diff --git a/docs/source/admin/fs/local-fs.rst b/docs/source/admin/fs/local-fs.rst
index 3178f3777..741858cd7 100644
--- a/docs/source/admin/fs/local-fs.rst
+++ b/docs/source/admin/fs/local-fs.rst
@@ -512,42 +512,53 @@ For example, a PDF document could generate:
 .. code:: json
 
    {
-      "date" : "2016-07-07T08:37:42Z",
-      "pdf:PDFVersion" : "1.5",
-      "xmp:CreatorTool" : "Microsoft Word",
-      "Keywords" : "keyword1, keyword2",
+      "access_permission:assemble_document" : "true",
+      "access_permission:can_modify" : "true",
+      "access_permission:can_print" : "true",
+      "access_permission:can_print_faithful" : "true",
+      "access_permission:extract_content" : "true",
+      "access_permission:extract_for_accessibility" : "true",
+      "access_permission:fill_in_form" : "true",
       "access_permission:modify_annotations" : "true",
-      "access_permission:can_print_degraded" : "true",
-      "subject" : "Test Tika Object",
+      "Content-Length" : "101643",
+      "Content-Type" : "application/pdf",
+      "Content-Type-Magic-Detected" : "application/pdf",
       "dc:creator" : "David Pilato",
+      "dc:format" : "application/pdf; version=1.5",
+      "dc:language" : "en-US",
+      "dc:subject" : "keyword1, keyword2",
+      "dc:title" : "Test Tika title",
       "dcterms:created" : "2016-07-07T08:37:42Z",
-      "Last-Modified" : "2016-07-07T08:37:42Z",
       "dcterms:modified" : "2016-07-07T08:37:42Z",
-      "dc:format" : "application/pdf; version=1.5",
-      "title" : "Test Tika title",
-      "Last-Save-Date" : "2016-07-07T08:37:42Z",
-      "access_permission:fill_in_form" : "true",
-      "meta:save-date" : "2016-07-07T08:37:42Z",
+      "pdf:charsPerPage" : "42",
+      "pdf:containsDamagedFont" : "false",
+      "pdf:containsNonEmbeddedFont" : "false",
+      "pdf:docinfo:created" : "2016-07-07T08:37:42Z",
+      "pdf:docinfo:creator" : "David Pilato",
+      "pdf:docinfo:creator_tool" : "Microsoft Word",
+      "pdf:docinfo:keywords" : "keyword1, keyword2",
+      "pdf:docinfo:modified" : "2016-07-07T08:37:42Z",
+      "pdf:docinfo:subject" : "Test Tika Object",
+      "pdf:docinfo:title" : "Test Tika title",
       "pdf:encrypted" : "false",
-      "dc:title" : "Test Tika title",
-      "modified" : "2016-07-07T08:37:42Z",
-      "cp:subject" : "Test Tika Object",
-      "Content-Type" : "application/pdf",
-      "X-Parsed-By" : "org.apache.tika.parser.DefaultParser",
-      "creator" : "David Pilato",
-      "meta:author" : "David Pilato",
-      "dc:subject" : "keyword1, keyword2",
-      "meta:creation-date" : "2016-07-07T08:37:42Z",
-      "created" : "Thu Jul 07 10:37:42 CEST 2016",
-      "access_permission:extract_for_accessibility" : "true",
-      "access_permission:assemble_document" : "true",
-      "xmpTPg:NPages" : "2",
-      "Creation-Date" : "2016-07-07T08:37:42Z",
-      "access_permission:extract_content" : "true",
-      "access_permission:can_print" : "true",
-      "meta:keyword" : "keyword1, keyword2",
-      "Author" : "David Pilato",
-      "access_permission:can_modify" : "true"
+      "pdf:eofOffsets" : "101460",
+      "pdf:hasCollection" : "false",
+      "pdf:hasMarkedContent" : "true",
+      "pdf:hasXFA" : "false",
+      "pdf:hasXMP" : "false",
+      "pdf:incrementalUpdateCount" : "1",
+      "pdf:num3DAnnotations" : "0",
+      "pdf:ocrPageCount" : "0",
+      "pdf:overallPercentageUnmappedUnicodeChars" : "0.0",
+      "pdf:PDFVersion" : "1.5",
+      "pdf:totalUnmappedUnicodeChars" : "0",
+      "pdf:unmappedUnicodeCharsPerPage" : "0",
+      "X-TIKA:Parsed-By" : "org.apache.tika.parser.pdf.PDFParser",
+      "X-TIKA:Parsed-By-Full-Set" : "org.apache.tika.parser.pdf.PDFParser",
+      "X-TIKA:resourceName" : "test.pdf",
+      "X-TIKA:versionCount" : "1",
+      "xmp:CreatorTool" : "Microsoft Word",
+      "xmpTPg:NPages" : "2"
    }
 
 Where a MP3 file would generate:
@@ -555,27 +566,26 @@ Where a MP3 file would generate:
 .. code:: json
 
    {
-      "xmpDM:genre" : "Vocal",
-      "X-Parsed-By" : "org.apache.tika.parser.DefaultParser",
-      "creator" : "David Pilato",
+      "channels" : "2",
+      "Content-Type" : "audio/mpeg",
+      "Content-Type-Magic-Detected" : "audio/mpeg",
+      "dc:creator" : "David Pilato",
+      "dc:title" : "Test Tika",
+      "samplerate" : "44100",
+      "version" : "MPEG 3 Layer III Version 1",
+      "X-TIKA:Parsed-By" : "org.apache.tika.parser.DefaultParser",
+      "X-TIKA:Parsed-By-Full-Set" : "org.apache.tika.parser.DefaultParser",
+      "X-TIKA:resourceName" : "test.mp3",
       "xmpDM:album" : "FS Crawler",
-      "xmpDM:trackNumber" : "1",
-      "xmpDM:releaseDate" : "2016",
-      "meta:author" : "David Pilato",
       "xmpDM:artist" : "David Pilato",
-      "dc:creator" : "David Pilato",
-      "xmpDM:audioCompressor" : "MP3",
-      "title" : "Test Tika",
       "xmpDM:audioChannelType" : "Stereo",
-      "version" : "MPEG 3 Layer III Version 1",
-      "xmpDM:logComment" : "Hello but reverted",
+      "xmpDM:audioCompressor" : "MP3",
       "xmpDM:audioSampleRate" : "44100",
-      "channels" : "2",
-      "dc:title" : "Test Tika",
-      "Author" : "David Pilato",
-      "xmpDM:duration" : "1018.775146484375",
-      "Content-Type" : "audio/mpeg",
-      "samplerate" : "44100"
+      "xmpDM:duration" : "1.0187751054763794",
+      "xmpDM:genre" : "Vocal",
+      "xmpDM:logComment" : "Hello but reverted",
+      "xmpDM:releaseDate" : "2016",
+      "xmpDM:trackNumber" : "1"
    }
 
 .. note::
@@ -879,31 +889,19 @@ Tika Config Path
 
 .. versionadded:: 2.10
 
-If you want to override the default tika parser configuration, you can set the path to a custom tika
-configuration file, which will be used instead.
+.. warning::
+
+   Since the upgrade to Apache Tika 4, ``fs.tika_config_path`` is **temporarily not supported**. Tika 4
+   removed the XML-based Tika configuration file mechanism, and there is currently no drop-in replacement
+   to assemble a parser from a configuration file. If this setting is provided, FSCrawler fails fast with
+   a clear configuration error. Support is expected to return on top of Tika 4's new JSON-based
+   configuration.
+
+This setting used to let you override the default Tika parser configuration by pointing to a custom Tika
+configuration file:
 
