Fix PURL test failures, migrate cyclonedx-bom to purl crate#746
Fix PURL test failures, migrate cyclonedx-bom to purl crate#746Shnatsel merged 7 commits intoCycloneDX:mainfrom
cyclonedx-bom to purl crate#746Conversation
…onstructor Signed-off-by: Sergey "Shnatsel" Davidoff <shnatsel@gmail.com>
Signed-off-by: Sergey "Shnatsel" Davidoff <shnatsel@gmail.com>
Signed-off-by: Sergey "Shnatsel" Davidoff <shnatsel@gmail.com>
…just use the `purl` crate. Signed-off-by: Sergey "Shnatsel" Davidoff <shnatsel@gmail.com>
Signed-off-by: Sergey "Shnatsel" Davidoff <shnatsel@gmail.com>
… for us Signed-off-by: Sergey "Shnatsel" Davidoff <shnatsel@gmail.com>
Signed-off-by: Sergey "Shnatsel" Davidoff <shnatsel@gmail.com>
|
Thank you for this. It looks good to me. Should phylum-dev/purl#11 be closed if it's handled now? |
|
Good point, I closed the upstream issue. |
|
@Shnatsel hey, can you tell me more about the purl vs. the packageurl crates? (My interest is as the creator of the PURL spec ;) ) |
|
@pombredanne as of this PR, the (I'd appreciate if you could take a look at that PR, it's a major interoperability hazard) Meanwhile the |
cargo cyclonedxDrop our custom percent encoding machinery now that the
purlcrate implements this for us. This was one hell of a rabbit hole, and it is a testament to the quality of thepurlcrate that they are actually handling this nonsense correctly now.Fixes #745
cyclonedx-bomMigrate from the very incomplete
packageurlto thepurlcrate already used bycargo cyclonedx. We used to have two different PURL crates in the tree, this brings the number back to one. Also provide a more informative error message on validation errors.Fixes #636
This is all still semver-compatible, because the
packageurlcrate was never actually exposed to the outside world. But this is a PR against main because I cannot be bothered to backport this.