Motivation
This is a bit of an ambitious one, but as a part of #241 I noticed that we are creating empty pyproject.toml if it doesn't already exist. But that seems unnecessary - we should be able to generate empty requirements.txt without creating an empty dependencies list.
But we can probably do better in this case; we can potentially infer requirements based on the project's imports. Of course, this could potentially be unreliable for cases where the import doesn't match the PyPI name (and potentially a security concern due to typosquatting on PyPI...) so I don't know if this is wise or not. Also should check if something off-the-shelf already exists.
Summary
Consider whether it makes sense to have something to infer requirements from imports and venv.
Motivation
This is a bit of an ambitious one, but as a part of #241 I noticed that we are creating empty
pyproject.tomlif it doesn't already exist. But that seems unnecessary - we should be able to generate emptyrequirements.txtwithout creating an empty dependencies list.But we can probably do better in this case; we can potentially infer requirements based on the project's imports. Of course, this could potentially be unreliable for cases where the import doesn't match the PyPI name (and potentially a security concern due to typosquatting on PyPI...) so I don't know if this is wise or not. Also should check if something off-the-shelf already exists.
Summary
Consider whether it makes sense to have something to infer requirements from imports and venv.