feat: Tika CLI improvements — metadata-only, quiet output, zio-logging

CHANGELOG.md

@@ -14,6 +14,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Wired scalafix into Mill build via mill-scalafix 0.6.0 plugin (`./mill __.fix`).
 - Updated scalafix OrganizeImports to coalesce all imports to wildcards with Scala 3 dialect.
 - Added RedundantSyntax scalafix rule.
+- Added opt-in CLI flag `--license-aware-capabilities` (and `XLCR_LICENSE_AWARE_CAPABILITIES` for server mode) to enable runtime Aspose license-aware capability checks while keeping default fast probing.
 
 ### Removed
 - `core-spark` module and all Spark pipeline sources/tests/docs.

README.md

@@ -101,6 +101,9 @@ xlcr split -i message.eml -d attachments/
 # Show document metadata
 xlcr info -i document.pdf
 
+# Show metadata/capabilities with runtime Aspose license checks (opt-in, slower)
+xlcr info -i document.pdf --license-aware-capabilities
+
 # List all supported conversions
 xlcr --backend-info
 
@@ -147,6 +150,7 @@ docker compose up server
 | `--lo-restart-after` | `XLCR_LO_RESTART_AFTER` | `200` | Restart LO after N conversions |
 | `--lo-task-timeout` | `XLCR_LO_TASK_TIMEOUT` | `120000` | Conversion timeout (ms) |
 | `--lo-queue-timeout` | `XLCR_LO_QUEUE_TIMEOUT` | `30000` | Queue wait timeout (ms) |
+| `--license-aware-capabilities` | `XLCR_LICENSE_AWARE_CAPABILITIES` | `false` | Use runtime Aspose license checks for `/capabilities`, `/info`, and convert/split preflight checks |
 
