fix(python): unknown type when submodule is not loaded #3049
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Callbacks may make references to types from submodules of loaded packages, however due to how the type registration works in the Python runtime, if the submodule has not been loaded, then the types it contains are not registered and will fail resolving. The simplest possible fix is to preemtively load all submodules upfront at the end of the root module's `__init__.py` file, which guarantees that all types have correctly been registered in the runtime (at the expense of some time and memory). Fixes aws/aws-cdk#16625
njlynch
approved these changes
Oct 8, 2021
…ules # Conflicts: # packages/jsii-calc/test/assembly.jsii
njlynch
approved these changes
Oct 11, 2021
|
@Mergifyio refresh |
|
Command
|
5 tasks
mergify bot
pushed a commit
that referenced
this pull request
Nov 18, 2021
The goal of this PR is, to find a way for generated Python code to work with namespaced modules regardless of the way it is consumed. While the current implementation of submodule imports seem to work fine when distributed as packages, it's breaking apart when the generated code is imported relatively as part of a single code base. That's what we do in cdktf and the reason why we have locked jsii to `1.37.0` at the moment. The general idea is, to generate relative imports rather than fqn imports in Python. This could be a potential way to address #3078 ```python # Before import jsii_calc.submodule import jsii_calc.submodule.nested_submodule import jsii_calc.submodule.nested_submodule.deeply_nested # After from . import submodule from .submodule import nested_submodule from .submodule.nested_submodule import deeply_nested ``` Besides from a few snapshots being changed, it seems like the tests are still working with this change. However, I'm not sure about the general implications. So, this entire PR is really meant to be a conversation starter. I have confirmed that an import like `import . from some_aws_namespace` would work for us in cdktf. ## References - #3078 - #3049 - hashicorp/terraform-cdk#1152 - hashicorp/terraform-cdk#1288 --- By submitting this pull request, I confirm that my contribution is made under the terms of the [Apache 2.0 license]. [Apache 2.0 license]: https://www.apache.org/licenses/LICENSE-2.0
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Callbacks may make references to types from submodules of loaded
packages, however due to how the type registration works in the Python
runtime, if the submodule has not been loaded, then the types it
contains are not registered and will fail resolving.
The simplest possible fix is to preemtively load all submodules upfront
at the end of the root module's
__init__.pyfile, which guaranteesthat all types have correctly been registered in the runtime (at the
expense of some time and memory).
Fixes aws/aws-cdk#16625
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.