Validate ExternalPackageRef #439
Conversation
Signed-off-by: Armin Tänzer <[email protected]>
Signed-off-by: Armin Tänzer <[email protected]>
Signed-off-by: Armin Tänzer <[email protected]>
Signed-off-by: Armin Tänzer <[email protected]>
Signed-off-by: Armin Tänzer <[email protected]>
meretp
left a comment
meretp
left a comment
There was a problem hiding this comment.
Thanks for the addition! I didn't look at each testcase and regex in detail, but overall this looks very good to me. Only two remarks from my side.
I'm open to discuss if a dictionary approach is suitable here but as I have another use case with the rdf writer, I think it makes sense to use such a mapping.
| CPE22TYPE_REGEX = r'^c[pP][eE]:/[AHOaho]?(:[A-Za-z0-9._\-~%]*){0,6}$' | ||
| CPE23TYPE_REGEX = r'^cpe:2\.3:[aho\*\-](:(((\?*|\*?)([a-zA-Z0-9\-\._]|(\\[\\\*\?!"#$$%&\'\(\)\+,\/:;<=>@\[\]\^`\{\|}~]))+(\?*|\*?))|[\*\-])){5}(:(([a-zA-Z]{2,3}(-([a-zA-Z]{2}|[0-9]{3}))?)|[\*\-]))(:(((\?*|\*?)([a-zA-Z0-9\-\._]|(\\[\\\*\?!"#$$%&\'\(\)\+,\/:;<=>@\[\]\^`\{\|}~]))+(\?*|\*?))|[\*\-])){4}$' | ||
| MAVEN_CENTRAL_REGEX = r'^[^:]+:[^:]+(:[^:]+)?$' | ||
| NPM_REGEX = r'^[^@]+@[^@]+$' | ||
| NUGET_REGEX = r'^[^/]+/[^/]+$' | ||
| BOWER_REGEX = r'^[^#]+#[^#]+$' | ||
| PURL_REGEX = r'^pkg:.+(\/.+)?\/.+(@.+)?(\?.+)?(#.+)?$' | ||
| SWH_REGEX = r'^swh:1:(snp|rel|rev|dir|cnt):[0-9a-fA-F]{40}$' | ||
| GITOID_REGEX = r'^gitoid:(blob|tree|commit|tag):(sha1:[0-9a-fA-F]{40}|sha256:[0-9a-fA-F]{64})$' |
There was a problem hiding this comment.
I would import these regex statements from external_package_ref_validator.py to prevent possible mismatch in the future if one of the regex needs changes and simply to avoid duplications.
| if reference_type == "npm": | ||
| return validate_against_regex(locator, NPM_REGEX, "npm", context) | ||
| if reference_type == "nuget": | ||
| return validate_against_regex(locator, NUGET_REGEX, "nuget", context) | ||
| if reference_type == "bower": | ||
| return validate_against_regex(locator, BOWER_REGEX, "bower", context) | ||
| if reference_type == "purl": | ||
| return validate_against_regex(locator, PURL_REGEX, "purl", context) |
There was a problem hiding this comment.
This is a lot of code duplication. We could use a dictionary instead to map ExternalPackageRefCategory to a set of valid types and map the regex expressions to the corresponding type.
As I am currently working on the rdf writer and there I need to check if the specified type is listed (so one of cpe22Type, cpe23Type, swid and so on) or not, I would need such a mapping as well.
There was a problem hiding this comment.
Very good point, the code is now much more concise. Please have a look if you can work with this in rdf now, too! :)
…fining them Signed-off-by: Armin Tänzer <[email protected]>
Signed-off-by: Armin Tänzer <[email protected]>
meretp
left a comment
meretp
left a comment
There was a problem hiding this comment.
Looks very good to me and yes, I can use it now, thanks a lot! 😊
2 participants
fixes #373