Handle licenses with empty license test in SPDX library

Signed-off-by: tdruez <[email protected]>

Author: tdruez

scanpipe/pipes/spdx.py

@@ -280,9 +280,14 @@ class ExtractedLicensingInfo:
 
     def as_dict(self):
         """Return the data as a serializable dict."""
+        if self.extracted_text.strip():
+            extracted_text = self.extracted_text
+        else:
+            extracted_text = "NOASSERTION"
+
         required_data = {
             "licenseId": self.license_id,
-            "extractedText": self.extracted_text,
+            "extractedText": extracted_text,
         }
 
         optional_data = {

scanpipe/tests/pipes/test_spdx.py

@@ -318,6 +318,15 @@ def test_spdx_extracted_licensing_info_as_dict(self):
         licensing_info = spdx.ExtractedLicensingInfo(**self.licensing_info_data)
         assert self.licensing_info_spdx_data == licensing_info.as_dict()
 
+    def test_spdx_extracted_licensing_info_empty_extracted_text(self):
+        licensing_info = spdx.ExtractedLicensingInfo(
+            **{
+                "license_id": "LicenseRef-1",
+                "extracted_text": " ",
+            }
+        )
+        assert "NOASSERTION" == licensing_info.as_dict()["extractedText"]
+
     def test_spdx_extracted_licensing_info_from_data(self):
         assert spdx.ExtractedLicensingInfo.from_data({})
         licensing_info = spdx.ExtractedLicensingInfo.from_data(