Better detection for Enso's NI when launching Language Server (#12034)

Previosly we were wrongly relying on the presence of the file. That way,
a bash script meant that NI integration was wrongly used.
This change uses Tika, but it has already been present in other
subprojects so no additional dependency shall be added.

Follow up on #11880.

Author: hubertp

build.sbt

@@ -4495,10 +4495,12 @@ lazy val `runtime-version-manager` = project
     libraryDependencies ++= Seq(
       "com.typesafe.scala-logging" %% "scala-logging"    % scalaLoggingVersion,
       "org.apache.commons"          % "commons-compress" % commonsCompressVersion,
+      "org.apache.tika"             % "tika-core"        % tikaVersion,
       "org.scalatest"              %% "scalatest"        % scalatestVersion % Test
     ),
     Compile / moduleDependencies ++= Seq(
       "org.apache.commons" % "commons-compress" % commonsCompressVersion,
+      "org.apache.tika"    % "tika-core"        % tikaVersion,
       "org.slf4j"          % "slf4j-api"        % slf4jVersion
     ),
     Compile / internalModuleDependencies := Seq(

distribution/launcher/THIRD-PARTY/NOTICE

@@ -91,6 +91,11 @@ The license file can be found at `licenses/APACHE2.0`.
 Copyright notices related to this dependency can be found in the directory `org.apache.commons.commons-compress-1.23.0`.
 
 
+'tika-core', licensed under the Apache License, Version 2.0, is distributed with the launcher.
+The license file can be found at `licenses/APACHE2.0`.
+Copyright notices related to this dependency can be found in the directory `org.apache.tika.tika-core-2.4.1`.
+
+
 'scala-java8-compat_2.13', licensed under the Apache-2.0, is distributed with the launcher.
 The license file can be found at `licenses/APACHE2.0`.
 Copyright notices related to this dependency can be found in the directory `org.scala-lang.modules.scala-java8-compat_2.13-1.0.0`.

distribution/launcher/THIRD-PARTY/org.apache.tika.tika-core-2.4.1/NOTICE

@@ -0,0 +1,8 @@
+
+Apache Tika core
+Copyright 2007-2022 The Apache Software Foundation
+
+This product includes software developed at
+The Apache Software Foundation (http://www.apache.org/).
+
+

distribution/launcher/THIRD-PARTY/org.apache.tika.tika-core-2.4.1/NOTICES

@@ -0,0 +1,9 @@
+    /**
+     * Information about rights held in and over the resource. Typically,
+     * a Rights element will contain a rights management statement for
+     * the resource, or reference a service providing such information.
+     * Rights information often encompasses Intellectual Property Rights
+     * (IPR), Copyright, and various Property Rights. If the Rights element
+     * is absent, no assumptions can be made about the status of these and
+     * other rights with respect to the resource.
+     */

distribution/project-manager/THIRD-PARTY/NOTICE

@@ -356,6 +356,11 @@ The license information can be found along with the copyright notices.
 Copyright notices related to this dependency can be found in the directory `org.apache.commons.commons-lang3-3.12.0`.
 
 
+'tika-core', licensed under the Apache License, Version 2.0, is distributed with the project-manager.
+The license file can be found at `licenses/APACHE2.0`.
+Copyright notices related to this dependency can be found in the directory `org.apache.tika.tika-core-2.4.1`.
+
+
 'reactive-streams', licensed under the CC0, is distributed with the project-manager.
 The license file can be found at `licenses/CC0`.
 Copyright notices related to this dependency can be found in the directory `org.reactivestreams.reactive-streams-1.0.3`.

distribution/project-manager/THIRD-PARTY/org.apache.tika.tika-core-2.4.1/NOTICE

@@ -0,0 +1,8 @@
+
+Apache Tika core
+Copyright 2007-2022 The Apache Software Foundation
+
+This product includes software developed at
+The Apache Software Foundation (http://www.apache.org/).
+
+

distribution/project-manager/THIRD-PARTY/org.apache.tika.tika-core-2.4.1/NOTICES

@@ -0,0 +1,9 @@
+    /**
+     * Information about rights held in and over the resource. Typically,
+     * a Rights element will contain a rights management statement for
+     * the resource, or reference a service providing such information.
+     * Rights information often encompasses Intellectual Property Rights
+     * (IPR), Copyright, and various Property Rights. If the Rights element
+     * is absent, no assumptions can be made about the status of these and
+     * other rights with respect to the resource.
+     */

lib/scala/runtime-version-manager/src/main/scala/org/enso/runtimeversionmanager/runner/NativeExecCommand.scala

