model: Use an unsorted set for storing authors

Continuing the story from [1], only sort authors on serialization. [1]: #6244 Signed-off-by: Sebastian Schuberth <sschuberth@gmail.com>
oss-review-toolkit · Jan 2, 2023 · 691ef43 · 691ef43
1 parent 2d29748
commit 691ef43
Show file tree

Hide file tree

Showing 30 changed files with 77 additions and 73 deletions.
diff --git a/analyzer/src/funTest/kotlin/managers/SpdxDocumentFileFunTest.kt b/analyzer/src/funTest/kotlin/managers/SpdxDocumentFileFunTest.kt
@@ -103,7 +103,7 @@ class SpdxDocumentFileFunTest : WordSpec({
                     id = curlId,
                     cpe = "cpe:2.3:a:http:curl:7.70.0:*:*:*:*:*:*:*",
                     definitionFilePath = vcsDir.getPathToRoot(curlPackageFile),
-                    authors = sortedSetOf("Daniel Stenberg (daniel@haxx.se)"),
+                    authors = setOf("Daniel Stenberg (daniel@haxx.se)"),
                     declaredLicenses = sortedSetOf("curl"),
                     vcs = VcsInfo(
                         type = VcsType.GIT,
@@ -124,7 +124,7 @@ class SpdxDocumentFileFunTest : WordSpec({
                     id = opensslId,
                     cpe = "cpe:2.3:a:a-name:openssl:1.1.1g:*:*:*:*:*:*:*",
                     definitionFilePath = vcsDir.getPathToRoot(opensslPackageFile),
-                    authors = sortedSetOf("OpenSSL Development Team"),
+                    authors = setOf("OpenSSL Development Team"),
                     declaredLicenses = sortedSetOf("Apache-2.0"),
                     vcs = VcsInfo(
                         type = VcsType.GIT,
@@ -145,7 +145,7 @@ class SpdxDocumentFileFunTest : WordSpec({
                     id = zlibId,
                     cpe = "cpe:/a:compress:zlib:1.2.11:::en-us",
                     definitionFilePath = vcsDir.getPathToRoot(zlibPackageFile),
-                    authors = sortedSetOf("Jean-loup Gailly", "Mark Adler"),
+                    authors = setOf("Jean-loup Gailly", "Mark Adler"),
                     declaredLicenses = sortedSetOf("Zlib"),
                     vcs = VcsInfo(
                         type = VcsType.GIT,

diff --git a/analyzer/src/main/kotlin/managers/Bower.kt b/analyzer/src/main/kotlin/managers/Bower.kt
@@ -101,8 +101,8 @@ class Bower(
          * there are two formats to specify the authors of a package (similar to NPM). The difference is that the
          * strings or objects are inside an array.
          */
-        private fun parseAuthors(node: JsonNode): SortedSet<String> =
-            sortedSetOf<String>().apply {
+        private fun parseAuthors(node: JsonNode): Set<String> =
+            mutableSetOf<String>().apply {
                 node["pkgMeta"]["authors"]?.mapNotNull { authorNode ->
                     when {
                         authorNode.isObject -> authorNode["name"]?.textValue()

diff --git a/analyzer/src/main/kotlin/managers/Bundler.kt b/analyzer/src/main/kotlin/managers/Bundler.kt
@@ -319,7 +319,7 @@ class Bundler(
             name = workingDir.name,
             version = "",
             homepageUrl = "",
-            authors = sortedSetOf(),
+            authors = emptySet(),
             declaredLicenses = sortedSetOf(),
             description = "",
             runtimeDependencies = emptySet(),
@@ -394,7 +394,7 @@ internal data class GemSpec(
     val name: String,
     val version: String,
     val homepageUrl: String,
-    val authors: SortedSet<String>,
+    val authors: Set<String>,
     val declaredLicenses: SortedSet<String>,
     val description: String,
     val runtimeDependencies: Set<String>,

