Add hostname and idn-hostname format validators (#82)

Related to #54

.github/workflows/check.yml

@@ -25,6 +25,9 @@ jobs:
   check:
     runs-on: ${{ inputs.run-on }}
     steps:
+      - name: 'Install native dependencies'
+        run: sudo apt-get install -y libunistring-dev
+        if: runner.os == 'Linux'
       - name: 'Checkout Repository'
         uses: actions/checkout@v4
         with:
@@ -45,6 +48,13 @@ jobs:
           key: ${{ runner.os }}-gradle-${{ hashFiles('*.gradle.kts') }}
           restore-keys: |
               ${{ runner.os }}-gradle-
+      - name: Cache unicode data
+        uses: actions/cache@v4
+        with:
+          path: build/unicode_dump
+          key: unicode-dump-${{ hashFiles('build/unicode_dump/*') }}
+          restore-keys: |
+            unicode-dump-
       - name: Setup Gradle
         uses: gradle/actions/setup-gradle@v3
         with:

.github/workflows/platform-benchmark.yml

@@ -57,6 +57,13 @@ jobs:
           key: ${{ runner.os }}-gradle-${{ hashFiles('*.gradle.kts') }}
           restore-keys: |
               ${{ runner.os }}-gradle-
+      - name: Cache unicode data
+        uses: actions/cache@v4
+        with:
+          path: build/unicode_dump
+          key: unicode-dump-${{ hashFiles('build/unicode_dump/*') }}
+          restore-keys: |
+            unicode-dump-
       - name: Setup Gradle
         uses: gradle/actions/setup-gradle@v3
         with:

.github/workflows/pull_request.yml

@@ -29,6 +29,13 @@ jobs:
           key: ${{ runner.os }}-gradle-${{ hashFiles('*.gradle.kts') }}
           restore-keys: |
             ${{ runner.os }}-gradle-
+      - name: Cache unicode data
+        uses: actions/cache@v4
+        with:
+          path: build/unicode_dump
+          key: unicode-dump-${{ hashFiles('build/unicode_dump/*') }}
+          restore-keys: |
+            unicode-dump-
       - name: Setup Gradle
         uses: gradle/actions/setup-gradle@v3
         with:

.github/workflows/release.yml

@@ -49,6 +49,13 @@ jobs:
           key: ${{ runner.os }}-gradle-${{ hashFiles('*.gradle.kts') }}
           restore-keys: |
             ${{ runner.os }}-gradle-
+      - name: Cache unicode data
+        uses: actions/cache@v4
+        with:
+          path: build/unicode_dump
+          key: unicode-dump-${{ hashFiles('build/unicode_dump/*') }}
+          restore-keys: |
+            unicode-dump-
       - name: Setup Gradle
         uses: gradle/actions/setup-gradle@v3
         with:

.github/workflows/snapshot_release.yml

@@ -35,6 +35,13 @@ jobs:
           key: ${{ runner.os }}-gradle-${{ hashFiles('*.gradle.kts') }}
           restore-keys: |
             ${{ runner.os }}-gradle-
+      - name: Cache unicode data
+        uses: actions/cache@v4
+        with:
+          path: build/unicode_dump
+          key: unicode-dump-${{ hashFiles('build/unicode_dump/*') }}
+          restore-keys: |
+            unicode-dump-
       - name: Build with Gradle
         uses: gradle/actions/setup-gradle@v3
         with:

README.md

@@ -342,6 +342,8 @@ The library supports `format` assertion. For now only a few formats are supporte
 * ipv4
 * ipv6
 * uuid
+* hostname
+* idn-hostname
 
 But there is an API to implement the user's defined format validation.
 The [FormatValidator](src/commonMain/kotlin/io/github/optimumcode/json/schema/ValidationError.kt) interface can be user for that.
@@ -389,6 +391,14 @@ You can see the results in the latest workflow execution.
 The update to Kotlin 1.9.22 came with an issue for JS incremental compilation.
 In case you see an error about main function that already bind please execute `clean` task.
 
