cdk-appsync-transformer
cdk-appsync-transformer copied to clipboard
Update to CDK v2.x
@kcwinner when do you plan to upgrade the project to CDK 2.x? Please let me know if I can help with anything. Feel free to ping me if you want some help with this, if you have any architectural patterns you want to follow, we can coordinate this before I actually get started.
Best regards,
Vin Busquet
@computationalcore you can try it out at v2.0.0-alpha.0
or use cdk-appsync-transformer@next
. I created a sample cdk app and tested a few things. But I don't have a live project right now to test against thoroughly.
@computationalcore you can try it out at
v2.0.0-alpha.0
or usecdk-appsync-transformer@next
. I created a sample cdk app and tested a few things. But I don't have a live project right now to test against thoroughly.
Got it. Thanks for the feedback and update. I can try later. Also I noticed what you mentioned later, that CDK 2 appsync support is still alpha for L2 constructs, only the L1 constructs are stable (since they are a direct map to Cloudformation).
Another question, have you checked out the new amplify data modeling directives that replaces @connection directive?
Do you think it's an interesting/feasible thing to add support? Again, if you want you can count on me for some cooperation, I just want to avoid blind contributions.
Best Regards,
Vin Busquet
@computationalcore I've been looking at supporting the new directives. This is looking more and more like a breaking change to support it. I've got a basic setup reading the new directives into the transformer, but the way it outputs uses CDK natively now instead of their cloudformation setup.
So, this library may make less sense the way it works now. I see a few paths:
- Convert the new amplify outputs to exactly what we have
- Output the CDK stacks from Amplify and let the owner determine what to do (would address #333 as well)
- Some combination of the two
- ???
@kcwinner got it. It seems to me that a combination of the two is the ideal, It seems to me that a combination of the two is ideal, but for compatibility reasons and perhaps narrowing the scope, I would vote to convert the new amplify outputs to exactly what we have (I'm also not sure if that's the simplest path).
This issue is now marked as stale because it hasn't seen activity for a while. Add a comment or it will be closed soon.
This issue is now marked as stale because it hasn't seen activity for a while. Add a comment or it will be closed soon.
Any update on the progress of this given CDK v1.x is officially in maintenance now too.
This issue is now marked as stale because it hasn't seen activity for a while. Add a comment or it will be closed soon.
Closing this issue as it hasn't seen activity for a while. Please add a comment @mentioning a maintainer to reopen.