This is an automated email from the ASF dual-hosted git repository.

tballison pushed a commit to branch branch_3x
in repository https://gitbox.apache.org/repos/asf/tika.git


The following commit(s) were added to refs/heads/branch_3x by this push:
     new 7bcd3d7827 Dependency updates branch 3x (#3629)
7bcd3d7827 is described below

commit 7bcd3d782791a7412bdb6defab18a342b60fac1a
Author: Tim Allison <[email protected]>
AuthorDate: Tue Nov 17 09:02:38 2026 -0500

    Dependency updates branch 3x (#3629)
    
    * Dependency updates for branch_3x; juniversalchardet 2.5.2
    
    * TIKA-4962 follow-up: gate case-insensitive html magic on markup-like file 
starts
---
 .mvn/wrapper/maven-wrapper.properties              |  2 +-
 CHANGES.txt                                        |  3 ++
 tika-app/pom.xml                                   |  2 +-
 .../org/apache/tika/mime/tika-mimetypes.xml        | 50 ++++++++++++++++++++--
 .../org/apache/tika/mime/MimeDetectionTest.java    | 43 +++++++++++++++++++
 tika-parent/pom.xml                                | 18 ++++----
 .../tika/parser/txt/UniversalEncodingListener.java |  4 ++
 .../org/apache/tika/parser/txt/TXTParserTest.java  | 16 +++++++
 tika-server/pom.xml                                |  2 +-
 tika-server/tika-server-core/pom.xml               |  2 +-
 10 files changed, 126 insertions(+), 16 deletions(-)

diff --git a/.mvn/wrapper/maven-wrapper.properties 
b/.mvn/wrapper/maven-wrapper.properties
index bb58b3c263..0d710c43f2 100644
--- a/.mvn/wrapper/maven-wrapper.properties
+++ b/.mvn/wrapper/maven-wrapper.properties
@@ -1,3 +1,3 @@
 wrapperVersion=3.3.4
 distributionType=only-script
-distributionUrl=https://repo.maven.apache.org/maven2/org/apache/maven/apache-maven/3.10.0/apache-maven-3.10.0-bin.zip
+distributionUrl=https://repo.maven.apache.org/maven2/org/apache/maven/apache-maven/3.10.1/apache-maven-3.10.1-bin.zip
diff --git a/CHANGES.txt b/CHANGES.txt
index ee7e520d7b..70eeff45dc 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -1,5 +1,8 @@
 Release 3.4.0 - ???
 