 ### LibreOffice Process Pooling
 
@@ -158,6 +162,9 @@ xlcr server start --lo-instances 4 --lo-restart-after 100
 
 # Or via environment variables
 XLCR_LO_INSTANCES=4 XLCR_LO_RESTART_AFTER=100 xlcr server start
+
+# Enable runtime license-aware capability checks (opt-in)
+xlcr server start --license-aware-capabilities
 ```
 
 Each instance runs as a separate LibreOffice process on a dedicated port (starting from 2002). JODConverter handles round-robin task distribution and automatic process restarts. Budget ~200-300MB RAM per instance.

core/package.mill

@@ -18,9 +18,10 @@ trait CoreModule extends build.XLCRModule {
   def testResources = Task.Sources(moduleDir / "test" / "resources")
 
   override def mvnDeps = Seq(
-    // Logging
+    // Logging: ZIO-native logging with SLF4J2 bridge (captures Tika/POI/Aspose logs)
     mvn"org.slf4j:slf4j-api:2.0.17",
-    mvn"ch.qos.logback:logback-classic:1.5.32",
+    mvn"dev.zio::zio-logging:2.5.3",
+    mvn"dev.zio::zio-logging-slf4j2-bridge:2.5.3",
     mvn"org.apache.logging.log4j:log4j-to-slf4j:2.25.3", // Bridge Log4j2 (from Tika) to SLF4J
     // ZIO
     mvn"dev.zio::zio:2.1.24",

core/src/main/scala/com/tjclp/xlcr/cli/Commands.scala

@@ -43,7 +43,8 @@ object Commands:
   case class InfoArgs(
     input: Path,
     extensionOnly: Boolean = false,
-    format: OutputFormat = OutputFormat.Text
+    format: OutputFormat = OutputFormat.Text,
+    licenseAwareCapabilities: Boolean = false
   )
 
   /** Arguments for the server command (pure data — no HTTP imports) */
@@ -54,7 +55,8 @@ object Commands:
     loInstances: Option[Int] = None,
     loRestartAfter: Option[Int] = None,
     loTaskTimeout: Option[Long] = None,
-    loQueueTimeout: Option[Long] = None
+    loQueueTimeout: Option[Long] = None,
+    licenseAwareCapabilities: Boolean = false
   )
 
   /** Output format for the info command */
@@ -140,6 +142,12 @@ object Commands:
       else OutputFormat.Text
     }
 
+  private val licenseAwareCapabilitiesFlag: Opts[Boolean] =
+    Opts.flag(
+      long = "license-aware-capabilities",
+      help = "Run runtime Aspose license checks when probing capabilities (slower, more accurate)"
+    ).orFalse
+
   private val backendOpt: Opts[Option[Backend]] =
     Opts.option[String](
       long = "backend",
@@ -279,7 +287,9 @@ object Commands:
     )(splitOpts.map(CliCommand.Split.apply))
 
   private val infoOpts: Opts[InfoArgs] =
-    (inputOpt, extensionOnlyFlag, outputFormatOpt).mapN(InfoArgs.apply)
+    (inputOpt, extensionOnlyFlag, outputFormatOpt, licenseAwareCapabilitiesFlag).mapN(
+      InfoArgs.apply
+    )
 
   val infoCmd: Opts[CliCommand] =
     Opts.subcommand(
@@ -353,7 +363,8 @@ object Commands:
       loInstancesOpt,
       loRestartAfterOpt,
       loTaskTimeoutOpt,
-      loQueueTimeoutOpt
+      loQueueTimeoutOpt,
+      licenseAwareCapabilitiesFlag
     ).mapN(ServerArgs.apply)
 
   private val serverStartCmd: Opts[CliCommand] =

core/test/src/cli/CommandsSpec.scala

@@ -198,6 +198,26 @@ class CommandsSpec extends AnyFlatSpec with Matchers:
       case _ => fail("Expected Info command")
   }
 
+  it should "parse info command with --license-aware-capabilities" in {
+    val result = parse(Seq("info", "-i", "document.pdf", "--license-aware-capabilities"))
+
+    result.isRight shouldBe true
+    result.toOption.get match
+      case CliCommand.Info(args) =>
+        args.licenseAwareCapabilities shouldBe true
+      case _ => fail("Expected Info command")
+  }
+
+  it should "parse server start with --license-aware-capabilities" in {
+    val result = parse(Seq("server", "start", "--license-aware-capabilities"))
+
+    result.isRight shouldBe true
+    result.toOption.get match
+      case CliCommand.Server(args) =>
+        args.licenseAwareCapabilities shouldBe true
+      case _ => fail("Expected Server command")
+  }
+
   // ============================================================================
   // Path Handling Tests
   // ============================================================================

scripts/xlcr-wrapper.sh

@@ -45,39 +45,5 @@ MEMORY_OPTS=(
     "-Xmx2g"
 )
 
-# Show backend status if requested
-if [[ "$1" == "--backend-info" ]]; then
-    echo "XLCR Backend Status:"
-    echo "  JAR: $XLCR_JAR"
-    echo ""
-
-    # Check for bundled license in JAR
-    if unzip -l "$XLCR_JAR" 2>/dev/null | grep -q "Aspose.Total.Java.lic"; then
-        echo "  Aspose: Licensed (bundled in JAR)"
-    elif [[ -n "$ASPOSE_TOTAL_LICENSE_B64" ]]; then
-        echo "  Aspose: Licensed (env ASPOSE_TOTAL_LICENSE_B64)"
-    elif [[ -f "Aspose.Total.Java.lic" ]]; then
-        echo "  Aspose: Licensed (local file)"
-    else
-        echo "  Aspose: Evaluation mode (no license found)"
-    fi
-
-    # Check LibreOffice
-    if [[ -d "/Applications/LibreOffice.app" ]]; then
