Run tests using reference data from purl-spec#80
Run tests using reference data from purl-spec#80pombredanne merged 4 commits intopackage-url:masterfrom
Conversation
Signed-off-by: Keshav Priyadarshi <git@keshav.space>
Signed-off-by: Keshav Priyadarshi <git@keshav.space>
Hey, thanks for doing this. I started looking at the failing tests and hope to be able to put up a PR soon. But there is indeed a number of fixes needed to the code, not least is it not very good at encoding purls according to spec. I based my code on your changes in this PR and have a few suggestions which I felt made life simpler when writing (and trying to fix broken) tests. Would you mind me commenting although I'm no repo maintainer? |
Yes please, suggestions are welcome. |
petergardfjall
left a comment
There was a problem hiding this comment.
I've proposed a few changes which made the tests a bit easier to work with for me at least. Feel free to incorporate them!
|
Hey, for you information I've opened a PR where I attempt to fix the code according to spec: #83 It still fails on a number of the new test cases but I've opened a few tickets since I don't understand the requirements that the test cases appear to place on library implementations: |
Signed-off-by: Keshav Priyadarshi <git@keshav.space>
|
@petergardfjall thanks, I've incorporated your suggestion into the PR. |
I might be wrong, but I think there was an in-flight PR by @TG1999 to fix the encoding related test in the purl-spec repo. I can't seem to find it. |
|
@petergardfjall do you want to get committer access? |
Hey @pombredanne! What does "committer access" mean in practice here? I'm not sure I would be comfortable making commits without someone knowledgable to review them. The purl spec is very important in the project I'm currently working on so quicker turnaround/responsiveness is desriable. |
|
@petergardfjall this would give you commit rights, but we are not committing things without reviews for sure! |
|
@shibumi @mcombuechen we need to adopt the new test suite here soon enough... any objection to this PR? Can review and approve? Next up will be to fix the tests. |
|
SGTM to me. We just have to fix the tests :) |
|
I think this is good to go. Would another maintainer care to review? With the code changes in #83 (using Merging this PR first and then merging #83 (after review, of course) seems like a sensible course of action to me. |
petergardfjall
left a comment
There was a problem hiding this comment.
Giving it my approval in case that helps.
Signed-off-by: Keshav Priyadarshi <git@keshav.space>
|
@petergardfjall thanks for the encoding fix in the purl-spec test fixture. Number of failing tests is now 65, down from 85. |
|
I've rebased #83 onto this PR and all tests now pass (if pending purl-spec testcase PRs are accepted)! So merging #80 into |
|
Unit tests are failing. Can you take a look? |
|
Thanks. Merging now! |
Adds test coverage using reference data provided by
purl-spechttps://github.com/aboutcode-org/purl-spec/tree/main/tests/types.The test schema is available https://github.com/aboutcode-org/purl-spec/blob/main/schemas/purl-test.schema.json.
Note
Out of 481 reference tests, 85 are failing.
Related issue: aboutcode-org/purl-spec#30