Closed nilshartmann closed 2 years ago
Hi Nils,
I agree with what you are suggesting.
Keeping a frontend could be interesting. The frontend part could me move to a dedicated spring-petclinic-react
which requires a GraphQL backend.
By the way, now that we have an official Spring GraphQL
project, does it make sense the spring-graphql
branch become the main/master
branch?
Hi Antoine,
thanks for your response.
Before pushing spring-graphql
to main/master
(should we rename to main
then?) I'd suggest to update to latest Spring Boot 2.7 Milestone. As I'm currently having some time, I'm already doing the update here on a local branch, think I will finish it, when Spring Boot 2.7M3 soon will be released. As part of the upgrade I also could create a spring-petclinic-react
(or spring-petclinic-graphql-react
) and move the existing client to that project.
Yes we could now use the main
branch name and set it to the default one. I still never do that, but let's try.
It's ok for me that you are using the latest Spring Boot Milestone. And I'm ok for the name spring-petclinic-graphql-react
(but we could rename spring-petclinic-angular
to spring-petclinic-angular-rest
). I think you have the permission to create a new repo.
Thanks a lot for your repsonse, Antoine! I will do so. If I'm having not enough permissions or other problems, I'll ping you again :)
Good morning @arey ,
on the spring-graphql
branch I already removed the configuration for mysql and postgres, so that branch only contains the hsql example 😬.
I removed postgres and mysql config, because I changed the hsql schema slightly and also created more sample data and hadn't had the time to adjust the config for the other DBs as well.
But: do you think we actually need mysql and postgres config? I'm a little bit unsure, because if one wants to learn how "GraphQL for Spring" works, there is imho no need for using another database, because spring-graphql doesn't matter about a concrete database (esp. if we use Spring Data JPA that abstracts the DB anyway). Personaly I find having multiple configurations a little distracting because it shifts the focus away from the "real" things needed for GraphQL . Also that means more work for us to maintain and complicates testing.
If you disagree, I'll re-add the config for postgres and mysql and adjust it with the changes I already did for hsql schema.
What do you think?
Hi Nils,
Supporting a production ready database like PostgreSQL or MySQL could be interesting and reassure some developers. But if you don't have time, no problem, it could wait an external contribution.
Personally, for sample applications like Petclinic I'm only using HSQL or H2 database.
Thus it's OK for me
Thanks for your feedback!
Then I will focus on hsql / h2 first.
Hi,
I'm going to upgrade the
spring-graphql
branch to latest spring boot and spring-graphql versions soon (waiting for SB 2.7.0 M3 to be released).And now I'm wondering, if we should remove the react frontend from the example at all and only provide the backend (like
spring-petclinic-rest
is doing).Reasons:
As an alternative we could provide frontends in own projects, as the
spring-petclinic-angular
project is doing for thespring-petclinic-rest
backend.I really enjoyed building the React frontend and still think it's a good demonstration of a powerful tech stack, I think it's too much for this spring-petclinic-graphql example 😢
What do you think?