-        LO_VERSION=$(/Applications/LibreOffice.app/Contents/MacOS/soffice --version 2>/dev/null | head -1 || echo "version unknown")
-        echo "  LibreOffice: Available (macOS app - $LO_VERSION)"
-    elif command -v soffice &> /dev/null; then
-        LO_VERSION=$(soffice --version 2>/dev/null | head -1 || echo "version unknown")
-        echo "  LibreOffice: Available ($LO_VERSION)"
-    elif [[ -d "${LIBREOFFICE_HOME:-/usr/lib/libreoffice}" ]]; then
-        echo "  LibreOffice: Available at ${LIBREOFFICE_HOME:-/usr/lib/libreoffice}"
-    else
-        echo "  LibreOffice: Not found"
-    fi
-
-    echo ""
-    # Continue to also show Java-side backend info
-fi
-
 # Run XLCR (filter out JVM module warnings)
 java "${JVM_OPTS[@]}" "${MEMORY_OPTS[@]}" ${JAVA_OPTS:-} -jar "$XLCR_JAR" "$@" 2> >(grep -v "^WARNING: package sun.misc not in java.base$" >&2)

xlcr/src/main/aspose/com/tjclp/xlcr/cli/BackendWiring.scala

@@ -28,11 +28,20 @@ object BackendWiring:
   ): ZIO[Any, TransformError, Chunk[DynamicFragment]] =
     AsposeTransforms.split(input, options)
 
-  def asposeCanConvert(from: Mime, to: Mime): Boolean =
-    AsposeTransforms.canConvertLicensed(from, to)
+  def asposeCanConvert(
+    from: Mime,
+    to: Mime,
+    licenseAwareCapabilities: Boolean = false
+  ): Boolean =
+    if licenseAwareCapabilities then AsposeTransforms.canConvertLicensed(from, to)
+    else AsposeTransforms.canConvert(from, to)
 
-  def asposeCanSplit(mime: Mime): Boolean =
-    AsposeTransforms.canSplitLicensed(mime)
+  def asposeCanSplit(
+    mime: Mime,
+    licenseAwareCapabilities: Boolean = false
+  ): Boolean =
+    if licenseAwareCapabilities then AsposeTransforms.canSplitLicensed(mime)
+    else AsposeTransforms.canSplit(mime)
 
   def checkAsposeStatus(): Unit =
     if AsposeLicenseV2.licenseAvailable then

xlcr/src/main/no-aspose/com/tjclp/xlcr/cli/BackendWiring.scala

@@ -28,9 +28,16 @@ object BackendWiring:
   ): ZIO[Any, TransformError, Chunk[DynamicFragment]] =
     ZIO.fail(UnsupportedConversion(input.mime, input.mime))
 
-  def asposeCanConvert(from: Mime, to: Mime): Boolean = false
+  def asposeCanConvert(
+    from: Mime,
+    to: Mime,
+    licenseAwareCapabilities: Boolean = false
+  ): Boolean = false
 
-  def asposeCanSplit(mime: Mime): Boolean = false
+  def asposeCanSplit(
+    mime: Mime,
+    licenseAwareCapabilities: Boolean = false
+  ): Boolean = false
 
   def checkAsposeStatus(): Unit =
     println("    Status: Not included in this build (no Aspose license detected at build time)")

xlcr/src/main/scala/com/tjclp/xlcr/cli/UnifiedTransforms.scala

@@ -80,14 +80,20 @@ object UnifiedTransforms:
    * When `backend` is None, returns true if ANY backend supports it. When a specific backend is
    * given, checks only that backend.
    */
-  def canConvert(from: Mime, to: Mime, backend: Option[Backend] = None): Boolean =
+  def canConvert(
+    from: Mime,
+    to: Mime,
+    backend: Option[Backend] = None,
+    licenseAwareCapabilities: Boolean = false
+  ): Boolean =
     backend match
-      case Some(Backend.Aspose)      => BackendWiring.asposeCanConvert(from, to)
+      case Some(Backend.Aspose) =>
+        BackendWiring.asposeCanConvert(from, to, licenseAwareCapabilities)
       case Some(Backend.LibreOffice) => libreOfficeAvailable &&
         LibreOfficeTransforms.canConvert(from, to)
       case Some(Backend.Xlcr) => XlcrTransforms.canConvert(from, to)
       case None               =>
-        BackendWiring.asposeCanConvert(from, to) ||
+        BackendWiring.asposeCanConvert(from, to, licenseAwareCapabilities) ||
         (libreOfficeAvailable && LibreOfficeTransforms.canConvert(from, to)) ||
         XlcrTransforms.canConvert(from, to)
 
@@ -127,13 +133,18 @@ object UnifiedTransforms:
    * When `backend` is None, returns true if ANY backend supports it. When a specific backend is
    * given, checks only that backend.
    */
-  def canSplit(mime: Mime, backend: Option[Backend] = None): Boolean =
+  def canSplit(
+    mime: Mime,
+    backend: Option[Backend] = None,
+    licenseAwareCapabilities: Boolean = false
+  ): Boolean =
     backend match
-      case Some(Backend.Aspose)      => BackendWiring.asposeCanSplit(mime)
+      case Some(Backend.Aspose) =>
+        BackendWiring.asposeCanSplit(mime, licenseAwareCapabilities)
       case Some(Backend.LibreOffice) => libreOfficeAvailable && LibreOfficeTransforms.canSplit(mime)
       case Some(Backend.Xlcr)        => XlcrTransforms.canSplit(mime)
       case None                      =>
-        BackendWiring.asposeCanSplit(mime) ||
+        BackendWiring.asposeCanSplit(mime, licenseAwareCapabilities) ||
         (libreOfficeAvailable && LibreOfficeTransforms.canSplit(mime)) ||
         XlcrTransforms.canSplit(mime)