+  * UTF-8 text that was misdetected as GB18030 is now detected as UTF-8
+    (juniversalchardet 2.5.2).
+
   * The http fetcher, the Solr and OpenSearch emitters and the Solr pipes 
iterator no longer
     send Authorization or Cookie headers, or answer a 401 with the configured 
password,
     after a redirect to another host, port or scheme. That includes http to 
https, so use
diff --git a/tika-app/pom.xml b/tika-app/pom.xml
index 7c21c9a15d..3b201072d1 100644
--- a/tika-app/pom.xml
+++ b/tika-app/pom.xml
@@ -243,7 +243,7 @@
           <plugin>
             <groupId>org.codehaus.mojo</groupId>
             <artifactId>build-helper-maven-plugin</artifactId>
-            <version>3.6.1</version>
+            <version>3.8.0</version>
             <executions>
               <execution>
                 <phase>package</phase>
diff --git 
a/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml 
b/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml
index 0c075a2271..2f12d5e793 100644
--- a/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml
+++ b/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml
@@ -7752,8 +7752,45 @@
       <match value="&lt;head" type="string" offset="0:64"/>
       <match value="&lt;TITLE" type="string" offset="0:64"/>
       <match value="&lt;title" type="string" offset="0:64"/>
-      <!-- the tag must end: <htmlPath> in an XML file is not HTML -->
-      <match value="(?i)&lt;html[\\s>]" type="regex" offset="0:128"/>
+      <!-- the tag must end: <htmlPath> in an XML file is not HTML. Short 
strings
+           rather than a regex: a pattern longer than xhtml's "<html xmlns=" 
would
+           sort ahead of it at this priority. -->
+      <match value="&lt;html " type="string" offset="0:128"/>
+      <match value="&lt;html>" type="string" offset="0:128"/>
+      <match value="&lt;html\n" type="string" offset="0:128"/>
+      <match value="&lt;html\r" type="string" offset="0:128"/>
+      <match value="&lt;html\x09" type="string" offset="0:128"/>
+      <match value="&lt;HTML " type="string" offset="0:64"/>
+      <match value="&lt;HTML>" type="string" offset="0:64"/>
+      <match value="&lt;HTML\n" type="string" offset="0:64"/>
+      <match value="&lt;HTML\r" type="string" offset="0:64"/>
+      <match value="&lt;HTML\x09" type="string" offset="0:64"/>
+      <!-- past 64 bytes, uppercase only when the file starts like markup (see 
below) -->
+      <match value="&lt;HTML " type="string" offset="64:128">
+        <match value="&lt;[^?]" type="regex" offset="0"/>
+        <match value="\\s" type="regex" offset="0"/>
+        <match value="0xEFBBBF" type="string" offset="0"/>
+      </match>
+      <match value="&lt;HTML>" type="string" offset="64:128">
+        <match value="&lt;[^?]" type="regex" offset="0"/>
+        <match value="\\s" type="regex" offset="0"/>
+        <match value="0xEFBBBF" type="string" offset="0"/>
+      </match>
+      <match value="&lt;HTML\n" type="string" offset="64:128">
+        <match value="&lt;[^?]" type="regex" offset="0"/>
+        <match value="\\s" type="regex" offset="0"/>
+        <match value="0xEFBBBF" type="string" offset="0"/>
+      </match>
+      <match value="&lt;HTML\r" type="string" offset="64:128">
+        <match value="&lt;[^?]" type="regex" offset="0"/>
+        <match value="\\s" type="regex" offset="0"/>
+        <match value="0xEFBBBF" type="string" offset="0"/>
+      </match>
+      <match value="&lt;HTML\x09" type="string" offset="64:128">
+        <match value="&lt;[^?]" type="regex" offset="0"/>
+        <match value="\\s" type="regex" offset="0"/>
+        <match value="0xEFBBBF" type="string" offset="0"/>
+      </match>
     </magic>
     <magic priority="20">
       <!-- Lower priority match for <html anywhere near the top of the file -->
@@ -7763,7 +7800,14 @@
            MimeTypes#getMinLength() bytes. It should also only start after
            the higher priority "start of file" one above
        -->
-      <match value="(?i)&lt;html[\\s>]" type="regex" offset="128:8192"/>
+      <match value="&lt;html[\\s>]" type="regex" offset="128:8192"/>
+      <!-- other cases only when the file starts like markup, so that code
+           building "<HTML>" strings or heredocs is not taken for html -->
+      <match value="(?i)&lt;html[\\s>]" type="regex" offset="128:8192">
+        <match value="&lt;[^?]" type="regex" offset="0"/>
+        <match value="\\s" type="regex" offset="0"/>
+        <match value="0xEFBBBF" type="string" offset="0"/>
+      </match>
     </magic>
     <glob pattern="*.html"/>
     <glob pattern="*.htm"/>
diff --git 
a/tika-core/src/test/java/org/apache/tika/mime/MimeDetectionTest.java 
b/tika-core/src/test/java/org/apache/tika/mime/MimeDetectionTest.java
index acbc583dd5..2c75a2e258 100644
--- a/tika-core/src/test/java/org/apache/tika/mime/MimeDetectionTest.java
+++ b/tika-core/src/test/java/org/apache/tika/mime/MimeDetectionTest.java
@@ -112,6 +112,49 @@ public class MimeDetectionTest {
         }
     }
 
