filtering, pagination, need for dynamism etc) came up that ended up being handled in a dozen different ways muddying up the waters.
To me, GraphQL built on top of REST to cover those bases and IMHO that's how these technology standards are supposed to evolve: thing comes up, it gets used till previously unknown requirements surface, new thing learning from old thing and addressing new requirements comes up.
I was really bothered when I realised that GraphQL doesn't allow for heterogeneous input-type unions though, cause that makes filtering a hassle (eg if you wanna have a query that can take an arbitrary number of filters for fields of different types).
I can second OData, but i think it's important to know that OData has nothing to do with Microsoft anymore ( officially). The only thing is that they support it officially in the dot net environment. It's now an open protocol with lot's of integrations: https://www.odata.org/libraries/
But it's great to use though. You'll love it if you like Linq/lambda's.
/odata/Clients?$filter=Status eq 0 or LastLoggedIn is null
IMHO what makes or breaks a paradigm like REST or GraphQL in real-life is the state of tooling made by third-parties to enable its adoption in the workplace or personal projects.
Working in Python-land I'm pretty confident REST wouldn't have been as ubiquitous if it wasn't for tools like Flask, marshmallow, and the myriad of other tools.
Having taken up GraphQL in personal projects I can say that at least in Python the tooling just ain't there so until it gets there I don't see GraphQL killing anything.
I think serverless GraphQL is the future of the backend. And honestly, not as familiar with the Python landscape but if you use tools like Prisma (Node-based) or AppSync (AWS Lambda based), a lot of your GraphQL queries and mutations can be generated automatically based on a typed schema file.
Have you tried Graphene? [1] I haven't, but have read its documentation a lot, and it looks quite powerful.
I'm not happy that it introduces yet another data object library, but at least it's possible to build plugins for existing ones -- SQLAlchemy and Django are already there, and it would be nice to add ones for attrs, Schematics and of course the 3.7 native data classes.
Regarding the popularity of REST -- even if it wasn't intended for CRUD originally, it fits perfectly with that paradigm, and I think this is why many people were able to implement it easily when they started using it.
I've actually worked with both Graphene and Graphene-SQLAlchemy a fair bit. Honestly I found that the documentation , albeit well-written, was insufficient (especially for the SQLAlchemy extension). That being said, having to expose an SQLAlchemy ORM via Graphene did not lead to much code-duplication as much of the ORM is automatically accounted-for in the GraphQL schema (there's a fair bit of duplication when it comes to mutations but I reckon that's just a matter of someone doing the work in the extension).
Moreover, the community around it in both its GitHub and StackOverflow isn't as active as the equivalent GraphQL communities in other languages/ecosystems.
All in all I believe that increased adoption of GraphQL will lead to even better tooling and I'm sincerely hoping it'll catch on :).
My primary 'beef' with REST has always been the "10 ways to do 1 thing" especially when it comes to less standardised features like pagination, type-definition/validation, etc. GraphQL introducing some of these things out-of-the-box would hopefully reduce the comfort-zone-based bikeshedding that comes with any new REST API being developed.
I haven't seen anything that would make me advise GraphQL over REST, specially now that REST tooling is finally catching up with validation and graphical generation.
All the presentations I have seen about how productive GraphQL is, seem to live in JavaScript with NoSQL databases silos.
Admittedly JS-land has some fantastically developed and documented tooling for GraphQL. I reckon that increased adoption, however, will see equivalent efforts being made in other ecosystems.
REST can practically never die - it's just describes how most of the internet works: HTTP + links.
Maybe people will stop using REST APIs, but REST will still be a thing as long as HTTP is.
GraphQL seems like it can solve some pain points of REST once you start using REST APIs at a massive scale. REST improves on SOAP in even the tiniest of cases.
I will admit that core REST doesn't support basic "querying" functionality, or even things like pagination, filtering, sorting, etc, which is why there are a set of standards, or best practices on top of REST that aim to standardize those commonly used patterns. [1] [2] [3]
Finally REST, whether by design or not, follows the KISS principle (Keep It Simple and Stupid), and that's probably why it's gained as much traction as it has over the years.
Want to get a user object? GET /user/<id> Want to update it? POST or PUT /user/<id> Get a list of all users? GET /user/
If you look at just the "basic" examples for GraphQL, you will understand why it's never going to replace REST in its current form.
[1] Microsoft's OData: https://www.odata.org/ [2] https://www.moesif.com/blog/technical/api-design/REST-API-De... [3] https://stackoverflow.com/questions/207477/restful-url-desig...