 .. code:: yaml
 
    name: "test"
    fs:
      tika_config_path: '/path/to/tikaConfig.xml'
-
-An example tika config file is shown below. See |Tika_configuring|_ for more information.
-
-.. code:: xml
-
-  <?xml version="1.0" encoding="UTF-8"?>
-  <properties>
-    <service-loader dynamic="true"/>
-    <service-loader loadErrorHandler="IGNORE"/>
-    <parsers>
-      <!-- Use Default Parser for files, but Default Parser will never use HTML parser -->
-      <parser class="org.apache.tika.parser.DefaultParser">
-        <parser-exclude class="org.apache.tika.parser.html.JsoupParser"/>
-      </parser>
-      <!-- Use a different parser for XHTML -->
-      <parser class="org.apache.tika.parser.xml.XMLParser">
-        <mime>application/xhtml+xml</mime>
-      </parser>
-    </parsers>
-  </properties>
diff --git a/docs/source/admin/fs/rest.rst b/docs/source/admin/fs/rest.rst
index 4ecc1c210..9f10191c2 100644
--- a/docs/source/admin/fs/rest.rst
+++ b/docs/source/admin/fs/rest.rst
@@ -78,7 +78,7 @@ It will give you a response similar to:
            "language" : "eng",
            "output_type" : "txt",
            "page_seg_mode" : 1,
-           "pdf_strategy" : "ocr_and_text",
+           "pdf_strategy" : "auto",
            "preserve_interword_spacing" : false
          },
          "raw_metadata" : true,
@@ -151,9 +151,12 @@ You will get back your document as it has been stored by elasticsearch:
        },
        "meta" : {
          "raw" : {
-           "X-Parsed-By" : "org.apache.tika.parser.DefaultParser",
            "Content-Encoding" : "ISO-8859-1",
-           "Content-Type" : "text/plain; charset=ISO-8859-1"
+           "Content-Type" : "text/plain; charset=ISO-8859-1",
+           "Content-Type-Magic-Detected" : "text/plain",
+           "X-TIKA:Parsed-By" : "org.apache.tika.parser.DefaultParser",
+           "X-TIKA:Parsed-By-Full-Set" : "org.apache.tika.parser.DefaultParser",
+           "X-TIKA:resourceName" : "test.txt"
          }
        },
        "path" : {
@@ -187,9 +190,12 @@ will give
        },
        "meta" : {
          "raw" : {
-           "X-Parsed-By" : "org.apache.tika.parser.DefaultParser",
            "Content-Encoding" : "ISO-8859-1",
-           "Content-Type" : "text/plain; charset=ISO-8859-1"
+           "Content-Type" : "text/plain; charset=ISO-8859-1",
+           "Content-Type-Magic-Detected" : "text/plain",
+           "X-TIKA:Parsed-By" : "org.apache.tika.parser.DefaultParser",
+           "X-TIKA:Parsed-By-Full-Set" : "org.apache.tika.parser.DefaultParser",
+           "X-TIKA:resourceName" : "test.txt"
          }
        },
        "path" : {
diff --git a/docs/source/fscrawler.ini b/docs/source/fscrawler.ini
index beabba034..70cb9a521 100644
--- a/docs/source/fscrawler.ini
+++ b/docs/source/fscrawler.ini
@@ -3,7 +3,7 @@ Version=2.10-SNAPSHOT
 JavaVersion=17
 
 [3rdParty]
-TikaVersion=3.3.1
+TikaVersion=4.0.0-beta-1
 ElasticsearchVersion7=7.17.29
 ElasticsearchVersion8=8.19.5
 ElasticsearchVersion9=9.3.0
diff --git a/docs/source/release/2.10.rst b/docs/source/release/2.10.rst
index 5022c6387..113e3c761 100644
--- a/docs/source/release/2.10.rst
+++ b/docs/source/release/2.10.rst
@@ -17,6 +17,21 @@ enough. Thanks to dadoonet.
 * We don't support anymore the ``elasticsearch.nodes.url`` setting. You need to use ``elasticsearch.urls``
   instead. Thanks to dadoonet.
 * The ``_upload`` REST endpoint has been removed. Please now use the ``_document`` endpoint. Thanks to dadoonet.
+* The Apache Tika 4 upgrade changes several ``meta.raw.*`` metadata key names. If you search, aggregate or
+  map these fields, you will need to update your queries and index templates accordingly:
+
+  * Image and EXIF metadata (JPEG, PNG, TIFF, …) keys are now namespaced under ``img:``. For example
+    ``Number of Tables`` becomes ``img:Number of Tables`` and ``Exif IFD0:Orientation`` becomes
+    ``img:Exif IFD0:Orientation``.
+  * The resource name key ``resourceName`` is renamed to ``X-TIKA:resourceName``.
+  * New keys are now produced, such as ``Content-Type-Magic-Detected``, ``X-TIKA:Parsed-By-Full-Set`` and,
+    for text documents, the encoding detection keys ``X-TIKA:detectedEncoding``,
+    ``X-TIKA:encodingDetectionTrace`` and ``X-TIKA:encodingDetector``.
+
+  Thanks to dadoonet.
+* The default ``fs.ocr.pdf_strategy`` is now ``auto`` instead of ``ocr_and_text``. With ``auto``, OCR is skipped on
+  PDF pages that already contain more than 10 characters of text. If you relied on the previous behaviour and want
+  OCR on every PDF page, explicitly set ``fs.ocr.pdf_strategy`` to ``ocr_and_text``. Thanks to dadoonet.
 
 New
 ---
@@ -38,6 +53,8 @@ New
 * You can now remove a document in Elasticsearch using FSCrawler ``_document`` endpoint. See :ref:`rest-service`. Thanks to dadoonet.
 * Implement our own HTTP Client for Elasticsearch. Thanks to dadoonet.
 * Add option to set path to custom tika config file. See :ref:`local-fs-settings`. Thanks to iadcode.
+  **Note**: this option is temporarily disabled after the Apache Tika 4 upgrade, which removed the XML-based
+  configuration file mechanism. It is expected to return on top of Tika 4's JSON-based configuration.
 * Support for Index Templates. See :ref:`mappings`. Thanks to dadoonet.
 * Support for Aliases. You can now index to an alias. Thanks to dadoonet.
 * Support for Access Token and Api Keys instead of Basic Authentication. See :ref:`credentials`. Thanks to dadoonet.
diff --git a/docs/source/user/ocr.rst b/docs/source/user/ocr.rst
index 83e65f5ff..fa564c232 100644
--- a/docs/source/user/ocr.rst
+++ b/docs/source/user/ocr.rst
@@ -30,7 +30,7 @@ Here is a list of OCR settings (under ``fs.ocr`` prefix):
 +---------------------------------------+-------------------------------------------------+------------------+-----------------------------------+
 | ``fs.ocr.output_type``                | ``FSCRAWLER_FS_OCR_OUTPUT_TYPE``                | ``txt``          | `OCR Output Type`_                |
 +---------------------------------------+-------------------------------------------------+------------------+-----------------------------------+
-| ``fs.ocr.pdf_strategy``               | ``FSCRAWLER_FS_OCR_PDF_STRATEGY``               | ``ocr_and_text`` | `OCR PDF Strategy`_               |
+| ``fs.ocr.pdf_strategy``               | ``FSCRAWLER_FS_OCR_PDF_STRATEGY``               | ``auto``         | `OCR PDF Strategy`_               |
 +---------------------------------------+-------------------------------------------------+------------------+-----------------------------------+
 | ``fs.ocr.page_seg_mode``              | ``FSCRAWLER_FS_OCR_PAGE_SEG_MODE``              | ``01``           | `OCR Page Seg Mode`_              |
 +---------------------------------------+-------------------------------------------------+------------------+-----------------------------------+
@@ -165,8 +165,8 @@ Supported strategies are:
 
 * ``ocr_and_text``: OCR and text extraction is performed.
 
-.. note:: When omitted, ``ocr_and_text`` value is used. If you have performance issues, it's worth using the ``auto`` option
-    instead as only documents with barely no text will go through the OCR process.
+.. note:: When omitted, ``auto`` value is used. OCR is skipped on PDF pages that already contain more than
+    10 characters of text. If you need OCR on every page regardless, set ``ocr_and_text`` or ``ocr_only``.
 
 OCR Page Seg Mode
 -----------------
diff --git a/elasticsearch-client/src/test/resources/log4j2.xml b/elasticsearch-client/src/test/resources/log4j2.xml
index aea0bc66b..d30e01bce 100644
--- a/elasticsearch-client/src/test/resources/log4j2.xml
+++ b/elasticsearch-client/src/test/resources/log4j2.xml
@@ -9,7 +9,7 @@
       <Logger name="fr.pilato.elasticsearch.crawler.fs" level="info" additivity="false">
          <AppenderRef ref="CONSOLE"/>
       </Logger>
-      <Logger name="fr.pilato.elasticsearch.crawler.fs.client" level="debug" additivity="false">
+      <Logger name="fr.pilato.elasticsearch.crawler.fs.client" level="info" additivity="false">
          <AppenderRef ref="CONSOLE"/>
       </Logger>
       <Logger name="org.glassfish" level="info" additivity="false">
diff --git a/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerImplAllDocumentsIT.java b/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerImplAllDocumentsIT.java
index c720136d0..4ede6f4b6 100644
--- a/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerImplAllDocumentsIT.java
+++ b/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerImplAllDocumentsIT.java
@@ -43,13 +43,14 @@
 import java.nio.file.Path;
 import org.apache.logging.log4j.LogManager;
 import org.apache.logging.log4j.Logger;
-import org.apache.tika.parser.external.ExternalParser;
+import org.apache.tika.exception.TikaConfigException;
+import org.apache.tika.parser.ocr.TesseractOCRParser;
 import org.assertj.core.api.Assertions;
-import org.assertj.core.api.Assumptions;
 import org.assertj.core.api.InstanceOfAssertFactories;
 import org.junit.jupiter.api.AfterAll;
 import org.junit.jupiter.api.BeforeAll;
 import org.junit.jupiter.api.Test;
+import org.junit.jupiter.api.condition.DisabledIf;
 
 /** Test all type of documents we have */
 @VerySlow
@@ -253,14 +254,28 @@ void chineseContent369() throws ElasticsearchClientException {
     }
 
