refactor(Repository): Remove vcs attributes

Remove `vcs` and `vcsProcessed` attributes from `Repository`, since
the `RepositoryProvenance` `provenance` is now the main attribute.

Add `RepositoryDeserializer` to support parsing of legacy `OrtResult` files
containing the previous attributes `vcs` and `vcsProcessed`, as well as the
new format. It also needs to handle the `nestedRepositories` and `config`
attributes for both legacy and recent `OrtResult` files.

Signed-off-by: Jens Keim <[email protected]>

Author: pepper-jk

cli-helper/src/main/kotlin/utils/Extensions.kt

@@ -716,7 +716,7 @@ internal fun OrtResult.getScanResultFor(packageConfiguration: PackageConfigurati
  * tree.
  */
 internal fun OrtResult.getRepositoryPaths(): Map<String, Set<String>> {
-    val result = mutableMapOf(repository.vcsProcessed.url to mutableSetOf(""))
+    val result = mutableMapOf(repository.provenance.vcsInfo.normalize().url to mutableSetOf(""))
 
     repository.nestedRepositories.mapValues { (path, vcsInfo) ->
         result.getOrPut(vcsInfo.url) { mutableSetOf() } += path

evaluator/src/main/kotlin/ProjectSourceRule.kt

@@ -93,7 +93,7 @@ open class ProjectSourceRule(
     /**
      * Return the [VcsType] of the project's code repository.
      */
-    fun projectSourceGetVcsType(): VcsType = ortResult.repository.vcsProcessed.type
+    fun projectSourceGetVcsType(): VcsType = ortResult.repository.provenance.vcsInfo.normalize().type
 
     /**
      * Return the file paths matching any of the given [glob expressions][patterns] with its file content matching

evaluator/src/main/kotlin/RuleSet.kt

@@ -159,7 +159,7 @@ fun ruleSet(
     licenseInfoResolver: LicenseInfoResolver = ortResult.createLicenseInfoResolver(),
     resolutionProvider: ResolutionProvider = DefaultResolutionProvider.create(),
     projectSourceResolver: SourceTreeResolver = SourceTreeResolver.forRemoteRepository(
-        ortResult.repository.vcsProcessed
+        ortResult.repository.provenance.vcsInfo.normalize()
     ),
     configure: RuleSet.() -> Unit = {}
 ) = RuleSet(ortResult, licenseInfoResolver, resolutionProvider, projectSourceResolver).apply(configure)

model/src/main/kotlin/Repository.kt

@@ -20,30 +20,26 @@
 package org.ossreviewtoolkit.model
 
 import com.fasterxml.jackson.annotation.JsonInclude
+import com.fasterxml.jackson.core.JsonParser
+import com.fasterxml.jackson.databind.DeserializationContext
+import com.fasterxml.jackson.databind.JsonNode
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize
+import com.fasterxml.jackson.databind.deser.std.StdDeserializer
+import com.fasterxml.jackson.module.kotlin.treeToValue
 
 import org.ossreviewtoolkit.model.config.RepositoryConfiguration
 import org.ossreviewtoolkit.utils.ort.ORT_REPO_CONFIG_FILENAME
 
 /**
  * A description of the source code repository that was used as input for ORT.
  */
+@JsonDeserialize(using = RepositoryDeserializer::class)
 data class Repository(
     /**
      * Provenance wrapper for original VCS information, if present.
      */
     val provenance: RepositoryProvenance,
 
-    /**
-     * Original VCS-related information from the working tree containing the analyzer root.
-     */
-    val vcs: VcsInfo = provenance.vcsInfo,
-
-    /**
-     * Processed VCS-related information from the working tree containing the analyzer root that has e.g. common
-     * mistakes corrected.
-     */
-    val vcsProcessed: VcsInfo = vcs.normalize(),
-
     /**
      * A map of nested repositories, for example Git submodules or Git-Repo modules. The key is the path to the
      * nested repository relative to the root of the main repository.
@@ -66,8 +62,6 @@ data class Repository(
                 vcsInfo = VcsInfo.EMPTY,
                 resolvedRevision = HashAlgorithm.SHA1.emptyValue
             ),
-            vcs = VcsInfo.EMPTY,
-            vcsProcessed = VcsInfo.EMPTY,
             nestedRepositories = emptyMap(),
             config = RepositoryConfiguration()
         )
@@ -82,8 +76,57 @@ data class Repository(
 
         val normalizedVcs = vcs.normalize()
 
-        if (vcsProcessed.matches(normalizedVcs)) return ""
+        if (provenance.vcsInfo.normalize().matches(normalizedVcs)) return ""
 