+When you build project for **linux** target you might get an error about missing native library.
+This is because `de.cketti.unicode:kotlin-codepoints` requires this library to perform string normalization.
+This is needed to support `idn-hostname` format. Install this library with the following command:
+
+```bash
+sudo apt-get install -y libunistring-dev
+```
+
 ### Devcontainer
 
 Devcontainers is a cool feature. However, by default in Codespaces and DevPod you will use [VS Code](https://code.visualstudio.com/).

build.gradle.kts

@@ -3,6 +3,7 @@ import org.jetbrains.kotlin.gradle.ExperimentalKotlinGradlePluginApi
 import org.jetbrains.kotlin.gradle.plugin.KotlinTarget
 import org.jetbrains.kotlin.gradle.plugin.KotlinTargetWithTests
 import org.jlleitschuh.gradle.ktlint.reporter.ReporterType
+import java.util.Locale
 
 plugins {
   alias(libs.plugins.kotlin.mutliplatform)
@@ -22,9 +23,109 @@ repositories {
 }
 
 apiValidation {
-  ignoredProjects += listOf("benchmark", "test-suites")
+  ignoredProjects += listOf("benchmark", "test-suites", "generator")
 }
 
+val generatedSourceDirectory: Provider<Directory> = layout.buildDirectory.dir("generated/source/unicode")
+
+//region Generation tasks block
+val generatorConfiguration: Configuration by configurations.creating
+
+dependencies {
+  generatorConfiguration(project(":generator"))
+}
+
+val dumpDir: Provider<Directory> = layout.buildDirectory.dir("unicode_dump")
+
+val dumpCharacterData by tasks.register<JavaExec>("dumpCharacterData") {
+  onlyIf {
+    dumpDir.get().asFile.run { !exists() || listFiles().isNullOrEmpty() }
+  }
+  outputs.dir(dumpDir)
+  classpath(generatorConfiguration)
+  mainClass.set("io.github.optimumcode.unocode.generator.Main")
+  args(
+    "dump",
+    "-o",
+    dumpDir.get(),
+  )
+}
+
+val generateCharacterDirectionData by tasks.register<JavaExec>("generateCharacterDirectionData") {
+  inputs.dir(dumpDir)
+  outputs.dir(generatedSourceDirectory)
+
+  dependsOn(dumpCharacterData)
+
+  classpath(generatorConfiguration)
+  mainClass.set("io.github.optimumcode.unocode.generator.Main")
+  args(
+    "character-direction",
+    "-p",
+    "io.github.optimumcode.json.schema.internal.unicode",
+    "-o",
+    generatedSourceDirectory.get(),
+    "-d",
+    dumpDir.get(),
+  )
+}
+
+val generateCharacterCategoryData by tasks.register<JavaExec>("generateCharacterCategoryData") {
+  inputs.dir(dumpDir)
+  outputs.dir(generatedSourceDirectory)
+
+  dependsOn(dumpCharacterData)
+
+  classpath(generatorConfiguration)
+  mainClass.set("io.github.optimumcode.unocode.generator.Main")
+  args(
+    "character-category",
+    "-p",
+    "io.github.optimumcode.json.schema.internal.unicode",
+    "-o",
+    generatedSourceDirectory.get(),
+    "-d",
+    dumpDir.get(),
+  )
+}
+
+val generateDerivedProperties by tasks.register<JavaExec>("generateDerivedProperties") {
+  val dataFile = layout.projectDirectory.dir("generator").dir("data").file("rfc5895_appendix_b_1.txt")
+  inputs.file(dataFile)
+  outputs.dir(generatedSourceDirectory)
+
+  classpath(generatorConfiguration)
+  mainClass.set("io.github.optimumcode.unocode.generator.Main")
+  args(
+    "derived-properties",
+    "-p",
+    "io.github.optimumcode.json.schema.internal.unicode",
+    "-o",
+    generatedSourceDirectory.get(),
+    "-d",
+    dataFile,
+  )
+}
+
+val generateJoiningTypes by tasks.register<JavaExec>("generateJoiningTypes") {
+  val dataFile = layout.projectDirectory.dir("generator").dir("data").file("DerivedJoiningType.txt")
+  inputs.file(dataFile)
+  outputs.dir(generatedSourceDirectory)
+
+  classpath(generatorConfiguration)
+  mainClass.set("io.github.optimumcode.unocode.generator.Main")
+  args(
+    "joining-types",
+    "-p",
+    "io.github.optimumcode.json.schema.internal.unicode",
+    "-o",
+    generatedSourceDirectory.get(),
+    "-d",
+    dataFile,
+  )
+}
+//endregion
+
 kotlin {
   explicitApi()
 