     @Test
+    @DisabledIf(value = "tesseractNotInstalled", disabledReason = "Tesseract is not installed")
     void ocr() throws ElasticsearchClientException {
-        Assumptions.assumeThat(ExternalParser.check("tesseract"))
-                .as("Tesseract is not installed so we are skipping this test")
-                .isTrue();
         runSearch("test-ocr.png", "words");
         runSearch("test-ocr.pdf", "words");
     }
 
+    /**
+     * Condition for {@link #ocr()}. Tika 4 removed {@code ExternalParser.check()}, so we reuse the same Tesseract
+     * availability check as production code. Returns {@code true} (disabling the test) when Tesseract is not installed
+     * or cannot be configured.
+     *
+     * @return {@code true} when Tesseract is unavailable
+     */
+    @SuppressWarnings("unused")
+    static boolean tesseractNotInstalled() {
+        try {
+            return !new TesseractOCRParser().hasTesseract();
+        } catch (TikaConfigException e) {
+            return true;
+        }
+    }
+
     @Test
     void shiftJisEncoding() throws ElasticsearchClientException {
         runSearch("issue-400-shiftjis.txt", "elasticsearch");
diff --git a/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerRestIT.java b/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerRestIT.java
index 3a5ee72ce..613d4a7f7 100644
--- a/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerRestIT.java
+++ b/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerRestIT.java
@@ -372,7 +372,7 @@ void documentWithExternalTags() throws Exception {
         checkDocument("replace_meta_only.txt", hit -> {
             Assertions.assertThat((String) JsonPath.read(hit.getSource(), "$.content"))
                     .contains("This file content will be extracted");
-            Assertions.assertThat((String) JsonPath.read(hit.getSource(), "$.meta.raw.resourceName"))
+            Assertions.assertThat((String) JsonPath.read(hit.getSource(), "$.meta.raw['X-TIKA:resourceName']"))
                     .isEqualTo("another-file-name.txt");
             Assertions.assertThatThrownBy(() -> JsonPath.read(hit.getSource(), "$.external"))
                     .isInstanceOf(PathNotFoundException.class);
diff --git a/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerTestOcrIT.java b/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerTestOcrIT.java
index 760774b7b..c0aa2606b 100644
--- a/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerTestOcrIT.java
+++ b/integration-tests/src/test/java/fr/pilato/elasticsearch/crawler/fs/test/integration/elasticsearch/FsCrawlerTestOcrIT.java
@@ -149,29 +149,9 @@ void ocr_disabled() throws Exception {
         Map<String, String> raw = JsonPath.read(jpgHitSource, "$.meta.raw");
         Assertions.assertThat(raw)
                 .as("test-ocr.jpg: raw metadata")
-                .hasSize(64)
-                .containsKey("Component 1")
-                .containsKey("Component 2")
-                .containsKey("Component 3")
-                .containsEntry("Compression Type", "Baseline")
+                .hasSize(66)
                 .containsEntry("Content-Type", "image/jpeg")
-                .containsEntry("Data Precision", "8 bits")
-                .containsEntry("Exif IFD0:Orientation", "Top, left side (Horizontal / normal)")
-                .containsEntry("Exif IFD0:Resolution Unit", "Inch")
-                .containsKey("Exif IFD0:X Resolution")
-                .containsKey("Exif IFD0:Y Resolution")
-                .containsEntry("Exif IFD0:YCbCr Positioning", "Center of pixel array")
-                .containsEntry("Exif SubIFD:Color Space", "sRGB")
-                .containsEntry("Exif SubIFD:Components Configuration", "YCbCr")
-                .containsEntry("Exif SubIFD:Exif Image Height", "622 pixels")
-                .containsEntry("Exif SubIFD:Exif Image Width", "982 pixels")
-                .containsEntry("Exif SubIFD:Exif Version", "2.21")
-                .containsEntry("Exif SubIFD:FlashPix Version", "1.00")
-                .containsEntry("Exif SubIFD:Scene Capture Type", "Standard")
-                .containsKey("File Modified Date")
-                .hasEntrySatisfying(
-                        "File Name", value -> Assertions.assertThat(value).startsWith("apache-tika-"))
-                .containsEntry("File Size", "41426 bytes")
+                .containsEntry("Content-Type-Magic-Detected", "image/jpeg")
                 .containsKey("ICC:Apple Multi-language Profile Name")
                 .containsKey("ICC:Blue Colorant")
                 .containsKey("ICC:Blue Parametric TRC")
@@ -201,11 +181,31 @@ void ocr_disabled() throws Exception {
                 .containsKey("ICC:Version")
                 .containsKey("ICC:Video Card Gamma")
                 .containsKey("ICC:XYZ values")
-                .containsEntry("Image Height", "622 pixels")
-                .containsEntry("Image Width", "982 pixels")
-                .containsEntry("Number of Components", "3")
-                .containsEntry("Number of Tables", "4 Huffman tables")
-                .containsEntry("resourceName", "test-ocr.jpg")
+                .containsKey("img:Component 1")
+                .containsKey("img:Component 2")
+                .containsKey("img:Component 3")
+                .containsEntry("img:Compression Type", "Baseline")
+                .containsEntry("img:Data Precision", "8 bits")
+                .containsEntry("img:Exif IFD0:Orientation", "Top, left side (Horizontal / normal)")
+                .containsEntry("img:Exif IFD0:Resolution Unit", "Inch")
+                .containsKey("img:Exif IFD0:X Resolution")
+                .containsKey("img:Exif IFD0:Y Resolution")
+                .containsEntry("img:Exif IFD0:YCbCr Positioning", "Center of pixel array")
+                .containsEntry("img:Exif SubIFD:Color Space", "sRGB")
+                .containsEntry("img:Exif SubIFD:Components Configuration", "YCbCr")
+                .containsEntry("img:Exif SubIFD:Exif Image Height", "622 pixels")
+                .containsEntry("img:Exif SubIFD:Exif Image Width", "982 pixels")
+                .containsEntry("img:Exif SubIFD:Exif Version", "2.21")
+                .containsEntry("img:Exif SubIFD:FlashPix Version", "1.00")
+                .containsEntry("img:Exif SubIFD:Scene Capture Type", "Standard")
+                .containsKey("img:File Modified Date")
+                .hasEntrySatisfying(
+                        "img:File Name", value -> Assertions.assertThat(value).startsWith("apache-tika-"))
+                .containsEntry("img:File Size", "41426 bytes")
+                .containsEntry("img:Image Height", "622 pixels")
+                .containsEntry("img:Image Width", "982 pixels")
+                .containsEntry("img:Number of Components", "3")
+                .containsEntry("img:Number of Tables", "4 Huffman tables")
                 .containsEntry("tiff:BitsPerSample", "8")
                 .containsEntry("tiff:ImageLength", "622")
                 .containsEntry("tiff:ImageWidth", "982")
@@ -214,7 +214,8 @@ void ocr_disabled() throws Exception {
                 .containsEntry("tiff:XResolution", "144.0")
                 .containsEntry("tiff:YResolution", "144.0")
                 .containsKey("X-TIKA:Parsed-By")
-                .containsKey("X-TIKA:Parsed-By-Full-Set");
+                .containsKey("X-TIKA:Parsed-By-Full-Set")
+                .containsEntry("X-TIKA:resourceName", "test-ocr.jpg");
 
         // test-ocr.png: OCR disabled → no content
         String pngHitSource = findHitByFilename(searchResponse, "test-ocr.png").getSource();
diff --git a/integration-tests/src/test/resources-binary/tags/document_with_external_tags/replace_meta_only.txt.json b/integration-tests/src/test/resources-binary/tags/document_with_external_tags/replace_meta_only.txt.json
index 617442d5d..4548718e1 100644
--- a/integration-tests/src/test/resources-binary/tags/document_with_external_tags/replace_meta_only.txt.json
+++ b/integration-tests/src/test/resources-binary/tags/document_with_external_tags/replace_meta_only.txt.json
@@ -1,7 +1,7 @@
 {
   "meta":{
     "raw":{
-      "resourceName":"another-file-name.txt"
+      "X-TIKA:resourceName":"another-file-name.txt"
     }
   }
 }
diff --git a/pom.xml b/pom.xml
index aa5b142c0..c4ad8f8c0 100644
--- a/pom.xml
+++ b/pom.xml
@@ -69,7 +69,7 @@
         <gestalt.version>0.38.0</gestalt.version>
         <snakeyaml.version>2.6</snakeyaml.version>
 
