Omit Detected licenses already in License

src/fosslight_source/_parsing_scancode_file_item.py

@@ -446,12 +446,44 @@ def _dedupe_detected_comment_expressions(expressions: list[str]) -> list[str]:
     return kept
 
 
-def build_detected_comment_from_dropped_matches(dropped_matches: list) -> str:
-    """Build ``Detected: a, b OR c`` from matches dropped by SPDX priority."""
-    displays = [
-        _display_expression_for_detected_comment(match)
-        for match in (dropped_matches or [])
+def _expression_covered_by_reported_licenses(
+    expression: str, reported_licenses: list
+) -> bool:
+    """
+    True when expression is already represented in License column values.
+
+    Matches the whole expression, or every AND/OR token, case-insensitively.
+    """
+    reported = {
+        (lic or "").strip().lower()
+        for lic in (reported_licenses or [])
+        if (lic or "").strip()
+    }
+    if not expression or not reported:
+        return False
+    expr = expression.strip()
+    if expr.lower() in reported:
+        return True
+    tokens = [
+        token.strip().lower()
+        for token in split_spdx_expression(expr)
+        if token and token.strip()
     ]
+    return bool(tokens) and all(token in reported for token in tokens)
+
+
+def build_detected_comment_from_dropped_matches(
+    dropped_matches: list, reported_licenses: list | None = None
+) -> str:
+    """Build ``Detected: a, b OR c`` from matches dropped by SPDX priority."""
+    displays = []
+    for match in (dropped_matches or []):
+        display = _display_expression_for_detected_comment(match)
+        if not display:
+            continue
+        if _expression_covered_by_reported_licenses(display, reported_licenses or []):
+            continue
+        displays.append(display)
     deduped = _dedupe_detected_comment_expressions(displays)
     if not deduped:
         return ""
@@ -840,7 +872,7 @@ def parsing_scancode(
                         match for match in all_matches if id(match) not in process_ids
                     ]
                     detected_comment = build_detected_comment_from_dropped_matches(
-                        dropped_matches
+                        dropped_matches, result_item.licenses
                     )
                     if detected_comment:
                         result_item.comment = detected_comment

tests/test_parsing_unknown_spdx.py

@@ -703,6 +703,110 @@ def test_dedupe_detected_comment_examples(dropped_exprs, expected_comment):
     assert build_detected_comment_from_dropped_matches(dropped) == expected_comment
 
 
+def test_detected_comment_excludes_licenses_already_in_license_column():
+    from fosslight_source._parsing_scancode_file_item import (
+        build_detected_comment_from_dropped_matches,
+    )
+    dropped = [
+        {
+            "license_expression": "apache-2.0",
+            "license_expression_spdx": "Apache-2.0",
+            "matched_text": "Licensed under the Apache License, Version 2.0",
+        },
+        {
+            "license_expression": "mit",
+            "license_expression_spdx": "MIT",
+            "matched_text": "Permission is hereby granted",
+        },
+    ]
+    assert (
+        build_detected_comment_from_dropped_matches(dropped, ["Apache-2.0"])
+        == "Detected: MIT"
+    )
+    assert build_detected_comment_from_dropped_matches(dropped, ["Apache-2.0", "MIT"]) == ""
+
+
+def test_detected_comment_uses_final_licenses_after_column_limit():
+    """License dropped by column length limit must still be allowed in Detected."""
+    fillers = [f"AaaLic{i:02d}-{'y' * 80}" for i in range(20)]
+    spdx_matches = [
+        {
+            "license_expression": name.lower(),
+            "license_expression_spdx": name,
+            "matched_text": f"SPDX-License-Identifier: {name}",
+        }
+        for name in fillers + ["ZebraExtra"]
+    ]
+    scancode_file_list = [{
+        "path": "crowded.c",
+        "type": "file",
+        "detected_license_expression": " AND ".join(
+            name.lower() for name in fillers + ["ZebraExtra"]
+        ),
+        "license_detections": [
+            {"matches": spdx_matches},
+            {
+                "matches": [{
+                    "license_expression": "zebraextra",
+                    "license_expression_spdx": "ZebraExtra",
+                    "matched_text": "This file is under ZebraExtra terms.",
+                }],
+            },
+        ],
+        "copyrights": [],
+    }]
+
+    success, results, _messages, _ = parsing_scancode(scancode_file_list)
+
+    assert success is True
+    assert "ZebraExtra" not in results[0].licenses
+    assert "Detected: ZebraExtra" in results[0].comment
+
+
+def test_readme_spdx_apache_does_not_repeat_in_detected_comment():
+    """Body apache-2.0 match dropped by SPDX priority must not echo License column."""
+    scancode_file_list = [{
+        "path": "README.md",
+        "type": "file",
+        "detected_license_expression": "apache-2.0",
+        "detected_license_expression_spdx": "Apache-2.0",
+        "license_detections": [
+            {
+                "matches": [{
+                    "license_expression": "apache-2.0",
+                    "license_expression_spdx": "Apache-2.0",
+                    "matched_text": "SPDX-License-Identifier: Apache-2.0",
+                }],
+            },
+            {
+                "matches": [
+                    {
+                        "license_expression": "apache-2.0",
+                        "license_expression_spdx": "Apache-2.0",
+                        "matched_text": (
+                            'Licensed under the Apache License, Version 2.0 '
+                            '(the "License"); you may not use this file except '
+                            "in compliance with the License."
+                        ),
+                    },
+                    {
+                        "license_expression": "apache-2.0",
+                        "license_expression_spdx": "Apache-2.0",
+                        "matched_text": "SPDX-License-Identifier: Apache-2.0",
+                    },
+                ],
+            },
+        ],
+        "copyrights": [],
+    }]
+
+    success, results, _messages, _ = parsing_scancode(scancode_file_list)
+
+    assert success is True
+    assert results[0].licenses == ["Apache-2.0"]
+    assert results[0].comment == ""
+
+
 def test_spdx_priority_detected_comment_uses_spdx_ids():
     """Dropped body matches use license_expression_spdx in Detected comment."""
     scancode_file_list = [{