Include coverage information in the license Range data. (#1470)

Author: isoos

CHANGELOG.md

@@ -1,7 +1,7 @@
 ## 0.22.23
 
 - Upgraded SDK constraint: `^3.8.0`.
-- Updated `License` with `start` and `end` `Position` fields that identifies the recognized block.
+- Updated `License` with `Range? range` field that identifies the recognized block.
 
 ## 0.22.22
 

lib/src/license.dart

@@ -8,6 +8,7 @@ import 'dart:io';
 
 import 'package:meta/meta.dart';
 import 'package:path/path.dart' as p;
+import 'package:source_span/source_span.dart';
 
 import 'license_detection/license_detector.dart' hide License, Range;
 import 'model.dart';
@@ -41,6 +42,11 @@ Future<List<License>> detectLicenseInFile(
   return licenses;
 }
 
+/// Characters and expression that are accepted as non-relevant range gaps,
+/// and consecutive [Range] values can be merged if they have only these
+/// between.
+final _rangeMergeRegexp = RegExp(r'^[\s\.\-\(\)\*]+$');
+
 /// Returns the license(s) detected from the [SPDX-corpus][1].
 ///
 /// [1]: https://spdx.org/licenses/
@@ -55,22 +61,52 @@ Future<List<License>> detectLicenseInContent(
     return <License>[];
   }
 
+  List<int> buildCoverages(LicenseMatch match) {
+    final ranges = <({int start, int end})>[];
+    // ignore: invalid_use_of_visible_for_testing_member
+    for (final token in match.tokens) {
+      // check to merge into last range
+      final last = ranges.lastOrNull;
+      if (last != null) {
+        var mergeWithLast = false;
+        if (last.end == token.span.start.offset) {
+          mergeWithLast = true;
+        } else {
+          final textBetween = content.substring(
+            last.end,
+            token.span.start.offset,
+          );
+          if (_rangeMergeRegexp.matchAsPrefix(textBetween) != null) {
+            mergeWithLast = true;
+          }
+        }
+        if (mergeWithLast) {
+          ranges[ranges.length - 1] = (
+            start: last.start,
+            end: token.span.end.offset,
+          );
+          continue;
+        }
+      }
+      // fallback: start a new range
+      ranges.add((start: token.span.start.offset, end: token.span.end.offset));
+    }
+    return ranges.expand((e) => [e.start, e.end]).toList();
+  }
+
   return licenseResult.matches.map((e) {
     return License(
       path: relativePath,
       spdxIdentifier: e.identifier,
       range: Range(
-        start: Position(
-          offset: e.start.offset,
-          line: e.start.line,
-          column: e.start.column,
-        ),
-        end: Position(
-          offset: e.end.offset,
-          line: e.end.line,
-          column: e.end.column,
-        ),
+        start: e.start.toPosition(),
+        end: e.end.toPosition(),
+        coverages: buildCoverages(e),
       ),
     );
   }).toList();
 }
+
+extension on SourceLocation {
+  Position toPosition() => Position(offset: offset, line: line, column: column);
+}

lib/src/model.dart

@@ -191,7 +191,11 @@ class Range {
   final Position start;
   final Position end;
 
-  Range({required this.start, required this.end});
+  /// Pair of values that indicate start-end offsets that inside the range are
+  /// covering the license text.
+  final List<int> coverages;
+
+  Range({required this.start, required this.end, required this.coverages});
 
   factory Range.fromJson(Map<String, dynamic> json) => _$RangeFromJson(json);
 

lib/src/model.g.dart

@@ -110,7 +110,16 @@ void main() {
             .replaceAll(
               RegExp('that was published [0-9]+ days ago'),
               'that was published N days ago',
-            );
+            )
+            .replaceAllMapped(RegExp(r'"coverages":\[(\d+(\,\d+)+)\]'), (m) {
+              final parts = m.group(1)!.split(',');
+              final remaning = parts.indexed
+                  .where((p) => p.$1 < 6 || p.$1 >= parts.length - 6)
+                  .map((p) => p.$2)
+                  .join(',');
+              return '"coverages":[$remaning]';
+            });
+
         actualMap = json.decode(updated) as Map<String, Object?>;
       });
 

test/end2end_test.dart

@@ -34,7 +34,13 @@
           "offset": 1523,
           "line": 25,
           "column": 68
-        }
+        },
+        "coverages": [
+          63,
+          836,
+          838,
+          1523
+        ]
       }
     }
   ],
@@ -107,7 +113,13 @@
             "offset": 1523,
             "line": 25,
             "column": 68
-          }
+          },
+          "coverages": [
+            63,
+            836,
+            838,
+            1523
+          ]
         }
       }
     ],

test/goldens/end2end/_dummy_pkg-1.0.0-null-safety.1.json

@@ -44,7 +44,13 @@
           "offset": 1503,
           "line": 26,
           "column": 68
-        }
+        },
+        "coverages": [
+          44,
+          816,
+          818,
+          1503
+        ]
       }
     }
   ],
@@ -145,7 +151,13 @@
             "offset": 1503,
             "line": 26,
             "column": 68
-          }
+          },
+          "coverages": [
+            44,
+            816,
+            818,
+            1503
+          ]
         }
       }
     ],

test/goldens/end2end/async-2.11.0.json

@@ -84,7 +84,21 @@
           "offset": 1036,
           "line": 19,
           "column": 9
-        }
+        },
+        "coverages": [
+          14,
+          24,
+          25,
+          27,
+          28,
+          34,
+          1011,
+          1019,
+          1020,
+          1022,
+          1023,
+          1036
+        ]
       }
     }
   ],
@@ -225,7 +239,21 @@
             "offset": 1036,
             "line": 19,
             "column": 9
-          }
+          },
+          "coverages": [
+            14,
+            24,
+            25,
+            27,
+            28,
+            34,
+            1011,
+            1019,
+            1020,
+            1022,
+            1023,
+            1036
+          ]
         }
       }
     ],

test/goldens/end2end/audio_service-0.18.17.json

@@ -35,7 +35,21 @@
           "offset": 1046,
           "line": 19,
           "column": 13
-        }
+        },
+        "coverages": [
+          24,
+          34,
+          35,
+          37,
+          38,
+          44,
+          1015,
+          1029,
+          1030,
+          1032,
+          1033,
+          1046
+        ]
       }
     }
   ],
@@ -115,7 +129,21 @@
             "offset": 1046,
             "line": 19,
             "column": 13
-          }
+          },
+          "coverages": [
+            24,
+            34,
+            35,
+            37,
+            38,
+            44,
+            1015,
+            1029,
+            1030,
+            1032,
+            1033,
+            1046
+          ]
         }
       }
     ],

test/goldens/end2end/bulma_min-0.7.4.json

@@ -40,7 +40,15 @@
           "offset": 1097,
           "line": 20,
           "column": 9
-        }
+        },
+        "coverages": [
+          59,
+          193,
+          195,
+          655,
+          657,
+          1097
+        ]
       }
     }
   ],
@@ -129,7 +137,15 @@
             "offset": 1097,
             "line": 20,
             "column": 9
-          }
+          },
+          "coverages": [
+            59,
+            193,
+            195,
+            655,
+            657,
+            1097
+          ]
         }
       }
     ],