+    /** An uppercase html tag past the first 128 bytes counts only if the file 
starts like markup. */
+    @Test
+    public void testDeepUppercaseHtml() throws Exception {
+        String pad = "x = 1;\n".repeat(30);
+        String js = "var w = window.open();\n" + pad + 
"w.document.write(\"<HTML><BODY>\");\n";
+        assertEquals(MediaType.TEXT_PLAIN, detect(js, null));
+        assertEquals(MediaType.text("javascript"), detect(js, "popup.js"));
+
+        String perl = "package Foo;\n" + pad + "print 
<<EOF;\n<HTML>\n<HEAD><TITLE>x</TITLE>\n";
+        assertEquals(MediaType.text("x-perl"), detect(perl, "Foo.pm"));
+
+        String php = "<?\n" + pad + "?>\n<HTML>\n<TITLE>x</TITLE>\n";
+        assertEquals(MediaType.text("x-php"), detect(php, "index.php"));
+
+        // within the first 128 bytes too
+        String perlNear = "#\n# Sort hosts\n#\n&make_info();\n\nprint CLIENT 
<<EOF;\n" +
+                "    ".repeat(5) + "<HTML>\n<HEAD><TITLE>x</TITLE>\n";
+        assertEquals(MediaType.text("x-perl"), detect(perlNear, 
"satan_counts.pl"));
+        String phpNear = "<? include(\"dbconnect.php3\"); ?>\n<? $sitename = 
doins_var(\"name\");\n" +
+                "print \"<HTML> <HEAD><TITLE>About</TITLE> </HEAD>\";";
+        assertEquals(MediaType.text("x-php"), detect(phpNear, "about.php3"));
+        assertEquals(MediaType.TEXT_HTML,
+                detect("<!-- top.src START -->" + " ".repeat(60) + 
"<HTML>\n<HEAD>", null));
+
+        String[] pages = {
+                "\r\n\t".repeat(100) + "<HTML>\n<BODY>x</BODY></HTML>",
+                "<SCRIPT SRC=\"/cookie.js\"></SCRIPT>\n" + " ".repeat(200) + 
"<HTML>\n<BODY>x</BODY>",
+                "" + " ".repeat(200) + "<HTML>\n<BODY>x</BODY>"
+        };
+        for (String p : pages) {
+            assertEquals(MediaType.TEXT_HTML, detect(p, null), p);
+        }
+    }
+
+    /** A bare html tag must not outrank xhtml's "html xmlns=" match, and a 
.html name still wins. */
+    @Test
+    public void testXhtmlNamespaceWithoutDoctype() throws Exception {
+        String xhtml = "<!DOCTYPE PUBLIC \"-//W3C//DTD HTML 4.01//EN\">\n" +
+                "<html xmlns=\"http://www.w3.org/1999/xhtml\";>\n<head><meta 
charset=utf-8></head>";
+        assertEquals(MediaType.application("xhtml+xml"), detect(xhtml, null));
+        assertEquals(MediaType.TEXT_HTML, detect(xhtml, "page.html"));
+    }
+
     private static MediaType detect(String content, String name) throws 
IOException {
         Metadata metadata = new Metadata();
         if (name != null) {
diff --git a/tika-parent/pom.xml b/tika-parent/pom.xml
index 6262f69a27..5f82f2bec8 100644
--- a/tika-parent/pom.xml
+++ b/tika-parent/pom.xml
@@ -316,7 +316,7 @@
     <!-- dependency versions -->
     <!-- change threetenbp exclusion version -->
     <google.cloud.version>2.78.0</google.cloud.version>
-    <aws2.version>2.59.0</aws2.version>
+    <aws2.version>2.59.1</aws2.version>
     <!-- WARNING: when you upgrade asm make sure that you update the
         OpCode in the initializer in 
org.apache.tika.parser.asm.XHTMLClassVisitor
         See TIKA-2992.
@@ -394,8 +394,7 @@
     <jsr305.version>3.0.2</jsr305.version>
     <!-- junit 6.x requires jdk 17 -->
     <junit5.version>5.14.4</junit5.version>
-    <!-- TODO 2.5.1 fails in  Apache Tika text parser module -->
-    <juniversalchardet.version>2.5.0</juniversalchardet.version>
+    <juniversalchardet.version>2.5.2</juniversalchardet.version>
     <junrar.version>8.1.2</junrar.version>
     <jwarc.version>0.37.0</jwarc.version>
     <kafka.version>4.4.0</kafka.version>
@@ -426,8 +425,9 @@
     <parso.version>2.0.14</parso.version>
     <pax.exam.version>4.14.0</pax.exam.version>
     <pdfbox.version>3.0.8</pdfbox.version>
+    <!-- 6.x requires jdk17 -->
     <poi.version>5.5.1</poi.version>
-    <protobuf.version>4.33.6</protobuf.version>
+    <protobuf.version>4.36.2</protobuf.version>
     <quartz.version>2.5.2</quartz.version>
     <reactor.core.version>3.8.8</reactor.core.version>
     <reactor.netty.version>1.3.8</reactor.netty.version>
@@ -455,10 +455,10 @@
     <woodstox.core.version>7.3.0</woodstox.core.version>
     <xmpcore.version>6.1.11</xmpcore.version>
     <!-- This has to be < 3.9 to work with Solr 8's integration tests (doesn't 
fail on windows) -->
-    <zookeeper.version>3.8.7</zookeeper.version>
-    <zstd.version>1.5.7-26</zstd.version>
+    <zookeeper.version>3.8.8</zookeeper.version>
+    <zstd.version>1.5.7-28</zstd.version>
     <nimbus-jose-jwt.version>10.10</nimbus-jose-jwt.version>
-    <logback.version>1.6.5</logback.version>
+    <logback.version>1.6.6</logback.version>
     <javacpp.version>1.5.14</javacpp.version>
     <maven.site.version>3.22.0</maven.site.version>
     <maven.exec.version>3.7.0</maven.exec.version>
@@ -1128,7 +1128,7 @@
       <plugin>
         <groupId>org.apache.maven.plugins</groupId>
         <artifactId>maven-enforcer-plugin</artifactId>
-        <version>3.6.3</version>
+        <version>3.7.0</version>
       </plugin>
       </plugins>
     </pluginManagement>
@@ -1153,7 +1153,7 @@
       <plugin>
         <groupId>org.codehaus.mojo</groupId>
         <artifactId>versions-maven-plugin</artifactId>
-        <version>2.25.1</version>
+        <version>2.25.2</version>
         <configuration>
           <generateBackupPoms>false</generateBackupPoms>
         </configuration>
diff --git 
a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-text-module/src/main/java/org/apache/tika/parser/txt/UniversalEncodingListener.java
 