-        <tika.version>3.3.1</tika.version>
+        <tika.version>4.0.0-beta-1</tika.version>
         <jackson.version>3.1.4</jackson.version>
         <!--
             Jackson 2.x is still pulled transitively by Tika (jackson-databind). Pin it to a patched
@@ -394,6 +394,7 @@
                 <groupId>org.apache.tika</groupId>
                 <artifactId>tika-parsers-standard-package</artifactId>
                 <version>${tika.version}</version>
+                <type>pom</type>
             </dependency>
             <dependency>
                 <groupId>org.apache.tika</groupId>
diff --git a/settings/src/main/java/fr/pilato/elasticsearch/crawler/fs/settings/Ocr.java b/settings/src/main/java/fr/pilato/elasticsearch/crawler/fs/settings/Ocr.java
index 7fa00cf79..7873a05c9 100644
--- a/settings/src/main/java/fr/pilato/elasticsearch/crawler/fs/settings/Ocr.java
+++ b/settings/src/main/java/fr/pilato/elasticsearch/crawler/fs/settings/Ocr.java
@@ -44,7 +44,7 @@ public class Ocr {
     @Nullable
     private String outputType;
     // Pdf OCR Strategy
-    @Config(defaultVal = "ocr_and_text")
+    @Config(defaultVal = "auto")
     private String pdfStrategy;
     // PDF Page Seg Mode
     @Config
@@ -110,7 +110,7 @@ public void setPageSegMode(Integer pageSegMode) {
     }
 
     /**
-     * Get the PDF Strategy. Could be "no_ocr", "auto", "ocr_only" or "ocr_and_text" (default)
+     * Get the PDF Strategy. Could be "no_ocr", "auto" (default), "ocr_only" or "ocr_and_text"
      *
      * @return the PDF Strategy
      */
diff --git a/settings/src/main/resources/fr/pilato/elasticsearch/crawler/fs/settings/fscrawler-default.properties b/settings/src/main/resources/fr/pilato/elasticsearch/crawler/fs/settings/fscrawler-default.properties
index 249cb38f0..5534299e0 100644
--- a/settings/src/main/resources/fr/pilato/elasticsearch/crawler/fs/settings/fscrawler-default.properties
+++ b/settings/src/main/resources/fr/pilato/elasticsearch/crawler/fs/settings/fscrawler-default.properties
@@ -25,7 +25,7 @@ fs.follow_symlinks=false
 fs.ocr.enabled=true
 fs.ocr.language=eng
 fs.ocr.output_type=txt
-fs.ocr.pdf_strategy=ocr_and_text
+fs.ocr.pdf_strategy=auto
 fs.ocr.page_seg_mode=1
 fs.ocr.preserve_interword_spacing=false
 
diff --git a/settings/src/main/resources/fr/pilato/elasticsearch/crawler/fs/settings/fscrawler-default.yaml b/settings/src/main/resources/fr/pilato/elasticsearch/crawler/fs/settings/fscrawler-default.yaml
index f7563e929..a95a7e252 100644
--- a/settings/src/main/resources/fr/pilato/elasticsearch/crawler/fs/settings/fscrawler-default.yaml
+++ b/settings/src/main/resources/fr/pilato/elasticsearch/crawler/fs/settings/fscrawler-default.yaml
@@ -85,7 +85,7 @@
     # optional: the OCR output type, either txt or hocr
     #output_type: txt
     # optional: the OCR strategy, either auto, no_ocr, ocr_only, ocr_and_text
-    #pdf_strategy: ocr_and_text
+    #pdf_strategy: auto
     # optional: the mode to use for page orientation from 0 to 13
     #page_seg_mode: 1 # automatic mode
     # optional: if true, we will try to preserve interword spacing
diff --git a/settings/src/test/java/fr/pilato/elasticsearch/crawler/fs/settings/FsSettingsLoaderTest.java b/settings/src/test/java/fr/pilato/elasticsearch/crawler/fs/settings/FsSettingsLoaderTest.java
index fe1f6f1eb..1f99ae68a 100644
--- a/settings/src/test/java/fr/pilato/elasticsearch/crawler/fs/settings/FsSettingsLoaderTest.java
+++ b/settings/src/test/java/fr/pilato/elasticsearch/crawler/fs/settings/FsSettingsLoaderTest.java
@@ -230,7 +230,7 @@ private FsSettings generateExpectedDefaultFsSettings() {
         ocr.setEnabled(true);
         ocr.setLanguage("eng");
         ocr.setOutputType("txt");
-        ocr.setPdfStrategy("ocr_and_text");
+        ocr.setPdfStrategy("auto");
         ocr.setPageSegMode(1);
         fs.setOcr(ocr);
         expected.setFs(fs);
diff --git a/tika/pom.xml b/tika/pom.xml
index f6e8f76a2..177609f3f 100644
--- a/tika/pom.xml
+++ b/tika/pom.xml
@@ -42,6 +42,7 @@
         <dependency>
             <groupId>org.apache.tika</groupId>
             <artifactId>tika-parsers-standard-package</artifactId>
+            <type>pom</type>
         </dependency>
         <dependency>
             <groupId>org.apache.tika</groupId>
diff --git a/tika/src/main/java/fr/pilato/elasticsearch/crawler/fs/tika/TikaInstance.java b/tika/src/main/java/fr/pilato/elasticsearch/crawler/fs/tika/TikaInstance.java
index 9b778bd18..2c6d5ae7a 100644
--- a/tika/src/main/java/fr/pilato/elasticsearch/crawler/fs/tika/TikaInstance.java
+++ b/tika/src/main/java/fr/pilato/elasticsearch/crawler/fs/tika/TikaInstance.java
@@ -20,6 +20,7 @@
  */
 package fr.pilato.elasticsearch.crawler.fs.tika;
 
+import fr.pilato.elasticsearch.crawler.fs.framework.FsCrawlerIllegalConfigurationException;
 import fr.pilato.elasticsearch.crawler.fs.settings.Fs;
 import java.io.File;
 import java.io.IOException;
@@ -27,15 +28,17 @@
 import java.util.Arrays;
 import java.util.HashSet;
 import java.util.List;
+import java.util.Locale;
 import java.util.Set;
 import org.apache.logging.log4j.LogManager;
 import org.apache.logging.log4j.Logger;
 import org.apache.tika.config.ServiceLoader;
-import org.apache.tika.config.TikaConfig;
 import org.apache.tika.exception.TikaConfigException;
 import org.apache.tika.exception.TikaException;
 import org.apache.tika.exception.WriteLimitReachedException;
 import org.apache.tika.exception.ZeroByteFileException;
+import org.apache.tika.io.TemporaryResources;
+import org.apache.tika.io.TikaInputStream;
 import org.apache.tika.language.detect.LanguageDetector;
 import org.apache.tika.metadata.Metadata;
 import org.apache.tika.metadata.TikaCoreProperties;
@@ -54,7 +57,9 @@
 import org.apache.tika.parser.image.TiffParser;
 import org.apache.tika.parser.ocr.TesseractOCRConfig;
 import org.apache.tika.parser.ocr.TesseractOCRParser;
+import org.apache.tika.parser.pdf.OcrConfig;
 import org.apache.tika.parser.pdf.PDFParser;
+import org.apache.tika.parser.pdf.PDFParserConfig;
 import org.apache.tika.sax.BodyContentHandler;
 import org.apache.tika.sax.WriteOutContentHandler;
 import org.xml.sax.SAXException;
@@ -95,40 +100,40 @@ class TikaInstance {
 
     private void initParser(Fs fs) {
         if (fs.getTikaConfigPath() != null) {
-            if (!(new File(fs.getTikaConfigPath())).exists()) {
-                throw new RuntimeException("Tika configuration file " + fs.getTikaConfigPath() + " not found!");
-            }
-            logger.info("Using custom tika configuration from [{}].", fs.getTikaConfigPath());
-            TikaConfig config = null;
-            try {
-                config = new TikaConfig(fs.getTikaConfigPath());
-            } catch (TikaException | IOException | SAXException e) {
-                logger.error("Can not configure Tika: {}", e.getMessage());
-                logger.debug("Fullstack trace error for Tika", e);
-            }
+            // Apache Tika 4 removed the XML-based Tika configuration file mechanism (configuration moved to
+            // per-component JSON, and there is no drop-in way to assemble an AutoDetectParser from a file).
+            // Fail fast with a clear message instead of silently ignoring the setting.
+            // TODO Re-introduce custom Tika configuration on top of Tika 4's JSON config mechanism (temporary removal).
+            throw new FsCrawlerIllegalConfigurationException(
+                    "fs.tika_config_path is not supported with Apache Tika 4.x: "
+                            + "the XML Tika configuration file mechanism was removed upstream. Please remove this setting ["
+                            + fs.getTikaConfigPath() + "].");
+        }
 
-            parser = new AutoDetectParser(config);
-        } else {
-            PDFParser pdfParser = new PDFParser();
-            DefaultParser defaultParser;
-            TesseractOCRParser ocrParser;
-            Set<MediaType> exclude = new HashSet<>();
-            exclude.add(MediaType.image("png"));
-            exclude.add(MediaType.image("jpeg"));
-            exclude.add(MediaType.image("bmp"));
-            exclude.add(MediaType.image("gif"));
+        PDFParser pdfParser = new PDFParser();
+        DefaultParser defaultParser;
+        Set<MediaType> exclude = new HashSet<>();
+        exclude.add(MediaType.image("png"));
+        exclude.add(MediaType.image("jpeg"));
+        exclude.add(MediaType.image("bmp"));
+        exclude.add(MediaType.image("gif"));
 
