Use refactored version of `tools-python`
meretp opened this issue · comments
Hi!
I am a maintainer of the tools-python
. We refactored the code base and are looking for early adopters to give us some feedback on the refactored code. For now the refactored code is only available on the main branch of the tools, but we will soon make an alpha release of this refactored version to make it available via pip
. We wrote a guide to help with the migration and are happy to support you with the migration if you are interested in using this refactored version.
Another advantage of you using the refactored version of tools-python
would be that some features in the spdx-online-tools
which are using the tools-java
and need some bridging for this could be replaced by the refactored version of the tools-python
.
So, what do you think about this?
Hi @meretp:
First, thank you for the great work on tools-python
!
Second, I'm glad to work on the migration and get your support if it is available. When you cut the alpha release, please let me know and I'll create a branch of ntia-conformance-checker
that uses the refactored version. I can then work on making necessary changes to fix whatever breaks :)
Third, on the aspects related to spdx-online-tools
, @goneall is the expert there, so I defer to him.
Hi @jspeed-meyers!
Great that you want to use the refactored version! Working with a separate branch for this sounds good to me. We actually did the release today, so it's now available on PyPI 🎉
I think the migration should be pretty straight forward. We changed the overall structure, so the imports need to be adapted and some classes also changed, but I think the ntia-confromance-checker
is only affected by the change from NoAssert
to SpdxNoAssertion
.
Thanks, actually the idea to replace some functions by tools-python
came from Gary.
@meretp, congrats on the release 😄
I'll open up a branch and PR tomorrow. I'll ping you again then.
@meretp, I started a new branch using the pre-release version. The next step is to try to fix the errors that are now resulting. I'll take a shot at that on Monday. Edit: I took a quick stab, but looks like it will take at least a little work since it wasn't dead simple.
If it's clear to you how to fix this first error (see the associated PR #114), I'm glad for the advice but I can probably figure it out myself too :)