DataLoader and the Problem it solves in GraphQL
Vložit
- čas přidán 22. 07. 2024
- Interested in learning more about GraphQL? checkout my course "GraphQL for Beginners with JavaScript" at: knowthen.com/graphql
When you first start learning GraphQL, there’s a pretty good chance, that you’re first attempt at creating a GraphQL API, will have one big Problem, that you really shouldn’t live with.
What is the problem I'm referring to?
Watch and find out.
Source Code: github.com/knowthen/dataloade...
This is probably the clearest tutorial I've ever seen about anything in programming related topics. Pretty brilliant. Congratulations and thank you.
off course
Super awesome! the clearest tutorial about Dataloader i have ever seen! Thanks a lot
Yeah
one of the best video on N+1 problem along with solution
absolutely brilliant tutorial on the Dataloader. I finally get the intuition behind what it actually does now, thanks so much
Incredibly clear ! Never watched/hearded so clear a tuto ! Thanks !
Absolutely brilliant. I have been watching a hell lot of tutorials, and i can say, you are one of the best, if not the best.
The clearest explanation with brilliant slides and editing. Thank you and congratulations
You deserve an award for this tutorial. I was finding it so difficult to teach some junior folks about how to implement dataloader properly. This solved my problem.
As a beginner, that's what I was lookin for. Finally, got an freaking awesome tutorial. Thank you bro!
Thank you! This was exactly what I needed. I've been struggling with the implementation of dataloaders in a large existing graph server and you made things click.
Never commented on youtube. But the knowledge, the premiere pro, the anticipation ... everything is just awesome mate. Strongly recommented channel
This is BY FAR, the BEST tutorial that i have ever seen! OMG! Thanks a LOT!
First video I saw from this channel. Automatically subscribed. Note that I was watching an introduction to DataLoader from the creator of the package himself and I was really in the dark. Now I see clear as day. Thanks!!
love every video James makes, very calming, very informative
this is one of the best tutorials i've seen for awhile
super clear and easy to understand video. Great job. Thank you!
Best dataloader explanation out there. Thank you very much!
Finally a video that I can completely understand. Thanks you Sir...
I'm rather surprised you only have around 5k subs. I really like your pacing and overall production quality. The animations are top notch!
i am lucky to watch this before i implement my graphql API!
The best explanation about Dataloader.
Amazing tutorial. Thank you !!
Perfectly explained. Bravo.
Great explanation. Thank you man!
wow, your explaination is the best.
this is a great explanation. thank-you for this
Great and clean explanation! Keep up the good work 🙌
Thank you very much for the explanation
AWESOME! Clearly explained
Amazing! This is brilliant! Thank you.
Really great video. Thanks
Perfect explanation!
brilliant teaching
great explanation, thanks!
Such a nice tutorial, awesome
Thanks for the video, it was a clear explanation.
still very solid video in 2020 🎉
Awesome tutorial!
This is amazing, thank you for this great video
The best video i found about data loader, Thanks !
Edit: please how can i log the time or the db query in the console are u using a package or just console loging it manually ?
Very nice explaination. Though I prefer passing arrays as first arg in lodash compared to ramda
Awesome...thanks for everything
Cool! I've been using GraphQL only once during a gql workshops and instead of using a DB we were using REST API with static data as a data source. Similar structrure although it were restaurants instead of books :D The problem was that the reviews contained only the ID of the author and the name could be accessed from other endpoint specific for reviews. Ive spotted that it would be inefficient to call each author by id and instead it would be better to just call an endpoint containing all of them and just filter them. Instead of 100 API requests(which could be pricy in real world app) it did one. Ofc that the filtering would is beeing done by SQL query in this case but i think the problem is the same :)
@knowthen, please provide a Patreon link! I'd love to repay you for this, the presentation quality was the best I've seen since on a coding tutorial. Thank you so much :) if I could suggest analyzing some GraphQL techniques such as schema and type generation (eg PostGraphile/Hasura and TypeGraphQL), as well as scaling techniques (gql modules, SDL/schema- vs code-first etc). Once again, thank you!
Amazing, thanks a lot
great!!!! the best explanation!!!!!
Thank you bro. From Viet Nam with love
Super awesome tutorial
Wow wow wow wow 👍...
Thanks you so much.
Awesome video, You won a sub 😁
mah doubts are clear now. Thanks for the awesome explanation.
Very nicely explained. Is it possible for you to give solution using .NET?
Could you please explain how to log the query you send to the datase to the console?
Plz do some more stuff like this
thanks for the great tutorial! quick question: do I need to have a separate data loader for each resolver? or a data loader for each separate collection?
yes you can create loader for each resolver.
This is the best.
Super!
Great!!!
Amazingg!
thanks
I fell in graphql trap without knowing its not better than REST except for standard 200 error code even for exception messages
Love the video, very informative! Also, your verbal mannerisms remind me of Barack Obama (which is a good thing).
Excellent Content, Great Job
An alternative to two functions from the grouping of summary data: codepen.io/starwrapper/pen/orqjQB
it's like SIMD cpu
8:00
When I started developing SQL for relational databases 20+ years ago they all supported joins and the problem defined would be resolved by use of a join. The use of manually coding loops through one table to find entries in another is non-sensical and either reflects a very poor product (GraphQL) or lack of understanding. I assume this guy knows what he is talking about thus GraphQL seems needlessly complex and under-featured. or am I missing something somewhere?
- Quarry, Corey, Quarries
Nice video though.
What i don't understand here is, why are we using different resolver for the reviews? Why not just fetch the reviews in the same books resolver? That way we wouldn't have this problem in the first place.
Because if we don't need the reviews, you are fetching them from the database every time you request a book anyways, making the database do unnecessary work
Or just use Phoenix/Ecto...
"Quarry?"
"Qurry"
Quarry: A quarry is a place, typically a large, deep pit, from which stone or other materials are or have been extracted.
Query: A query is a request for data or information from a database table or combination of tables. This data may be generated as results returned by Structured Query Language (SQL) or as pictorials, graphs or complex results, e.g., trend analyses from data-mining tools.
You're saying it wrong.
That's just his accent man.