-            Parser gdalParser = ParserDecorator.withoutTypes(new GDALParser(), exclude);
+        Parser gdalParser = ParserDecorator.withoutTypes(new GDALParser(), exclude);
 
-            // To solve https://issues.apache.org/jira/browse/TIKA-3364
-            // PDF content might be extracted multiple times.
-            pdfParser.getPDFParserConfig().setExtractBookmarksText(false);
+        // To solve https://issues.apache.org/jira/browse/TIKA-3364
+        // PDF content might be extracted multiple times.
+        pdfParser.getPDFParserConfig().setExtractBookmarksText(false);
 
-            if (ocrActivated) {
-                logger.debug("OCR is activated.");
-                ocrParser = new TesseractOCRParser();
+        if (ocrActivated) {
+            logger.debug("OCR is activated.");
+            // In Tika 4 the Tesseract paths live on TesseractOCRConfig (registered in the parse context), not on
+            // the parser. We reuse the very config built in initContext() so hasTesseract() checks our paths, and
+            // we build the parser from it. Any misconfiguration (e.g. a wrong path) throws a TikaConfigException,
+            // in which case we simply disable OCR — mirroring the historical "silently skip OCR" behaviour.
+            TesseractOCRConfig ocrConfig = context.get(TesseractOCRConfig.class);
+            try {
                 if (fs.getOcr().getPath() != null) {
-                    // Tika's setTesseractPath() wants the directory containing the tesseract executable, but
+                    // Tika's tesseractPath wants the directory containing the tesseract executable, but
                     // fs.ocr.path is documented as the path to the binary itself. Accept both: if the path
                     // points to an existing file, use its parent directory instead.
                     File ocrPath = new File(fs.getOcr().getPath());
@@ -137,61 +142,99 @@ private void initParser(Fs fs) {
                             ? ocrPath.getAbsoluteFile().getParent()
                             : fs.getOcr().getPath();
                     logger.debug("Tesseract Path set to [{}].", tesseractPath);
-                    ocrParser.setTesseractPath(tesseractPath);
+                    ocrConfig.setTesseractPath(tesseractPath);
                 }
                 if (fs.getOcr().getDataPath() != null) {
                     logger.debug("Tesseract Data Path set to [{}].", fs.getOcr().getDataPath());
-                    ocrParser.setTessdataPath(fs.getOcr().getDataPath());
+                    ocrConfig.setTessdataPath(fs.getOcr().getDataPath());
                 }
-                try {
-                    if (ocrParser.hasTesseract()) {
-                        logger.debug(
-                                "OCR strategy for PDF documents is [{}] and tesseract was found.",
-                                fs.getOcr().getPdfStrategy());
-                        pdfParser.setOcrStrategy(fs.getOcr().getPdfStrategy());
-                    } else {
-                        logger.debug("But Tesseract is not installed so we won't run OCR.");
-                        ocrActivated = false;
-                        pdfParser.setOcrStrategy("no_ocr");
-                    }
-                } catch (TikaConfigException e) {
-                    logger.debug("Tesseract is not correctly set up so we won't run OCR. Error is: {}", e.getMessage());
-                    logger.debug("Fullstack trace error for Tesseract", e);
+                TesseractOCRParser ocrParser = new TesseractOCRParser(ocrConfig);
+                if (ocrParser.hasTesseract()) {
+                    logger.debug(
+                            "OCR strategy for PDF documents is [{}] and tesseract was found.",
+                            fs.getOcr().getPdfStrategy());
+                    setPdfOcrStrategy(pdfParser, fs.getOcr().getPdfStrategy());
+                } else {
+                    logger.debug("But Tesseract is not installed so we won't run OCR.");
                     ocrActivated = false;
-                    pdfParser.setOcrStrategy("no_ocr");
+                    setPdfOcrStrategy(pdfParser, "no_ocr");
                 }
+            } catch (TikaConfigException e) {
+                logger.debug("Tesseract is not correctly set up so we won't run OCR. Error is: {}", e.getMessage());
+                logger.debug("Fullstack trace error for Tesseract", e);
+                ocrActivated = false;
+                setPdfOcrStrategy(pdfParser, "no_ocr");
             }
+        }
 
-            if (ocrActivated) {
-                logger.info("OCR is enabled. This might slowdown the process.");
-                // We are excluding the pdf parser as we built one that we want to use instead.
-                defaultParser = new DefaultParser(
-                        MediaTypeRegistry.getDefaultRegistry(),
-                        new ServiceLoader(),
-                        List.of(PDFParser.class, GDALParser.class));
-            } else {
-                logger.info("OCR is disabled.");
-                TesseractOCRConfig config = context.get(TesseractOCRConfig.class);
-                if (config != null) {
-                    config.setSkipOcr(true);
-                }
-                // We are excluding the pdf parser as we built one that we want to use instead
-                // and the OCR Parser as it's explicitly disabled.
-                defaultParser = new DefaultParser(
-                        MediaTypeRegistry.getDefaultRegistry(),
-                        new ServiceLoader(),
-                        Arrays.asList(PDFParser.class, TesseractOCRParser.class));
+        if (ocrActivated) {
+            logger.info("OCR is enabled. This might slowdown the process.");
+            // We are excluding the pdf parser as we built one that we want to use instead.
+            defaultParser = new DefaultParser(
+                    MediaTypeRegistry.getDefaultRegistry(),
+                    new ServiceLoader(),
+                    List.of(PDFParser.class, GDALParser.class));
+        } else {
+            logger.info("OCR is disabled.");
+            TesseractOCRConfig config = context.get(TesseractOCRConfig.class);
+            if (config != null) {
+                config.setSkipOcr(true);
             }
-            parser = new AutoDetectParser(
-                    defaultParser,
-                    pdfParser,
-                    gdalParser,
-                    new BPGParser(),
-                    new TiffParser(),
-                    new HeifParser(),
-                    new ImageParser(),
-                    new JpegParser());
+            // We are excluding the pdf parser as we built one that we want to use instead
+            // and the OCR Parser as it's explicitly disabled.
+            defaultParser = new DefaultParser(
+                    MediaTypeRegistry.getDefaultRegistry(),
+                    new ServiceLoader(),
+                    Arrays.asList(PDFParser.class, TesseractOCRParser.class));
+        }
+        parser = new AutoDetectParser(
+                defaultParser,
+                pdfParser,
+                gdalParser,
+                new BPGParser(),
+                new TiffParser(),
+                new HeifParser(),
+                new ImageParser(),
+                new JpegParser());
+    }
+
+    /**
+     * Applies the FSCrawler PDF OCR strategy on the given PDF parser. In Tika 4 the strategy moved from
+     * {@code PDFParser.setOcrStrategy(String)} to {@link PDFParserConfig#setOcr(OcrConfig)} with an
+     * {@link OcrConfig.Strategy} enum, so we translate our documented string values accordingly.
+     *
+     * @param pdfParser the PDF parser to configure
+     * @param strategy one of {@code no_ocr}, {@code auto}, {@code ocr_only} or {@code ocr_and_text}
+     */
+    private static void setPdfOcrStrategy(PDFParser pdfParser, String strategy) {
+        PDFParserConfig pdfConfig = pdfParser.getPDFParserConfig();
+        OcrConfig ocrConfig = pdfConfig.getOcr();
+        if (ocrConfig == null) {
+            ocrConfig = new OcrConfig();
+        }
+        ocrConfig.setStrategy(mapPdfOcrStrategy(strategy));
+        pdfConfig.setOcr(ocrConfig);
+    }
+
+    /**
+     * Maps an FSCrawler PDF OCR strategy string to the Tika 4 {@link OcrConfig.Strategy} enum. When the strategy is
+     * null, {@link OcrConfig.Strategy#AUTO} is used (the FSCrawler default). The FSCrawler value {@code ocr_and_text}
+     * maps to {@link OcrConfig.Strategy#OCR_AND_TEXT_EXTRACTION}. Unknown values fall back to
+     * {@link OcrConfig.Strategy#AUTO}.
+     *
+     * @param strategy the FSCrawler strategy string (may be null)
+     * @return the matching Tika strategy
+     */
+    private static OcrConfig.Strategy mapPdfOcrStrategy(String strategy) {
+        if (strategy == null) {
+            return OcrConfig.Strategy.AUTO;
         }
+        return switch (strategy.toLowerCase(Locale.ROOT)) {
+            case "no_ocr" -> OcrConfig.Strategy.NO_OCR;
+            case "ocr_only" -> OcrConfig.Strategy.OCR_ONLY;
+            case "ocr_and_text", "ocr_and_text_extraction" -> OcrConfig.Strategy.OCR_AND_TEXT_EXTRACTION;
+            default -> OcrConfig.Strategy.AUTO;
+        };
     }
 