b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-text-module/src/main/java/org/apache/tika/parser/txt/UniversalEncodingListener.java
index 8458aabe8d..d20f88dfbb 100644
--- 
a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-text-module/src/main/java/org/apache/tika/parser/txt/UniversalEncodingListener.java
+++ 
b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-text-module/src/main/java/org/apache/tika/parser/txt/UniversalEncodingListener.java
@@ -56,6 +56,10 @@ class UniversalEncodingListener implements CharsetListener {
     }
 
     public void report(String name) {
+        // juniversalchardet 2.5.2 reports US-ASCII; bytes past the probe may 
not be
+        if (Constants.CHARSET_US_ASCII.equals(name)) {
+            name = Constants.CHARSET_WINDOWS_1252;
+        }
         if (Constants.CHARSET_WINDOWS_1252.equals(name)) {
             if (hint != null) {
                 // Use the encoding hint when available
diff --git 
a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-text-module/src/test/java/org/apache/tika/parser/txt/TXTParserTest.java
 
b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-text-module/src/test/java/org/apache/tika/parser/txt/TXTParserTest.java
index f39be16c47..6a04c29426 100644
--- 
a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-text-module/src/test/java/org/apache/tika/parser/txt/TXTParserTest.java
+++ 
b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-text-module/src/test/java/org/apache/tika/parser/txt/TXTParserTest.java
@@ -87,6 +87,22 @@ public class TXTParserTest extends TikaTest {
         assertEquals("\n", handler.toString());
     }
 
+    // juniversalchardet reports US-ASCII from the probe alone; the tail must 
still decode
+    @Test
+    public void testAsciiPrefixLatin1Tail() throws Exception {
+        StringBuilder sb = new StringBuilder();
+        while (sb.length() < 20000) {
+            sb.append("plain ascii text\n");
+        }
+        sb.append("caf\u00e9\n");
+        Metadata metadata = new Metadata();
+        BodyContentHandler handler = new BodyContentHandler(-1);
+        parser.parse(new 
ByteArrayInputStream(sb.toString().getBytes(ISO_8859_1)), handler,
+                metadata, new ParseContext());
+        assertEquals("text/plain; charset=ISO-8859-1", 
metadata.get(Metadata.CONTENT_TYPE));
+        assertContains("caf\u00e9", handler.toString());
+    }
+
     /**
      * Test for the heuristics that we use to assign an eight-bit character
      * encoding to mostly ASCII sequences. If a more specific match can not
diff --git a/tika-server/pom.xml b/tika-server/pom.xml
index e37c717f0a..58c3499ce3 100644
--- a/tika-server/pom.xml
+++ b/tika-server/pom.xml
@@ -41,7 +41,7 @@
     <maven.compiler.target>17</maven.compiler.target>
     <maven.compiler.release>17</maven.compiler.release>
     <cxf.version>4.2.4</cxf.version>
-    <jetty.version>12.1.14</jetty.version>
+    <jetty.version>12.1.15</jetty.version>
     <jakarta.ws.rs.version>4.0.0</jakarta.ws.rs.version>
   </properties>
 
diff --git a/tika-server/tika-server-core/pom.xml 
b/tika-server/tika-server-core/pom.xml
index 3dc6fe5834..0a74a801ec 100644
--- a/tika-server/tika-server-core/pom.xml
+++ b/tika-server/tika-server-core/pom.xml
@@ -304,7 +304,7 @@
           <plugin>
             <groupId>org.codehaus.mojo</groupId>
             <artifactId>exec-maven-plugin</artifactId>
-            <version>{maven.exec.version}</version>
+            <version>${maven.exec.version}</version>
             <executions>
               <execution>
                 <phase>test</phase>

Reply via email to