-
Notifications
You must be signed in to change notification settings - Fork 587
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
fix: in some cases, try to use pom info to guess name and version to top level jar #2080
Conversation
Signed-off-by: Will Murphy <will.murphy@anchore.com>
Signed-off-by: Will Murphy <will.murphy@anchore.com>
All tests pass, but hacky. Signed-off-by: Will Murphy <will.murphy@anchore.com>
Signed-off-by: Will Murphy <will.murphy@anchore.com>
Benchmark Test ResultsBenchmark results from the latest changes vs base branch
|
pomMatches := j.fileManifest.GlobMatch(pomXMLGlob) | ||
var pomPropertiesObject pkg.PomProperties | ||
var pomProjectObject pkg.PomProject | ||
if len(pomPropertyMatches) == 1 && len(pomMatches) == 1 { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
it's possible that we could have only a pom.xml and no properties, I think we should account for that as well (allow for 0 to 1 pom properties)
|
||
func TestWarCatalogedCorrectlyIfRenamed(t *testing.T) { | ||
// install hudson-war@2.2.1 and renames the file to `/hudson.war` | ||
// TODO: is this better expressed in syft/pkg/cataloger/java/archive_parser_test.go? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
agreed, I'm not against having this integration test at all, so consider this comment non-blocking, however, testing specific business logic should be pushed as low as possible test-level-wise (unit is preferred for these kinds of assertions). We do have some regression tests at the integration level to test against the specific asset where we found the regression, so keeping this test here is ok.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the suggestion! I think I'll leave this here for now; it was reported as an end to end regression, and the parser was still finding the correct package, it was just also finding an incorrect extra package, so the end-to-end flow seems a good place of the test.
I'll push a commit to remove the TODO.
return &pkg.Package{ | ||
Name: selectName(manifest, j.fileInfo), | ||
Version: selectVersion(manifest, j.fileInfo), | ||
// TODO: maybe select name should just have a pom properties in it? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
or maybe rename selectName
to something more specific like selectNameFromManifest
(same for the version function) -- the upside is that you can write a selectNameAndVersion
wrapper that is called here and pushes some of this lower level logic to another function, and the existing tests for selectName don't really need to get modified from a logic sense.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the suggestion! I think I'll do this as a follow up.
Signed-off-by: Will Murphy <will.murphy@anchore.com>
Signed-off-by: Will Murphy <will.murphy@anchore.com>
…top level jar (anchore#2080) Otherwise, small renames like 'hudson-war-2.2.1.war' to 'hudson.war', would cause syft to incorrectly catolog the archive. Signed-off-by: Will Murphy <will.murphy@anchore.com>
Implements the following changes in the java archive parser's
discoverMainPackage()
method:if there is exactly one pom.xml and pom.properties in the jar and if the filename is a prefix of the artifact ID on the pom.properties, prefer, in order, for name and version:
Definitely open to feedback and refactoring here, but wanted to show a working fix before spending too much time making it look great.
Fixes #2077