     private void initContext(Fs fs) {
@@ -234,8 +277,11 @@ private void initContext(Fs fs) {
      */
     String extractText(int indexedChars, InputStream stream, Metadata metadata) throws IOException, TikaException {
         WriteOutContentHandler handler = new WriteOutContentHandler(indexedChars);
-        try {
-            parser.parse(stream, new BodyContentHandler(handler), metadata, context);
+        // Tika 4 requires a TikaInputStream for Parser.parse(). Use TemporaryResources so Tika can spool/cache
+        // without closing the caller's stream (required for FTP/SFTP wrappers that finalize the transfer on close).
+        try (TemporaryResources tmp = new TemporaryResources()) {
+            TikaInputStream tikaInputStream = TikaInputStream.get(stream, tmp, metadata);
+            parser.parse(tikaInputStream, new BodyContentHandler(handler), metadata, context);
         } catch (WriteLimitReachedException e) {
             String resourceName = metadata.get(TikaCoreProperties.RESOURCE_NAME_KEY);
             logger.debug("We reached the limit we set ({}) for {}: {}", indexedChars, resourceName, e.getMessage());
diff --git a/tika/src/test/java/fr/pilato/elasticsearch/crawler/fs/tika/TikaDocParserTest.java b/tika/src/test/java/fr/pilato/elasticsearch/crawler/fs/tika/TikaDocParserTest.java
index 001776670..05ae9a477 100644
--- a/tika/src/test/java/fr/pilato/elasticsearch/crawler/fs/tika/TikaDocParserTest.java
+++ b/tika/src/test/java/fr/pilato/elasticsearch/crawler/fs/tika/TikaDocParserTest.java
@@ -22,6 +22,7 @@
 
 import com.carrotsearch.randomizedtesting.jupiter.RandomizedTest;
 import fr.pilato.elasticsearch.crawler.fs.beans.Doc;
+import fr.pilato.elasticsearch.crawler.fs.framework.FsCrawlerIllegalConfigurationException;
 import fr.pilato.elasticsearch.crawler.fs.framework.FsCrawlerUtil;
 import fr.pilato.elasticsearch.crawler.fs.settings.FsSettings;
 import fr.pilato.elasticsearch.crawler.fs.settings.FsSettingsLoader;
@@ -29,7 +30,6 @@
 import java.io.ByteArrayInputStream;
 import java.io.File;
 import java.io.IOException;
-import java.io.InputStream;
 import java.nio.file.Files;
 import java.nio.file.Path;
 import java.nio.file.Paths;
@@ -194,11 +194,12 @@ void xmlIssue163() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(4)
+                .hasSize(5)
                 .containsEntry("Content-Type", "application/xml")
-                .containsEntry("resourceName", "issue-163.xml")
+                .containsEntry("Content-Type-Magic-Detected", "application/xml")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.DefaultParser")
-                .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser");
+                .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser")
+                .containsEntry("X-TIKA:resourceName", "issue-163.xml");
     }
 
     @Test
@@ -220,8 +221,10 @@ void extractFromDoc() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(25)
+                .hasSize(28)
+                .containsEntry("Content-Length", "24576")
                 .containsEntry("Content-Type", "application/msword")
+                .containsEntry("Content-Type-Magic-Detected", "application/msword")
                 .containsEntry("cp:category", "test")
                 .containsEntry("cp:revision", "2")
                 .containsEntry("custom:N° du document", "1234")
@@ -241,10 +244,11 @@ void extractFromDoc() throws IOException {
                 .containsEntry("meta:last-author", "David Pilato")
                 .containsEntry("meta:page-count", "2")
                 .containsEntry("meta:word-count", "19")
-                .containsEntry("resourceName", "test.doc")
+                .containsEntry("msoffice:comment-person-display-name", "Unknown")
                 .containsEntry("w:Comments", "Comments")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.DefaultParser")
                 .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser")
+                .containsEntry("X-TIKA:resourceName", "test.doc")
                 .containsEntry("xmpTPg:NPages", "2");
     }
 
@@ -268,9 +272,13 @@ void extractFromDocx() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(31)
+                .hasSize(34)
+                .containsEntry("Content-Length", "14242")
                 .containsEntry(
                         "Content-Type", "application/vnd.openxmlformats-officedocument.wordprocessingml.document")
+                .containsEntry(
+                        "Content-Type-Magic-Detected",
+                        "application/vnd.openxmlformats-officedocument.wordprocessingml.document")
                 .containsEntry("cp:category", "test")
                 .containsEntry("cp:revision", "4")
                 .containsEntry("custom:N° du document", "1234")
@@ -297,10 +305,11 @@ void extractFromDocx() throws IOException {
                 .containsEntry("meta:page-count", "2")
                 .containsEntry("meta:paragraph-count", "2")
                 .containsEntry("meta:word-count", "19")
-                .containsEntry("resourceName", "test.docx")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.DefaultParser")
                 .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser")
-                .containsEntry("xmpTPg:NPages", "2");
+                .containsEntry("X-TIKA:resourceName", "test.docx")
+                .containsEntry("xmpTPg:NPages", "2")
+                .containsEntry("zip:detectorZipFileOpened", "true");
     }
 
     @Test
@@ -321,22 +330,29 @@ void extractFromHtml() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(15)
-                .containsEntry("Content-Encoding", "UTF-8")
+                .hasSize(17)
+                .containsEntry("Content-Encoding", "x-MacRoman")
                 .containsEntry("Content-Location", "Web%20page")
-                .containsEntry("Content-Type", "text/html; charset=UTF-8")
+                .containsEntry("Content-Type", "text/html; charset=x-MacRoman")
                 .containsEntry("Content-Type-Hint", "text/html; charset=macintosh")
+                .containsEntry("Content-Type-Magic-Detected", "text/html")
                 .containsEntry("dc:title", "Test Tika title")
-                .containsEntry("Generator", "Microsoft Word 15")
-                .containsEntry("Mots clés", "keyword1, keyword2")
-                .containsEntry("Originator", "Microsoft Word 15")
-                .containsEntry("ProgId", "Word.Document")
-                .containsEntry("resourceName", "test.html")
-                .containsEntry("Titre", "Test Tika title")
-                .containsEntry("X-TIKA:detectedEncoding", "UTF-8")
-                .containsEntry("X-TIKA:encodingDetector", "UniversalEncodingDetector")
+                .containsEntry("html:Generator", "Microsoft Word 15")
+                // Tika 4 mis-decodes non-ASCII HTML metadata attribute names: the key "html:Mots clés" is
+                // returned as mojibake ("html:Mots clés") while its value stays correct. Tracked upstream in
+                // https://issues.apache.org/jira/browse/TIKA-4774 (open). Change with containsEntry when fixed:
+                .doesNotContainEntry("html:Mots clés", "keyword1, keyword2")
+                .containsEntry("html:Originator", "Microsoft Word 15")
+                .containsEntry("html:ProgId", "Word.Document")
+                .containsEntry("html:Titre", "Test Tika title")
+                .containsEntry("X-TIKA:detectedEncoding", "x-MacRoman")
+                .containsEntry(
+                        "X-TIKA:encodingDetectionTrace",
+                        "HtmlEncodingDetector->x-MacRoman[DECLARATIVE], MojibusterEncodingDetector->UTF-8[STRUCTURAL] [junk-filter-prefer-declarative]")
+                .containsEntry("X-TIKA:encodingDetector", "HtmlEncodingDetector")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.DefaultParser")
-                .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser");
+                .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser")
+                .containsEntry("X-TIKA:resourceName", "test.html");
     }
 
