spdx / spdx/Spdx-Java-Library

Include hasFiles and documentDescribes information in relationship comparison

Offen
#153 4 Kommentare 0 Reaktionen 0 zugewiesene Personen Auf GitHub ansehen

Dieses Issue hat noch niemand übernommen.

wontfix
Vorherrschende Sprache
Java
Sterne
71
Forks
44
Ø Merge
12 Std. 54 Min.
Gemergte PRs (30 T.)
7

Beschreibung

I had a look into the comparison method in the tools-java to find out if we can also use them in the testbed. As this method originates from this repo, I decided to open the issue here. So, apart from the fact that we would need to change one namespace to compare two equal docs, I ran into another issue. This is also an issue within the testbed (#51).
When comparing the doument

SPDXVersion: SPDX-2.3
DataLicense: CC0-1.0
DocumentNamespace: https://some.namespace
DocumentName: document name
SPDXID: SPDXRef-DOCUMENT

## Creation Information
Creator: Tool: test-tool
Created: 2022-01-01T00:00:00Z
## Relationships
Relationship: SPDXRef-DOCUMENT DESCRIBES SPDXRef-fileA
Relationship: SPDXRef-DOCUMENT DESCRIBES SPDXRef-fileB

FileName: ./fileA.c
SPDXID: SPDXRef-fileA
FileChecksum: SHA1: d6a770ba38583ed4bb4525bd96e50461655d2758
LicenseConcluded: NOASSERTION
## Relationships
Relationship: SPDXRef-fileA DEPENDS_ON SPDXRef-fileB
Relationship: SPDXRef-fileA DESCRIBED_BY SPDXRef-DOCUMENT
FileName: ./fileB.c
SPDXID: SPDXRef-fileB
FileChecksum: SHA1: d6a770ba38583ed4bb4525bd96e50461655d2758
LicenseConcluded: NOASSERTION
## Relationships
Relationship: SPDXRef-fileB DEPENDENCY_OF SPDXRef-fileA
Relationship: SPDXRef-fileB DESCRIBED_BY SPDXRef-DOCUMENT

with

SPDXVersion: SPDX-2.3
DataLicense: CC0-1.0
DocumentNamespace: https://some.namespace2
DocumentName: document name
SPDXID: SPDXRef-DOCUMENT

## Creation Information
Creator: Tool: test-tool
Created: 2022-01-01T00:00:00Z
## Relationships
Relationship: SPDXRef-DOCUMENT DESCRIBES SPDXRef-fileA
Relationship: SPDXRef-DOCUMENT DESCRIBES SPDXRef-fileB

FileName: ./fileA.c
SPDXID: SPDXRef-fileA
FileChecksum: SHA1: d6a770ba38583ed4bb4525bd96e50461655d2758
LicenseConcluded: NOASSERTION
## Relationships
Relationship: SPDXRef-fileA DEPENDS_ON SPDXRef-fileB
Relationship: SPDXRef-fileA DESCRIBED_BY SPDXRef-DOCUMENT
FileName: ./fileB.c
SPDXID: SPDXRef-fileB
FileChecksum: SHA1: d6a770ba38583ed4bb4525bd96e50461655d2758
LicenseConcluded: NOASSERTION
## Relationships
Relationship: SPDXRef-fileB DEPENDENCY_OF SPDXRef-fileA

the resulting xlsx file lists a difference in the file relationships (which is understandable as the relationship Relationship: SPDXRef-fileB DESCRIBED_BY SPDXRef-DOCUMENT is missing -although it is a duplicate of SPDXRef-DOCUMENT DESCRIBES SPDXRef-fileB). But the result also marks document describes with diff although the values in the rows below are the same. The same holds for the file relationships of ./fileA.c.
Why is this marked as diff?
In general I would expect that the comparison is somehow independant from the direction of the relationship although I see that this is a rather complex topic and I can understand that the comparison would only be about the "actual present" relationships.

Another problem, which is related to this, is the following: For json, yaml and xml there are additionally the tags "documentDescribes" and "hasFiles" in packages to represent DESCRIBES, resp. CONTAINS-relationships. The tools-python avoid duplications when serialising and do not write for example SPDXRef-DOCUMENT DESCRIBES SPDXRef-File additionally out, because this information is already mapped in "documentDescribes: [SPDXRef-File]". This leads then with the comparison however again to the fact that two documents are not evaluated as equal.

Do you think it would make sense to add some additional logic to the comparison that checks for this kind of semantic equivalence of the relationships ( so including the information from hasFilesand documentDescribes) and not only for actual existence?

Beitragsleitfaden

Beitragsleitfaden öffnen

Erste Schritte

  1. Lies das ganze Issue und danach den Beitragsleitfaden des Projekts.
  2. Schreib ins Issue, dass du es übernimmst — das erspart doppelte Arbeit.
  3. Forke das Repository und arbeite in einem Branch.
  4. Öffne einen Pull Request, der die Issue-Nummer nennt.

Rechercherichtung

Beginne mit der Vergleichsmethode in tools-java und untersuche, wie der tools-python-Serializer hasFiles und documentDescribes darstellt. Vergleiche die Verarbeitung von Beziehungen für beide Beispieldokumente, einschließlich umgekehrter oder weggelassener Beziehungseinträge. Als erledigt gilt, dass semantisch äquivalente Dokumente als gleich verglichen werden, ohne fälschlicherweise Unterschiede bei Dokument- oder Dateibeziehungen zu melden.

Vom Indexierungsmodell aus dem Issue-Text verfasst.

Bewertung

Tech-Stack
java
Bereich
tooling
Issue-Typ
Feature
Schwierigkeit
5/5
Geschätzter Aufwand
Über eine Woche
Aktivitätsstatus
Veraltet
Klarheit
Größtenteils klar
Anfängerfreundlichkeit
25/100

Neue Issues direkt in Ihr Postfach

Eine kurze Übersicht über anfängerfreundliche GitHub-Issues.