Dockerfile: Install pydantic with --no-binary#17
Dockerfile: Install pydantic with --no-binary#17anthonyfok wants to merge 1 commit intoOpenDRR:masterfrom
Conversation
This is to reduce the size of the resulting Docker image, apparently from 969MB down to 861MB Fixes OpenDRR#15
9fe8a8d to
87090a5
Compare
jvanulde
left a comment
There was a problem hiding this comment.
I think it should be an optional setting. We can note that it's a necessity for AWS Lambda. Perhaps we should propose to the pygeoapi community as well via a PR?
|
Turns out I probably changed the wrong Dockerfile: instead of the one in the top-level directory, I probably should have changed aws-lambda/container/Dockerfile instead. (Thanks to @arashmalekz's comment #15 (comment) telling me about the aws-lambda directory which I neglected.) See also geopython#709 for the origin of aws-lambda/container/Dockerfile. Will revisit this in a future sprint (say Sprint 52?) |
|
Note that since we don't deploy on Lambda any longer we don't need to worry about this. |
Thanks so much Joost! I was wondering about that ('cause it was mentioned at our cloud tech meetings) but wasn't sure of the details. I guess what is left to do in this PR is to (1) apply the change to the correct Dockerfile, and (2) submit a PR upstream, and then we can close this PR. Postponing this till we have more free time. :-) |
This is to reduce the size of the resulting Docker image,
apparently from 969MB down to 861MB
Fixes #15
Questions: Is it a good idea to push this to
masterbranch? Or to keep it just for a special pydantic-no-binary branch for AWS Lambda only? Many thanks!