diff --git a/analyzer/src/main/kotlin/managers/Cargo.kt b/analyzer/src/main/kotlin/managers/Cargo.kt
@@ -345,5 +345,5 @@ private fun getResolvedVersion(
 /**
  * Parse information about authors from the given [node] with package metadata.
  */
-private fun parseAuthors(node: JsonNode?): SortedSet<String> =
-    node?.mapNotNullTo(sortedSetOf()) { parseAuthorString(it.textValue()) } ?: sortedSetOf()
+private fun parseAuthors(node: JsonNode?): Set<String> =
+    node?.mapNotNullTo(mutableSetOf()) { parseAuthorString(it.textValue()) } ?: emptySet()
diff --git a/analyzer/src/main/kotlin/managers/Carthage.kt b/analyzer/src/main/kotlin/managers/Carthage.kt
@@ -81,7 +81,7 @@ class Carthage(
                         version = projectInfo.revision.orEmpty()
                     ),
                     definitionFilePath = VersionControlSystem.getPathInfo(definitionFile).path,
-                    authors = sortedSetOf(),
+                    authors = emptySet(),
                     declaredLicenses = sortedSetOf(),
                     vcs = VcsInfo.EMPTY,
                     vcsProcessed = processProjectVcs(workingDir, VcsInfo.EMPTY),
@@ -189,7 +189,7 @@ class Carthage(
                 name = vcsHost?.getProject(projectUrl).orEmpty(),
                 version = revision
             ),
-            authors = sortedSetOf(),
+            authors = emptySet(),
             declaredLicenses = sortedSetOf(),
             description = "",
             homepageUrl = projectUrl.removeSuffix(".git"),
@@ -213,7 +213,7 @@ class Carthage(
                 name = fileUrl.substringAfterLast("/"),
                 version = revision
             ),
-            authors = sortedSetOf(),
+            authors = emptySet(),
             declaredLicenses = sortedSetOf(),
             description = "",
             homepageUrl = "",
@@ -231,7 +231,7 @@ class Carthage(
                 name = id.substringAfterLast("/").removeSuffix(".json"),
                 version = revision
             ),
-            authors = sortedSetOf(),
+            authors = emptySet(),
             declaredLicenses = sortedSetOf(),
             description = "",
             homepageUrl = "",

diff --git a/analyzer/src/main/kotlin/managers/CocoaPods.kt b/analyzer/src/main/kotlin/managers/CocoaPods.kt
@@ -159,7 +159,7 @@ class CocoaPods(
                     version = ""
                 ),
                 definitionFilePath = VersionControlSystem.getPathInfo(definitionFile).path,
-                authors = sortedSetOf(),
+                authors = emptySet(),
                 declaredLicenses = sortedSetOf(),
                 vcs = VcsInfo.EMPTY,
                 vcsProcessed = processProjectVcs(workingDir),
@@ -185,7 +185,7 @@ class CocoaPods(
 
         return Package(
             id = id,
-            authors = sortedSetOf(),
+            authors = emptySet(),
             declaredLicenses = podspec.license.takeUnless { it.isEmpty() }?.let { sortedSetOf(it) } ?: sortedSetOf(),
             description = podspec.summary,
             homepageUrl = podspec.homepage,

diff --git a/analyzer/src/main/kotlin/managers/Conan.kt b/analyzer/src/main/kotlin/managers/Conan.kt
@@ -462,6 +462,6 @@ class Conan(
      * Parse information about the package author from the given JSON [node]. If present, return a set containing the
      * author name; otherwise, return an empty set.
      */
-    private fun parseAuthors(node: JsonNode): SortedSet<String> =
-        parseAuthorString(node["author"]?.textValue(), '<', '(')?.let { sortedSetOf(it) } ?: sortedSetOf()
+    private fun parseAuthors(node: JsonNode): Set<String> =
+        parseAuthorString(node["author"]?.textValue(), '<', '(')?.let { setOf(it) } ?: emptySet()
 }
diff --git a/analyzer/src/main/kotlin/managers/GoDep.kt b/analyzer/src/main/kotlin/managers/GoDep.kt
@@ -135,7 +135,7 @@ class GoDep(
 
             val pkg = Package(
                 id = Identifier("Go", "", name, normalizeModuleVersion(version)),
-                authors = sortedSetOf(),
+                authors = emptySet(),
                 declaredLicenses = sortedSetOf(),
                 description = "",
                 homepageUrl = "",
@@ -174,7 +174,7 @@ class GoDep(
                         version = projectVcs.revision
                     ),
                     definitionFilePath = VersionControlSystem.getPathInfo(definitionFile).path,
-                    authors = sortedSetOf(),
+                    authors = emptySet(),
                     declaredLicenses = sortedSetOf(),
                     vcs = VcsInfo.EMPTY,
                     vcsProcessed = projectVcs,