         return nestedRepositories.entries.find { (_, nestedVcs) -> nestedVcs.normalize().matches(normalizedVcs) }?.key
     }
 }
+
+/**
+ * A custom deserializer for [Repository] to support the legacy [vcs] and [vcsProcessed] attributes.
+ */
+private class RepositoryDeserializer : StdDeserializer<Repository>(Repository::class.java) {
+    override fun deserialize(p: JsonParser, ctxt: DeserializationContext): Repository {
+        val node = p.codec.readTree<JsonNode>(p)
+        val parsedProvenance = when {
+            node.has("vcs") -> {
+                // Parse [vcs] and [vcsProcessed] attributes.
+                val vcs = jsonMapper.treeToValue<VcsInfo>(node["vcs"])
+                val vcsProcess = jsonMapper.treeToValue<VcsInfo>(node["vcs_processed"])
+
+                // Fall back to [vcsProcessed], if [vcs] is empty.
+                val vcsInfo = if (vcs != VcsInfo.EMPTY) vcs else vcsProcess
+
+                // Get the [vcs]'s revision.
+                // Fall back to [vcsProcessed], if [vcs] has empty revision.
+                val resolvedRevision = vcs.revision.ifEmpty {
+                    vcsProcess.revision.ifEmpty {
+                        HashAlgorithm.SHA1.emptyValue
+                    }
+                }
+
+                // Build a RepositoryProvenance from the parsed VcsInfo fields.
+                RepositoryProvenance(vcsInfo, resolvedRevision)
+            }
+
+            else -> {
+                // Parse the [provenance], if no legacy fields are present.
+                jsonMapper.treeToValue<RepositoryProvenance>(node["provenance"])
+            }
+        }
+
+        val nestedRepositories = if (node.has("nested_repositories")) {
+            jsonMapper.treeToValue<Map<String, VcsInfo>>(node["nested_repositories"])
+        } else {
+            emptyMap()
+        }
+
+        val config = if (node.has("config")) {
+            jsonMapper.treeToValue<RepositoryConfiguration>(node["config"])
+        } else {
+            RepositoryConfiguration()
+        }
+
+        return Repository(provenance = parsedProvenance, nestedRepositories, config)
+    }
+}

model/src/test/kotlin/OrtResultTest.kt

@@ -379,8 +379,10 @@ class OrtResultTest : WordSpec({
 
             val ortResult = OrtResult.EMPTY.copy(
                 repository = Repository.EMPTY.copy(
-                    vcs = vcs,
-                    vcsProcessed = vcs,
+                    provenance = RepositoryProvenance(
+                        vcsInfo = vcs,
+                        resolvedRevision = vcs.revision
+                    ),
                     config = RepositoryConfiguration(
                         excludes = Excludes(
                             paths = listOf(

plugins/reporters/cyclonedx/src/main/kotlin/CycloneDxReporter.kt

@@ -154,7 +154,7 @@ class CycloneDxReporter(
                         // There is no component type for repositories.
                         type = Component.Type.FILE
 
-                        with(input.ortResult.repository.vcsProcessed) {
+                        with(input.ortResult.repository.provenance.vcsInfo.normalize()) {
                             bomRef = "$url@$revision"
 
                             name = url
@@ -178,7 +178,7 @@ class CycloneDxReporter(
             // distributable), just create a single BOM for all projects in that case for now. As there also is no
             // single correct project to pick for adding external references in that case, simply only use the global
             // repository VCS information here.
-            val vcs = input.ortResult.repository.vcsProcessed
+            val vcs = input.ortResult.repository.provenance.vcsInfo.normalize()
             bom.addExternalReference(
                 ExternalReference.Type.VCS,
                 vcs.url,

plugins/reporters/opossum/src/main/kotlin/OpossumReporter.kt

@@ -127,7 +127,7 @@ class OpossumReporter(
         private val externalAttributionSources: MutableMap<String, OpossumExternalAttributionSource> = mutableMapOf()
 
         internal fun create(input: ReporterInput, maxDepth: Int = Int.MAX_VALUE): OpossumInput {
-            addBaseUrl("/", input.ortResult.repository.vcs)
+            addBaseUrl("/", input.ortResult.repository.provenance.vcsInfo)
 
             SpdxLicense.entries.forEach {
                 val licenseText = input.licenseTextProvider.getLicenseText(it.id)

plugins/reporters/spdx/src/testFixtures/kotlin/TestData.kt

@@ -71,8 +71,7 @@ val ORT_RESULT = OrtResult(
                 )
             )
         ),
-        provenance = ANALYZED_PROVENANCE,
-        vcsProcessed = ANALYZED_VCS
+        provenance = ANALYZED_PROVENANCE
     ),
     analyzer = AnalyzerRun.EMPTY.copy(
         result = AnalyzerResult(

plugins/reporters/static-html/src/main/kotlin/TablesReportModelMapper.kt

@@ -48,7 +48,7 @@ internal object TablesReportModelMapper {
             .sortedBy { it.id }
 
         return TablesReport(
-            input.ortResult.repository.vcsProcessed,
+            input.ortResult.repository.provenance.vcsInfo.normalize(),
             input.ortResult.repository.config,
             ruleViolations,
             getAnalyzerIssueSummaryTable(input),

scanner/src/funTest/kotlin/scanners/ScannerIntegrationFunTest.kt

@@ -36,6 +36,7 @@ import org.ossreviewtoolkit.model.PackageReference
 import org.ossreviewtoolkit.model.PackageType
 import org.ossreviewtoolkit.model.Project
 import org.ossreviewtoolkit.model.Repository
+import org.ossreviewtoolkit.model.RepositoryProvenance
 import org.ossreviewtoolkit.model.ScanSummary
 import org.ossreviewtoolkit.model.Scope
 import org.ossreviewtoolkit.model.TextLocation
@@ -167,8 +168,10 @@ private fun createAnalyzerResultWithProject(project: Project, vararg packages: P
     return OrtResult.EMPTY.copy(
         analyzer = analyzerRun,
         repository = Repository.EMPTY.copy(
-            vcsProcessed = projectWithScope.vcsProcessed,
-            vcs = projectWithScope.vcs
+            provenance = RepositoryProvenance(
+                vcsInfo = projectWithScope.vcs,
+                resolvedRevision = projectWithScope.vcs.revision
+            )
         )
     )
 }