@@ -4,6 +4,9 @@ import org.enso.cli.OS
 import org.enso.distribution.{DistributionManager, Environment}
 import org.enso.runtimeversionmanager.components.Engine
 
+import org.apache.tika.config.TikaConfig
+import org.apache.tika.Tika
+
 import java.nio.file.Path
 
 case class NativeExecCommand(executablePath: Path) extends ExecCommand {
@@ -26,7 +29,22 @@ object NativeExecCommand {
     val fullExecPath =
       dm.paths.engines.resolve(version).resolve("bin").resolve(execName)
 
-    if (fullExecPath.toFile.exists()) Some(NativeExecCommand(fullExecPath))
-    else None
+    if (fullExecPath.toFile.exists() && isBinary(fullExecPath)) {
+      Some(NativeExecCommand(fullExecPath))
+    } else None
+  }
+
+  private def isBinary(path: Path): Boolean = {
+    try {
+      val config    = TikaConfig.getDefaultConfig()
+      val tika      = new Tika(config)
+      val mimeTypes = config.getMimeRepository
+      val mime      = tika.detect(path);
+      val tpe       = mimeTypes.forName(mime).getType.getType
+      tpe != null && tpe == "application"
+    } catch {
+      case _: Throwable =>
+        false
+    }
   }
 }

tools/legal-review/engine/report-state

@@ -1,3 +1,3 @@
-6EB80DB9BFB5BAC58FFE8EAB48A509A40EAC27733FCD0F46BC5FC3FDDB82339C
+1C7BD42739A2FD536962BEEA3DC052C025F06FFA1E25E9D7BA0D41E9714BCD24
 29565A85352D2C3F92648DF46C368298D320747F00AE06FD83197DD6E4CC1DF7
 0

tools/legal-review/launcher/org.apache.tika.tika-core-2.4.1/copyright-ignore

@@ -0,0 +1,32 @@
+"The copyright information."
+ARTWORK_OR_OBJECT_DETAIL_COPYRIGHT_NOTICE, ARTWORK_OR_OBJECT_DETAIL_CREATOR,
+COPYRIGHT_OWNER_ID, COPYRIGHT_OWNER_NAME, IMAGE_CREATOR_ID, IMAGE_CREATOR_NAME,
+Contains any necessary copyright notice for claiming the intellectual
+Copyright 2016 Norconex Inc.
+Copyright Owner, Image Supplier and Licensor may be the same or different
+Copyright ownership can be expressed in a more controlled way using the
+DATE_CREATED, DESCRIPTION_WRITER, INSTRUCTIONS, JOB_ID, TITLE, COPYRIGHT_NOTICE,
+Owner or owners of the copyright in the licensed image.
+PREFIX_IPTC_EXT + TikaCoreProperties.NAMESPACE_PREFIX_DELIMITER + "AOCopyrightNotice");
+PREFIX_PLUS + TikaCoreProperties.NAMESPACE_PREFIX_DELIMITER + "CopyrightOwner");
+PREFIX_PLUS + TikaCoreProperties.NAMESPACE_PREFIX_DELIMITER + "CopyrightOwnerID"),
+PREFIX_PLUS + TikaCoreProperties.NAMESPACE_PREFIX_DELIMITER + "CopyrightOwnerId";
+PREFIX_PLUS + TikaCoreProperties.NAMESPACE_PREFIX_DELIMITER + "CopyrightOwnerName");
+Property ARTWORK_OR_OBJECT_DETAIL_COPYRIGHT_NOTICE = Property.internalTextBag(
+Property COPYRIGHT = Property.externalText("xmpDM:copyright");
+Property COPYRIGHT_NOTICE = DublinCore.RIGHTS;
+Property COPYRIGHT_OWNER = Property.internalTextBag(
+Property COPYRIGHT_OWNER_ID = Property.composite(Property.internalTextBag(
+Property COPYRIGHT_OWNER_NAME = Property.internalTextBag(
+Serves to identify the rights holder/s for the image. The Copyright
+String COPYRIGHT_OWNER_ID_WRONG_CASE =
+The ID of the owner or owners of the copyright in the licensed image.
+The name of the owner or owners of the copyright in the licensed image.
+current owner of the copyright of this work with associated intellectual
+in the CopyrightNotice.
+new Property[]{Property.internalTextBag(IPTC.COPYRIGHT_OWNER_ID_WRONG_CASE)});
+regarding copyright ownership.  The ASF licenses this file
+the copyrights of this image other rights properties may be used.
+standard. These parts Copyright 2010 International Press Telecommunications
+this work for additional information regarding copyright ownership.
+@deprecated use {@link IPTC#COPYRIGHT_OWNER_ID}