     /**
@@ -349,28 +365,19 @@ void extractFromMp3() throws IOException {
 
         // Extracted content
         Assertions.assertThat(doc.getContent()).contains("Test Tika");
-
-        // Content Type
-        Assertions.assertThat(doc.getFile().getContentType()).isEqualTo("audio/mpeg");
-
-        // Metadata
-        Assertions.assertThat(doc.getMeta().getAuthor()).isEqualTo("David Pilato");
-        Assertions.assertThat(doc.getMeta().getDate()).isNull();
-        Assertions.assertThat(doc.getMeta().getKeywords()).isNull();
-        Assertions.assertThat(doc.getMeta().getTitle()).isEqualTo("Test Tika");
-
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(19)
+                .hasSize(20)
                 .containsEntry("channels", "2")
                 .containsEntry("Content-Type", "audio/mpeg")
+                .containsEntry("Content-Type-Magic-Detected", "audio/mpeg")
                 .containsEntry("dc:creator", "David Pilato")
                 .containsEntry("dc:title", "Test Tika")
-                .containsEntry("resourceName", "test.mp3")
                 .containsEntry("samplerate", "44100")
                 .containsEntry("version", "MPEG 3 Layer III Version 1")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.DefaultParser")
                 .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser")
+                .containsEntry("X-TIKA:resourceName", "test.mp3")
                 .containsEntry("xmpDM:album", "FS Crawler")
                 .containsEntry("xmpDM:artist", "David Pilato")
                 .containsEntry("xmpDM:audioChannelType", "Stereo")
@@ -378,8 +385,19 @@ void extractFromMp3() throws IOException {
                 .containsEntry("xmpDM:audioSampleRate", "44100")
                 .containsEntry("xmpDM:duration", "1.0187751054763794")
                 .containsEntry("xmpDM:genre", "Vocal")
+                .containsKey("xmpDM:logComment")
                 .containsEntry("xmpDM:releaseDate", "2016")
                 .containsEntry("xmpDM:trackNumber", "1");
+
+        // Content Type
+        Assertions.assertThat(doc.getFile().getContentType()).isEqualTo("audio/mpeg");
+
+        // Metadata
+        Assertions.assertThat(doc.getMeta().getAuthor()).isEqualTo("David Pilato");
+        Assertions.assertThat(doc.getMeta().getDate()).isNull();
+        Assertions.assertThat(doc.getMeta().getKeywords()).isNull();
+        Assertions.assertThat(doc.getMeta().getTitle()).isEqualTo("Test Tika");
+
         Assertions.assertThat(raw)
                 .extractingByKey("xmpDM:logComment")
                 .satisfies(rawField -> Assertions.assertThat(rawField).containsAnyOf("Hello but reverted"));
@@ -404,8 +422,10 @@ void extractFromOdt() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(22)
+                .hasSize(25)
+                .containsEntry("Content-Length", "6236")
                 .containsEntry("Content-Type", "application/vnd.oasis.opendocument.text")
+                .containsEntry("Content-Type-Magic-Detected", "application/vnd.oasis.opendocument.text")
                 .containsEntry("cp:subject", "Test Tika Object")
                 .containsEntry("custom:Terminé le", "2016-07-06T22:00:00Z")
                 .containsEntry("dc:creator", "David Pilato")
@@ -423,10 +443,11 @@ void extractFromOdt() throws IOException {
                 .containsEntry("meta:paragraph-count", "1")
                 .containsEntry("meta:word-count", "12")
                 .containsEntry("odf:version", "1.2")
-                .containsEntry("resourceName", "test.odt")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.DefaultParser")
                 .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser")
-                .containsEntry("xmpTPg:NPages", "1");
+                .containsEntry("X-TIKA:resourceName", "test.odt")
+                .containsEntry("xmpTPg:NPages", "1")
+                .containsEntry("zip:detectorZipFileOpened", "true");
     }
 
     @Test
@@ -448,7 +469,7 @@ void extractFromPdf() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(42)
+                .hasSize(47)
                 .containsEntry("access_permission:assemble_document", "true")
                 .containsEntry("access_permission:can_modify", "true")
                 .containsEntry("access_permission:can_print", "true")
@@ -457,7 +478,9 @@ void extractFromPdf() throws IOException {
                 .containsEntry("access_permission:extract_for_accessibility", "true")
                 .containsEntry("access_permission:fill_in_form", "true")
                 .containsEntry("access_permission:modify_annotations", "true")
+                .containsEntry("Content-Length", "101643")
                 .containsEntry("Content-Type", "application/pdf")
+                .containsEntry("Content-Type-Magic-Detected", "application/pdf")
                 .containsEntry("dc:creator", "David Pilato")
                 .containsEntry("dc:format", "application/pdf; version=1.5")
                 .containsEntry("dc:language", "en-US")
@@ -476,18 +499,21 @@ void extractFromPdf() throws IOException {
                 .containsEntry("pdf:docinfo:subject", "Test Tika Object")
                 .containsEntry("pdf:docinfo:title", "Test Tika title")
                 .containsEntry("pdf:encrypted", "false")
+                .containsEntry("pdf:eofOffsets", "101460")
                 .containsEntry("pdf:hasCollection", "false")
                 .containsEntry("pdf:hasMarkedContent", "true")
                 .containsEntry("pdf:hasXFA", "false")
                 .containsEntry("pdf:hasXMP", "false")
+                .containsEntry("pdf:incrementalUpdateCount", "1")
                 .containsEntry("pdf:num3DAnnotations", "0")
                 .containsEntry("pdf:overallPercentageUnmappedUnicodeChars", "0.0")
                 .containsEntry("pdf:PDFVersion", "1.5")
                 .containsEntry("pdf:totalUnmappedUnicodeChars", "0")
                 .containsEntry("pdf:unmappedUnicodeCharsPerPage", "0")
-                .containsEntry("resourceName", "test.pdf")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.pdf.PDFParser")
                 .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.pdf.PDFParser")
+                .containsEntry("X-TIKA:resourceName", "test.pdf")
+                .containsEntry("X-TIKA:versionCount", "1")
                 .containsEntry("xmp:CreatorTool", "Microsoft Word")
                 .containsEntry("xmpTPg:NPages", "2");
         Assertions.assertThat(raw)
@@ -520,29 +546,10 @@ void extractFromJpg() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(64)
-                .containsKey("Component 1")
-                .containsKey("Component 2")
-                .containsKey("Component 3")
-                .containsEntry("Compression Type", "Baseline")
+                .hasSize(66)
+                .containsEntry("Content-Length", "41426")
                 .containsEntry("Content-Type", "image/jpeg")
-                .containsEntry("Data Precision", "8 bits")
-                .containsEntry("Exif IFD0:Orientation", "Top, left side (Horizontal / normal)")
-                .containsEntry("Exif IFD0:Resolution Unit", "Inch")
-                .containsKey("Exif IFD0:X Resolution")
-                .containsKey("Exif IFD0:Y Resolution")
-                .containsEntry("Exif IFD0:YCbCr Positioning", "Center of pixel array")
-                .containsEntry("Exif SubIFD:Color Space", "sRGB")
-                .containsEntry("Exif SubIFD:Components Configuration", "YCbCr")
-                .containsEntry("Exif SubIFD:Exif Image Height", "622 pixels")
-                .containsEntry("Exif SubIFD:Exif Image Width", "982 pixels")
-                .containsEntry("Exif SubIFD:Exif Version", "2.21")
-                .containsEntry("Exif SubIFD:FlashPix Version", "1.00")
-                .containsEntry("Exif SubIFD:Scene Capture Type", "Standard")
-                .containsKey("File Modified Date")
-                .hasEntrySatisfying(
-                        "File Name", value -> Assertions.assertThat(value).startsWith("apache-tika-"))
-                .containsEntry("File Size", "41426 bytes")
+                .containsEntry("Content-Type-Magic-Detected", "image/jpeg")
                 .containsKey("ICC:Apple Multi-language Profile Name")
                 .containsKey("ICC:Blue Colorant")
                 .containsKey("ICC:Blue Parametric TRC")
@@ -572,11 +579,31 @@ void extractFromJpg() throws IOException {
                 .containsKey("ICC:Version")
                 .containsKey("ICC:Video Card Gamma")
                 .containsKey("ICC:XYZ values")
-                .containsEntry("Image Height", "622 pixels")
-                .containsEntry("Image Width", "982 pixels")
-                .containsEntry("Number of Components", "3")
-                .containsEntry("Number of Tables", "4 Huffman tables")
-                .containsEntry("resourceName", "test.jpg")
+                .containsKey("img:Component 1")
+                .containsKey("img:Component 2")
+                .containsKey("img:Component 3")
+                .containsEntry("img:Compression Type", "Baseline")
+                .containsEntry("img:Data Precision", "8 bits")
+                .containsEntry("img:Exif IFD0:Orientation", "Top, left side (Horizontal / normal)")
+                .containsEntry("img:Exif IFD0:Resolution Unit", "Inch")
+                .containsKey("img:Exif IFD0:X Resolution")
+                .containsKey("img:Exif IFD0:Y Resolution")
+                .containsEntry("img:Exif IFD0:YCbCr Positioning", "Center of pixel array")
+                .containsEntry("img:Exif SubIFD:Color Space", "sRGB")
+                .containsEntry("img:Exif SubIFD:Components Configuration", "YCbCr")
+                .containsEntry("img:Exif SubIFD:Exif Image Height", "622 pixels")
+                .containsEntry("img:Exif SubIFD:Exif Image Width", "982 pixels")
+                .containsEntry("img:Exif SubIFD:Exif Version", "2.21")
+                .containsEntry("img:Exif SubIFD:FlashPix Version", "1.00")
+                .containsEntry("img:Exif SubIFD:Scene Capture Type", "Standard")
+                .containsKey("img:File Modified Date")
+                .hasEntrySatisfying(
+                        "img:File Name", value -> Assertions.assertThat(value).startsWith("apache-tika-"))
+                .containsEntry("img:File Size", "41426 bytes")
+                .containsEntry("img:Image Height", "622 pixels")
+                .containsEntry("img:Image Width", "982 pixels")
+                .containsEntry("img:Number of Components", "3")
+                .containsEntry("img:Number of Tables", "4 Huffman tables")
                 .containsEntry("tiff:BitsPerSample", "8")
                 .containsEntry("tiff:ImageLength", "622")
                 .containsEntry("tiff:ImageWidth", "982")
@@ -585,7 +612,8 @@ void extractFromJpg() throws IOException {
                 .containsEntry("tiff:XResolution", "144.0")
                 .containsEntry("tiff:YResolution", "144.0")
                 .containsKey("X-TIKA:Parsed-By")
-                .containsKey("X-TIKA:Parsed-By-Full-Set");
+                .containsKey("X-TIKA:Parsed-By-Full-Set")
+                .containsEntry("X-TIKA:resourceName", "test.jpg");
     }
 
