Analyzer improvements

analyzer/build.gradle.kts

@@ -49,6 +49,7 @@ dependencies {
     api(project(":model"))
 
     implementation(project(":downloader"))
+    implementation(project(":spdx-utils"))
     implementation(project(":utils"))
 
     implementation("com.fasterxml.jackson.module:jackson-module-kotlin:$jacksonVersion")

analyzer/src/main/kotlin/Analyzer.kt

@@ -71,7 +71,7 @@ class Analyzer(private val config: AnalyzerConfiguration) {
 
         log.debug { "Using the following configuration settings:\n$repositoryConfiguration" }
 
-        // Map files by the package manager factory that manages them.
+        // Associate files by the package manager factory that manages them.
         val factoryFiles = if (packageManagers.size == 1 && absoluteProjectPath.isFile) {
             // If only one package manager is activated, treat the given path as definition file for that package
             // manager despite its name.
@@ -80,6 +80,7 @@ class Analyzer(private val config: AnalyzerConfiguration) {
             PackageManager.findManagedFiles(absoluteProjectPath, packageManagers).toMutableMap()
         }
 
+        // Associate mapped files by the package manager that manages them.
         val managedFiles = factoryFiles.mapNotNull { (factory, files) ->
             val manager = factory.create(absoluteProjectPath, config, repositoryConfiguration)
             val mappedFiles = manager.mapDefinitionFiles(files)

analyzer/src/main/kotlin/PackageManager.kt

@@ -42,6 +42,7 @@ import org.ossreviewtoolkit.model.VcsType
 import org.ossreviewtoolkit.model.config.AnalyzerConfiguration
 import org.ossreviewtoolkit.model.config.RepositoryConfiguration
 import org.ossreviewtoolkit.model.createAndLogIssue
+import org.ossreviewtoolkit.spdx.VCS_DIRECTORIES
 import org.ossreviewtoolkit.utils.collectMessagesAsString
 import org.ossreviewtoolkit.utils.isSymbolicLink
 import org.ossreviewtoolkit.utils.log
@@ -70,13 +71,7 @@ abstract class PackageManager(
          */
         val ALL by lazy { LOADER.iterator().asSequence().toList() }
 
-        private val IGNORED_DIRECTORY_MATCHERS = listOf(
-            // Ignore VCS configuration directories.
-            ".git",
-            ".hg",
-            ".repo",
-            ".svn",
-            "CVS",
+        private val PACKAGE_MANAGER_DIRECTORIES = listOf(
             // Ignore intermediate build system directories.
             ".gradle",
             "node_modules",
@@ -86,7 +81,9 @@ abstract class PackageManager(
             // Ignore virtual environments in Python.
             "lib/python2.*/dist-packages",
             "lib/python3.*/site-packages"
-        ).map {
+        )
+
+        private val IGNORED_DIRECTORY_MATCHERS = (VCS_DIRECTORIES + PACKAGE_MANAGER_DIRECTORIES).map {
             FileSystems.getDefault().getPathMatcher("glob:**/$it")
         }
 

analyzer/src/main/kotlin/managers/Unmanaged.kt

@@ -34,7 +34,9 @@ import org.ossreviewtoolkit.model.config.RepositoryConfiguration
 import org.ossreviewtoolkit.utils.log
 
 /**
- * A fake [PackageManager] for projects that do not use any of the known package managers.
+ * A fake [PackageManager] for projects that do not use any of the known package managers, or no package manager at all.
+ * It is required as in ORT's data model e.g. scan results need to be attached to projects (or packages), so files that
+ * do not belong to any other project need to be attached to somewhere.
  */
 class Unmanaged(
     name: String,

spdx-utils/src/main/kotlin/SpdxUtils.kt

@@ -42,7 +42,7 @@ val NON_LICENSE_FILENAMES = listOf(
 /**
  * A list of directories used by version control systems to store metadata.
  */
-internal val VCS_DIRECTORIES = listOf(
+val VCS_DIRECTORIES = listOf(
     ".git",
     ".hg",
     ".repo",