diff --git a/analyzer/src/main/kotlin/managers/GoMod.kt b/analyzer/src/main/kotlin/managers/GoMod.kt
@@ -133,7 +133,7 @@ class GoMod(
                             version = projectVcs.revision
                         ),
                         definitionFilePath = VersionControlSystem.getPathInfo(definitionFile).path,
-                        authors = sortedSetOf(), // Go mod doesn't support author information.
+                        authors = emptySet(), // Go mod doesn't support author information.
                         declaredLicenses = sortedSetOf(), // Go mod doesn't support declared licenses.
                         vcs = projectVcs,
                         vcsProcessed = projectVcs,
@@ -305,7 +305,7 @@ class GoMod(
 
         return Package(
             id = toId(),
-            authors = sortedSetOf(), // Go mod doesn't support author information.
+            authors = emptySet(), // Go mod doesn't support author information.
             declaredLicenses = sortedSetOf(), // Go mod doesn't support declared licenses.
             description = "",
             homepageUrl = "",

diff --git a/analyzer/src/main/kotlin/managers/Gradle.kt b/analyzer/src/main/kotlin/managers/Gradle.kt
@@ -258,7 +258,7 @@ class Gradle(
                 val project = Project(
                     id = projectId,
                     definitionFilePath = VersionControlSystem.getPathInfo(definitionFile).path,
-                    authors = sortedSetOf(),
+                    authors = emptySet(),
                     declaredLicenses = sortedSetOf(),
                     vcs = VcsInfo.EMPTY,
                     vcsProcessed = processProjectVcs(definitionFile.parentFile),

diff --git a/analyzer/src/main/kotlin/managers/Pub.kt b/analyzer/src/main/kotlin/managers/Pub.kt
@@ -473,7 +473,7 @@ class Pub(
                     var rawName = ""
                     var homepageUrl = ""
                     var vcs = VcsInfo.EMPTY
-                    var authors: SortedSet<String> = sortedSetOf<String>()
+                    var authors = mutableSetOf<String>()
 
                     when {
                         pkgInfoFromLockFile["source"].textValueOrEmpty() == "path" -> {
@@ -672,8 +672,8 @@ class Pub(
 /**
  * Extract information about package authors from the given [pubspec].
  */
-private fun parseAuthors(pubspec: JsonNode): SortedSet<String> =
-    (listOfNotNull(pubspec["author"]) + pubspec["authors"]?.toList().orEmpty()).mapNotNullTo(sortedSetOf()) {
+private fun parseAuthors(pubspec: JsonNode): MutableSet<String> =
+    (setOfNotNull(pubspec["author"]) + pubspec["authors"]?.toSet().orEmpty()).mapNotNullTo(mutableSetOf()) {
         parseAuthorString(it.textValue())
     }
 

diff --git a/analyzer/src/main/kotlin/managers/SpdxDocumentFile.kt b/analyzer/src/main/kotlin/managers/SpdxDocumentFile.kt
@@ -199,20 +199,20 @@ private fun String.sanitize(): String = replace(':', ' ').collapseWhitespace()
 /**
  * Wrap any "present" SPDX value in a sorted set, or return an empty sorted set otherwise.
  */
-private fun String?.wrapPresentInSortedSet(): SortedSet<String> {
+private fun String?.wrapPresentInSet(): Set<String> {
     if (SpdxConstants.isPresent(this)) {
         withoutPrefix(SpdxConstants.PERSON)?.let { persons ->
             // In case of a person, allow a comma-separated list of persons.
-            return persons.split(',').mapTo(sortedSetOf()) { it.trim() }
+            return persons.split(',').mapTo(mutableSetOf()) { it.trim() }
         }
 
         // Do not split an organization like "Acme, Inc." by comma.
         withoutPrefix(SpdxConstants.ORGANIZATION)?.let {
-            return sortedSetOf(it)
+            return setOf(it)
         }
     }
 
