Open thomasmaerten opened 4 years ago
This is something we need. A nested structure like the one below would help:
custom:
pythonRequirements:
layers:
layerName1:
layerName2:
At this point I don't understand why it doesn't behave the same way for functions.[].module but for layers.[].path: https://www.serverless.com/framework/docs/providers/aws/guide/layers
What I mean, is if the layer referenced in layers has a "requirements.txt" inside the "path" directory (similar to the "module" directory) then the same kind of pip install process can be carried
Is there any update on this?
Hey @fjpa121197 - no update on this so far, I honestly missed that issue and didn't have a chance to properly review it.
Do you have any suggestions on how the implementation for it should look like? Anyone interested in this issue, please share your proposals 👍
This feature would be a great addition. Please introduce this for custom layers. Let us define a requirements.txt in layer-properties, pick it up and package it.
We are currently struggling with current features as they do not build up our reality. But we need them and are willing and able to implement them and help others.
This addition would be great. Any updates here?
No updates there @maramizo, but if someone would be interested in picking this up, it would definitely be much appreciated 💯
Hello,
I would like to deploy multiple functions having different package dependencies within the same stack. As the packages are too heavy to be included in one layer, I would like to deploy multiple layers with different requirements text files.
Has someone already set up a work around or is it a feature to come ?
Thanks