You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In many places the parsed values are split at # to delete the namespace, e.g. here. This probably originates from the rdf parser, but this is also done in the jsonyamlxml-writer. As mentioned in #247 it should be validated if the part before #really is the namespace so that this part is not blindly regarded.
Related to that it would make sense to check that the namespace is deleted while parsing the file so that only the "raw" values occur in the internal data model and it is no longer required to split values while writing.
Also, the namespace should only be discarded if it actually matches the document namespace. If it is different (i.e. referencing an external element from another document) we have to keep it.
Activity
armintaenzertng commentedon Oct 27, 2022
Also, the namespace should only be discarded if it actually matches the document namespace. If it is different (i.e. referencing an external element from another document) we have to keep it.
zvr commentedon Oct 27, 2022
As @armintaenzertng wrote, this is really not a validation -- it's a check to see whether the namespace can be deleted as a shorthand.
[-]validate namespace before splitting values[/-][+]check if namespace can be deleted in SPDXIDs[/+]meretp commentedon Oct 27, 2022
related to #105
armintaenzertng commentedon Mar 30, 2023
This is fixed in the refactored version now found on main.