-    return sortedSetOf()
+    return setOf()
 }
 
 /**
@@ -315,7 +315,7 @@ class SpdxDocumentFile(
             id = id,
             purl = locateExternalReference(SpdxExternalReference.Type.Purl) ?: id.toPurl(),
             cpe = locateCpe(),
-            authors = originator.wrapPresentInSortedSet(),
+            authors = originator.wrapPresentInSet(),
             declaredLicenses = sortedSetOf(licenseDeclared),
             concludedLicense = getConcludedLicense(),
             description = packageDescription,
@@ -516,7 +516,7 @@ class SpdxDocumentFile(
             id = projectPackage.toIdentifier(),
             cpe = projectPackage.locateCpe(),
             definitionFilePath = VersionControlSystem.getPathInfo(definitionFile).path,
-            authors = projectPackage.originator.wrapPresentInSortedSet(),
+            authors = projectPackage.originator.wrapPresentInSet(),
             declaredLicenses = sortedSetOf(projectPackage.licenseDeclared),
             vcs = processProjectVcs(definitionFile.parentFile, VcsInfo.EMPTY),
             homepageUrl = projectPackage.homepage.mapNotPresentToEmpty(),

diff --git a/analyzer/src/main/kotlin/managers/Stack.kt b/analyzer/src/main/kotlin/managers/Stack.kt
@@ -349,7 +349,7 @@ class Stack(
                 .split(',')
                 .map(String::trim)
                 .filter(String::isNotEmpty)
-                .mapTo(sortedSetOf(), ::parseAuthorString),
+                .mapNotNullTo(mutableSetOf(), ::parseAuthorString),
             declaredLicenses = map["license"]?.let { sortedSetOf(it) } ?: sortedSetOf(),
             description = map["description"].orEmpty(),
             homepageUrl = homepageUrl,

diff --git a/analyzer/src/main/kotlin/managers/Yarn2.kt b/analyzer/src/main/kotlin/managers/Yarn2.kt
@@ -28,7 +28,6 @@ import com.fasterxml.jackson.module.kotlin.readValues
 import com.vdurmont.semver4j.Requirement
 
 import java.io.File
-import java.util.SortedSet
 
 import kotlinx.coroutines.Dispatchers
 import kotlinx.coroutines.async
@@ -742,6 +741,6 @@ class Yarn2(
         val homepage: String = "",
         val downloadUrl: String = "",
         val hash: Hash = Hash.NONE,
-        val author: SortedSet<String> = emptySet<String>().toSortedSet()
+        val author: Set<String> = emptySet()
     )
 }
diff --git a/analyzer/src/main/kotlin/managers/utils/NpmSupport.kt b/analyzer/src/main/kotlin/managers/utils/NpmSupport.kt
@@ -324,8 +324,8 @@ private fun getYarnWorkspaceSubmodules(definitionFiles: Set<File>): Set<File> {
  * https://docs.npmjs.com/files/package.json#people-fields-author-contributors, there are two formats to
  * specify the author of a package: An object with multiple properties or a single string.
  */
-fun parseNpmAuthors(json: JsonNode): SortedSet<String> =
-    sortedSetOf<String>().apply {
+fun parseNpmAuthors(json: JsonNode): Set<String> =
+    mutableSetOf<String>().apply {
         json["author"]?.let { authorNode ->
             when {
                 authorNode.isObject -> authorNode["name"]?.textValue()

diff --git a/analyzer/src/main/kotlin/managers/utils/NuGetSupport.kt b/analyzer/src/main/kotlin/managers/utils/NuGetSupport.kt
@@ -339,11 +339,11 @@ private fun parseLicenses(spec: PackageSpec?): SortedSet<String> {
 /**
  * Parse information about the authors of a package from the given [spec].
  */
-private fun parseAuthors(spec: PackageSpec?): SortedSet<String> =
+private fun parseAuthors(spec: PackageSpec?): Set<String> =
     spec?.metadata?.authors?.split(',', ';').orEmpty()
         .map(String::trim)
         .filterNot(String::isEmpty)
-        .toSortedSet()
+        .toSet()
 
 /**
  * Try to find a .nuspec file for the given [definitionFile]. The file is looked up in the same directory.

diff --git a/analyzer/src/main/kotlin/managers/utils/PythonInspector.kt b/analyzer/src/main/kotlin/managers/utils/PythonInspector.kt
@@ -202,7 +202,7 @@ internal fun PythonInspector.Result.toOrtProject(
     return Project(
         id = id,
         definitionFilePath = VersionControlSystem.getPathInfo(definitionFile).path,
-        authors = projectData?.parties?.toAuthors() ?: sortedSetOf(),
+        authors = projectData?.parties?.toAuthors() ?: emptySet(),
         declaredLicenses = projectData?.declaredLicense?.getDeclaredLicenses() ?: sortedSetOf(),
         vcs = VcsInfo.EMPTY,
         vcsProcessed = PackageManager.processProjectVcs(definitionFile.parentFile, VcsInfo.EMPTY, homepageUrl),
@@ -332,8 +332,8 @@ internal fun List<PythonInspector.Package>.toOrtPackages(): Set<Package> =
         )
     }
 