     @Test
@@ -606,8 +634,9 @@ void extractFromRtf() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(15)
+                .hasSize(16)
                 .containsEntry("Content-Type", "application/rtf")
+                .containsEntry("Content-Type-Magic-Detected", "application/rtf")
                 .containsEntry("cp:category", "test")
                 .containsEntry("dc:creator", "David Pilato")
                 .containsEntry("dc:subject", "Test Tika Object")
@@ -618,9 +647,9 @@ void extractFromRtf() throws IOException {
                 .containsEntry("meta:keyword", "keyword1, keyword2")
                 .containsEntry("meta:page-count", "2")
                 .containsEntry("meta:word-count", "19")
-                .containsEntry("resourceName", "test.rtf")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.DefaultParser")
-                .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser");
+                .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser")
+                .containsEntry("X-TIKA:resourceName", "test.rtf");
         Assertions.assertThat(raw)
                 .containsKey("dcterms:created")
                 .extractingByKey("dcterms:created", InstanceOfAssertFactories.STRING)
@@ -645,14 +674,17 @@ void extractFromTxt() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(7)
-                .containsEntry("Content-Encoding", "ISO-8859-1")
-                .containsEntry("Content-Type", "text/plain; charset=ISO-8859-1")
-                .containsEntry("resourceName", "test.txt")
-                .containsEntry("X-TIKA:detectedEncoding", "ISO-8859-1")
-                .containsEntry("X-TIKA:encodingDetector", "UniversalEncodingDetector")
+                .hasSize(9)
+                .containsEntry("Content-Encoding", "windows-1252")
+                .containsEntry("Content-Type", "text/plain; charset=windows-1252")
+                .containsEntry("Content-Type-Magic-Detected", "text/plain")
+                .containsEntry("X-TIKA:detectedEncoding", "windows-1252")
+                .containsEntry(
+                        "X-TIKA:encodingDetectionTrace", "MojibusterEncodingDetector->windows-1252[STATISTICAL](0.10)")
+                .containsEntry("X-TIKA:encodingDetector", "MojibusterEncodingDetector")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.DefaultParser")
-                .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser");
+                .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser")
+                .containsEntry("X-TIKA:resourceName", "test.txt");
 
         Assertions.assertThat(doc.getAttachment()).isNull();
         Assertions.assertThat(doc.getFile().getChecksum()).isNull();
@@ -676,15 +708,16 @@ void extractFromWav() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(10)
+                .hasSize(11)
                 .containsEntry("bits", "16")
                 .containsEntry("channels", "2")
                 .containsEntry("Content-Type", "audio/vnd.wave")
+                .containsEntry("Content-Type-Magic-Detected", "audio/vnd.wave")
                 .containsEntry("encoding", "PCM_SIGNED")
-                .containsEntry("resourceName", "test.wav")
                 .containsEntry("samplerate", "44100.0")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.DefaultParser")
                 .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.DefaultParser")
+                .containsEntry("X-TIKA:resourceName", "test.wav")
                 .containsEntry("xmpDM:audioSampleRate", "44100")
                 .containsEntry("xmpDM:audioSampleType", "16Int");
     }
@@ -1080,35 +1113,15 @@ void ocrLanguageHeb() throws IOException {
     }
 
     @Test
-    void customTikaConfig() throws IOException {
-        InputStream tikaConfigIS = getClass().getResourceAsStream("/config/tikaConfig.xml");
-        Path testTikaConfig = testTmpDir.resolve("tika-config");
-        Files.createDirectories(testTikaConfig);
-        Files.copy(tikaConfigIS, testTikaConfig.resolve("tikaConfig.xml"));
-
+    void customTikaConfig() {
+        // Apache Tika 4 removed the XML Tika configuration file mechanism, so fs.tika_config_path is no longer
+        // supported. Setting it must fail fast with a clear configuration error rather than being silently ignored.
         FsSettings fsSettings = FsSettingsLoader.load();
-        fsSettings
-                .getFs()
-                .setTikaConfigPath(testTikaConfig.resolve("tikaConfig.xml").toString());
-
-        // Test that default parser for HTML is HTML parser
-        Doc doc = extractFromFile("test.html");
-        Assertions.assertThat(doc.getContent()).doesNotContain("Test Tika title");
-        Assertions.assertThat(doc.getContent()).contains("This second part of the text is in Page 2");
-
-        // Test HTML parser is never used, TXT parser used instead
-        doc = extractFromFile("test.html", fsSettings);
-        Assertions.assertThat(doc.getContent()).contains("<title>Test Tika title</title>");
-
-        // Test that default parser for XHTML is HTML parser
-        doc = extractFromFile("test.xhtml");
-        Assertions.assertThat(doc.getContent()).doesNotContain("Test Tika title");
-        Assertions.assertThat(doc.getContent()).contains("This is an example of XHTML");
-
-        // Test XML parser is used to parse XHTML
-        doc = extractFromFile("test.xhtml", fsSettings);
-        Assertions.assertThat(doc.getContent()).contains("Test Tika title");
-        Assertions.assertThat(doc.getContent()).doesNotContain("<title>Test Tika title</title>");
+        fsSettings.getFs().setTikaConfigPath("/any/path/tikaConfig.xml");
+
+        Assertions.assertThatThrownBy(() -> new TikaDocParser(fsSettings))
+                .isInstanceOf(FsCrawlerIllegalConfigurationException.class)
+                .hasMessageContaining("fs.tika_config_path is not supported with Apache Tika 4.x");
     }
 
     @Test
@@ -1146,7 +1159,7 @@ void pdfIssue1097() throws IOException {
 
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(36)
+                .hasSize(41)
                 .containsEntry("access_permission:assemble_document", "true")
                 .containsEntry("access_permission:can_modify", "true")
                 .containsEntry("access_permission:can_print", "true")
@@ -1155,7 +1168,9 @@ void pdfIssue1097() throws IOException {
                 .containsEntry("access_permission:extract_for_accessibility", "true")
                 .containsEntry("access_permission:fill_in_form", "true")
                 .containsEntry("access_permission:modify_annotations", "true")
+                .containsEntry("Content-Length", "13264")
                 .containsEntry("Content-Type", "application/pdf")
+                .containsEntry("Content-Type-Magic-Detected", "application/pdf")
                 .containsEntry("dc:creator", "Evangelos Vlachogiannis")
                 .containsEntry("dc:format", "application/pdf; version=1.4")
                 .containsEntry("dcterms:created", "2007-02-23T15:56:37Z")
@@ -1167,19 +1182,22 @@ void pdfIssue1097() throws IOException {
                 .containsEntry("pdf:docinfo:creator_tool", "Writer")
                 .containsEntry("pdf:docinfo:producer", "OpenOffice.org 2.1")
                 .containsEntry("pdf:encrypted", "false")
+                .containsEntry("pdf:eofOffsets", "13264")
                 .containsEntry("pdf:hasCollection", "false")
                 .containsEntry("pdf:hasMarkedContent", "false")
                 .containsEntry("pdf:hasXFA", "false")
                 .containsEntry("pdf:hasXMP", "false")
+                .containsEntry("pdf:incrementalUpdateCount", "0")
                 .containsEntry("pdf:num3DAnnotations", "0")
                 .containsEntry("pdf:overallPercentageUnmappedUnicodeChars", "0.0")
                 .containsEntry("pdf:PDFVersion", "1.4")
                 .containsEntry("pdf:producer", "OpenOffice.org 2.1")
                 .containsEntry("pdf:totalUnmappedUnicodeChars", "0")
                 .containsEntry("pdf:unmappedUnicodeCharsPerPage", "0")
-                .containsEntry("resourceName", "issue-1097.pdf")
                 .containsEntry("X-TIKA:Parsed-By", "org.apache.tika.parser.pdf.PDFParser")
                 .containsEntry("X-TIKA:Parsed-By-Full-Set", "org.apache.tika.parser.pdf.PDFParser")
+                .containsEntry("X-TIKA:resourceName", "issue-1097.pdf")
+                .containsEntry("X-TIKA:versionCount", "0")
                 .containsEntry("xmp:CreatorTool", "Writer")
                 .containsEntry("xmpTPg:NPages", "1");
         Assertions.assertThat(raw)
@@ -1204,9 +1222,10 @@ void emptyFileIssue834() throws IOException {
         // Metadata
         Map<String, String> raw = doc.getMeta().getRaw();
         Assertions.assertThat(raw)
-                .hasSize(2)
+                .hasSize(3)
                 .containsEntry("Content-Type", "text/plain")
-                .containsEntry("resourceName", "issue-834.txt");
+                .containsEntry("Content-Type-Magic-Detected", "text/plain")
+                .containsEntry("X-TIKA:resourceName", "issue-834.txt");
     }
 
     /** Test protected document */