Update minimum JDK version to 21

.github/workflows/publish.yml

@@ -43,7 +43,7 @@ jobs:
         uses: actions/setup-java@v4
         with:
           distribution: temurin
-          java-version: 17
+          java-version: 21
           cache: sbt
 
       - name: Setup SBT

.github/workflows/scala.yml

@@ -16,8 +16,6 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - os: ubuntu-latest
-            java: 17
           - os: ubuntu-latest
             java: 21
           - os: ubuntu-latest
@@ -53,7 +51,7 @@ jobs:
         uses: actions/setup-java@v4
         with:
           distribution: temurin
-          java-version: 17
+          java-version: 21
           cache: sbt
 
       - name: Setup SBT

README.md

@@ -120,14 +120,21 @@ And use the `--debug` option to get more information about any exceptions you en
 
 ## Alternative installation: JAR builds
 
-If for some reason the binaries wouldn't work for you, or you want to get better performance for large files (see *Performance considerations* below), you can use the JAR build. This build runs on any platform, as long as you have Java (min. version 17). Go to the **[releases page](https://github.com/Jelly-RDF/cli/releases/latest)** and download the `jelly-cli.jar` file. Then, run it like so:
+If for some reason the binaries wouldn't work for you, or you want to get better performance for large files (see *Performance considerations* below), you can use the JAR build. This build runs on any platform, as long as you have Java (min. version 21). Go to the **[releases page](https://github.com/Jelly-RDF/cli/releases/latest)** and download the `jelly-cli.jar` file. Then, run it like so:
 
 ```shell
 java -jar jelly-cli.jar --help
 ```
 
 For simple day-to-day use cases, we recommend using the binary distribution, because it has way faster startup times and doesn't require you to install Java.
 
+### Java version requirements
+
+Depending on the version of `jelly-cli`:
+
+- 0.10.x or later: Java 21+
+- 0.1.x – 0.9.x: Java 17+
+
 ### Performance considerations
 
 If you are bulk-converting large amounts of RDF data (>10M triples), you may want to use the JAR build instead of the pre-compiled binary. With the JAR, your JVM will perform just-in-time compilation, resulting in better optimized code. The application will take longer to start, but the overall throughput will be better for large files.