-private fun List<PythonInspector.Party>.toAuthors(): SortedSet<String> =
-    filter { it.role == "author" }.mapNotNullTo(sortedSetOf()) { party ->
+private fun List<PythonInspector.Party>.toAuthors(): Set<String> =
+    filter { it.role == "author" }.mapNotNullTo(mutableSetOf()) { party ->
         buildString {
             party.name?.let { append(it) }
             party.email?.let {

diff --git a/analyzer/src/test/kotlin/managers/BundlerTest.kt b/analyzer/src/test/kotlin/managers/BundlerTest.kt
@@ -66,7 +66,7 @@ class BundlerTest : WordSpec({
                 name = "rspec",
                 version = "3.7.0",
                 homepageUrl = "http://github.com/rspec",
-                authors = sortedSetOf("Steven Baker", "David Chelimsky", "Myron Marston"),
+                authors = setOf("Steven Baker", "David Chelimsky", "Myron Marston"),
                 declaredLicenses = sortedSetOf("MIT"),
                 description = "BDD for Ruby",
                 runtimeDependencies = setOf("rspec-core", "rspec-expectations", "rspec-mocks"),

diff --git a/model/src/main/kotlin/Package.kt b/model/src/main/kotlin/Package.kt
@@ -21,9 +21,11 @@ package org.ossreviewtoolkit.model
 
 import com.fasterxml.jackson.annotation.JsonAlias
 import com.fasterxml.jackson.annotation.JsonInclude
+import com.fasterxml.jackson.databind.annotation.JsonSerialize
 
 import java.util.SortedSet
 
+import org.ossreviewtoolkit.model.utils.StringSortedSetConverter
 import org.ossreviewtoolkit.model.utils.toPurl
 import org.ossreviewtoolkit.utils.ort.DeclaredLicenseProcessor
 import org.ossreviewtoolkit.utils.ort.ProcessedDeclaredLicense
@@ -61,7 +63,8 @@ data class Package(
      * The set of authors declared for this package.
      */
     @JsonInclude(JsonInclude.Include.NON_DEFAULT)
-    val authors: SortedSet<String> = sortedSetOf(),
+    @JsonSerialize(converter = StringSortedSetConverter::class)
+    val authors: Set<String> = emptySet(),
 
     /**
      * The set of licenses declared for this package. This does not necessarily correspond to the licenses as detected
@@ -139,7 +142,7 @@ data class Package(
         val EMPTY = Package(
             id = Identifier.EMPTY,
             purl = "",
-            authors = sortedSetOf(),
+            authors = emptySet(),
             declaredLicenses = sortedSetOf(),
             declaredLicensesProcessed = ProcessedDeclaredLicense.EMPTY,
             concludedLicense = null,

diff --git a/model/src/main/kotlin/PackageCurationData.kt b/model/src/main/kotlin/PackageCurationData.kt
@@ -22,8 +22,6 @@ package org.ossreviewtoolkit.model
 import com.fasterxml.jackson.annotation.JsonAlias
 import com.fasterxml.jackson.annotation.JsonInclude
 
-import java.util.SortedSet
-
 import org.ossreviewtoolkit.utils.common.zip
 import org.ossreviewtoolkit.utils.ort.DeclaredLicenseProcessor
 import org.ossreviewtoolkit.utils.spdx.SpdxExpression
@@ -54,7 +52,7 @@ data class PackageCurationData(
     /**
      * The set of authors of the package.
      */
-    val authors: SortedSet<String>? = null,
+    val authors: Set<String>? = null,
 
     /**
      * The concluded license as an [SpdxExpression]. It can be used to override the [declared][Package.declaredLicenses]
@@ -171,7 +169,7 @@ data class PackageCurationData(
             comment = setOfNotNull(comment, other.comment).joinToString("\n").takeIf { it.isNotEmpty() },
             purl = purl ?: other.purl,
             cpe = cpe ?: other.cpe,
-            authors = (authors.orEmpty() + other.authors.orEmpty()).toSortedSet(),
+            authors = authors.orEmpty() + other.authors.orEmpty(),
             concludedLicense = setOfNotNull(concludedLicense, other.concludedLicense).reduce(SpdxExpression::and),
             description = description ?: other.description,
             homepageUrl = homepageUrl ?: other.homepageUrl,