OutputHTML: Fix License Reference Table content

AUTHORS.rst

@@ -35,6 +35,7 @@ The following organizations or individuals have contributed to ScanCode:
 - Horie Issei @is2ei
 - James Ward @jamesward
 - Jelmer Vernooij @jelmer
+- Jens Keim @pepper-jk
 - Kunal Chhabra @kunalchhabra37
 - Jillian Daguil @jdaguil
 - Jiri Popelka @jpopelka

CHANGELOG.rst

@@ -1,6 +1,10 @@
 Changelog
 =========
 
+- Fix License Reference table in HTML Output by reverting to the
+  behavior of v32.0.0.
+  See https://github.com/aboutcode-org/scancode-toolkit/issues/4101
+
 - We now support new package manifest formats:
 
   - OpenWRT packages.

src/formattedcode/output_html.py

@@ -188,9 +188,13 @@ def generate_output(results, license_references, version, template):
     """
     # FIXME: This code is highly coupled with actual scans and may not
     # support adding new scans at all
+
+    from licensedcode.cache import get_licenses_db
+
     converted = {}
     converted_infos = {}
     converted_packages = {}
+    licenses = {}
 
     LICENSES = 'license_detections'
     COPYRIGHTS = 'copyrights'
@@ -223,6 +227,11 @@ def generate_output(results, license_references, version, template):
                     'value': license_expression,
                 })
 
+                if license_expression not in licenses:
+                    license_object = get_licenses_db().get(license_expression)
+                    if license_object != None:
+                        licenses[license_expression] = license_object
+
         if results:
             converted[path] = sorted(results, key=itemgetter('start'))
 
@@ -239,6 +248,9 @@ def generate_output(results, license_references, version, template):
         if PACKAGES in scanned_file:
             converted_packages[path] = scanned_file[PACKAGES]
 
+    licenses = dict(sorted(licenses.items()))
+    license_references = list(licenses.values())
+
     files = {
         'license_copyright': converted,
         'infos': converted_infos,