@@ -74,9 +175,18 @@ kotlin {
 
   sourceSets {
     commonMain {
+      kotlin.srcDirs(generatedSourceDirectory)
+
       dependencies {
         api(libs.kotlin.serialization.json)
         implementation(libs.uri)
+        // When using approach like above you won't be able to add because block
+        implementation(libs.kotlin.codepoints.get().toString()) {
+          because("simplifies work with unicode codepoints")
+        }
+        implementation(libs.normalize.get().toString()) {
+          because("provides normalization required by IDN-hostname format")
+        }
       }
     }
     commonTest {
@@ -94,6 +204,19 @@ kotlin {
     }
   }
 
+  targets.configureEach {
+    val capitalizedTargetName =
+      name.replaceFirstChar { if (it.isLowerCase()) it.titlecase(Locale.getDefault()) else it.toString() }
+    tasks.named("compileKotlin$capitalizedTargetName") {
+      dependsOn(
+        generateCharacterDirectionData,
+        generateCharacterCategoryData,
+        generateDerivedProperties,
+        generateJoiningTypes,
+      )
+    }
+  }
+
   afterEvaluate {
     fun Task.dependsOnTargetTests(targets: List<KotlinTarget>) {
       targets.forEach {
@@ -122,11 +245,48 @@ kotlin {
   }
 }
 
+afterEvaluate {
+  val taskNames = setOf("compile", "detekt", "runKtlint")
+  tasks.configureEach {
+    // There is something wrong with compileCommonMainKotlinMetadata task
+    // Gradle cannot find it, but this task uses the generated source directory
+    // and Gradle reports implicit dependency.
+    // As a workaround I do this - seems like it is working.
+    // However, I might be missing something. Need to revisit this later.
+
+    if (taskNames.any { name.startsWith(it) }) {
+      mustRunAfter(
+        generateCharacterDirectionData,
+        generateCharacterCategoryData,
+        generateDerivedProperties,
+        generateJoiningTypes,
+      )
+    }
+  }
+}
+
+koverReport {
+  filters {
+    excludes {
+      packages(
+        "io.github.optimumcode.json.schema.internal.unicode.*",
+        "io.github.optimumcode.json.schema.internal.unicode",
+      )
+    }
+  }
+}
+
 ktlint {
   version.set(libs.versions.ktlint)
   reporters {
     reporter(ReporterType.HTML)
   }
+  filter {
+    exclude { el ->
+      val absolutePath = el.file.absolutePath
+      absolutePath.contains("generated").and(!el.isDirectory)
+    }
+  }
 }
 
 afterEvaluate {

generator/build.gradle.kts

@@ -0,0 +1,52 @@
+import com.expediagroup.graphql.plugin.gradle.config.GraphQLSerializer
+import com.expediagroup.graphql.plugin.gradle.graphql
+import org.jlleitschuh.gradle.ktlint.reporter.ReporterType
+
+plugins {
+  // otherwise there is Gradle exception
+  // https://github.com/gradle/gradle/issues/20084
+  id(libs.plugins.kotlin.jvm.get().pluginId)
+  alias(libs.plugins.kotlin.serialization)
+  alias(libs.plugins.expediagroup.graphql)
+
+  alias(libs.plugins.detekt)
+  alias(libs.plugins.ktlint)
+}
+
+repositories {
+  mavenCentral()
+}
+
+kotlin {
+  jvmToolchain(11)
+}
+
+dependencies {
+  implementation(libs.kotlinpoet)
+  implementation(libs.graphql.ktor)
+  implementation(libs.clikt) {
+    because("cli for executing generation")
+  }
+}
+
+graphql {
+  client {
+    endpoint = "https://www.compart.com/en/unicode/graphql"
+    packageName = "io.github.optimumcode.unicode.generator.internal.graphql"
+    serializer = GraphQLSerializer.KOTLINX
+  }
+}
+
+ktlint {
+  version.set(libs.versions.ktlint)
+  debug.set(true)
+  reporters {
+    reporter(ReporterType.HTML)
+  }
+  filter {
+    exclude { el ->
+      val absolutePath = el.file.absolutePath
+      absolutePath.contains("generated").and(!el.isDirectory)
+    }
+  }
+}