TIKA-2645 -- make pool methods private for better encapsulation and add

a pool for DOM building
apache · May 23, 2018 · cdca0f7 · cdca0f7
1 parent c40045a
commit cdca0f7
Show file tree

Hide file tree

Showing 24 changed files with 297 additions and 253 deletions.
diff --git a/CHANGES.txt b/CHANGES.txt
@@ -1,13 +1,14 @@
 Release 1.19 ???
 
-
-   * Use a pool for SAXParsers rather than creating a new one for every parse.
+   * Use a pool for SAXParsers and DOMBuilders rather than creating
+     a new parser/builder for every parse.
      For better performance, set XMLReaderUtils.setPoolSize() to the
      number of threads you're using with Tika (TIKA-2645.
 
    * Add the RecursiveParserWrapperHandler to improve the RecursiveParserWrapper
      API slightly (TIKA-2644).
 
+
 Release 1.18 - 4/20/2018
 
    * Upgrade Jackson to 2.9.5 (TIKA-2634).
@@ -68,11 +69,26 @@ Release 1.18 - 4/20/2018
    * Fixed bug where TesseractOCRParser ignores configured ImageMagickPath,
      and set rotation script to ignore Python warnings (TIKA-2509)
 
-   * Upgrade geo-apis to 3.0.1 (TIKA-2535).
+   * Upgrade geo-apis to 3.0.1 (TIKA-2535)
+
+   * Mime definition and magic improvements for text-based programming
+     and config formats (TIKA-2554, TIKA-2567, TIKA-1141)
 
    * Added local Docker image build using dockerfile-maven-plugin to allow
      images to be built from source (TIKA-1518).
 
+   * Support for SAS7BDAT data files (TIKA-2462)
+
+   * Handle .epub files using .htm rather than .html extensions for the
+     embedded contents (TIKA-1288)
+
+   * Mime magic for ACES Images (TIKA-2628) and DPX Images (TIKA-2629)
+
+   * For sparse XLSX and XLSB files, always output missing cells to
+     the left of filled ones (matching XLS), and optionally output
+     missing rows on all 3 formats if requested via the
+     OfficeParserContext (TIKA-2479)
+
 Release 1.17 - 12/8/2017
 
   ***NOTE: THIS IS THE LAST VERSION OF TIKA THAT WILL RUN

diff --git a/tika-core/src/main/java/org/apache/tika/detect/XmlRootExtractor.java b/tika-core/src/main/java/org/apache/tika/detect/XmlRootExtractor.java
@@ -24,6 +24,7 @@
 
 import org.apache.tika.exception.TikaException;
 import org.apache.tika.io.CloseShieldInputStream;
+import org.apache.tika.parser.ParseContext;
 import org.apache.tika.sax.OfflineContentHandler;
 import org.apache.tika.utils.XMLReaderUtils;
 import org.xml.sax.Attributes;
@@ -37,6 +38,7 @@
  * @since Apache Tika 0.4
  */
 public class XmlRootExtractor {
+    private static final ParseContext EMPTY_CONTEXT = new ParseContext();
 
     public QName extractRootElement(byte[] data) {
         return extractRootElement(new ByteArrayInputStream(data));
@@ -47,15 +49,11 @@ public QName extractRootElement(byte[] data) {
      */
     public QName extractRootElement(InputStream stream) {
         ExtractorHandler handler = new ExtractorHandler();
-        SAXParser parser = null;
         try {
-            parser = XMLReaderUtils.acquireSAXParser();
-            parser.parse(
+            XMLReaderUtils.parseSAX(
                     new CloseShieldInputStream(stream),
-                    new OfflineContentHandler(handler));
+                    new OfflineContentHandler(handler), EMPTY_CONTEXT);
         } catch (Exception ignore) {
-        } finally {
-                XMLReaderUtils.releaseParser(parser);
         }
         return handler.rootElement;
     }

diff --git a/tika-core/src/main/java/org/apache/tika/mime/MimeTypesReader.java b/tika-core/src/main/java/org/apache/tika/mime/MimeTypesReader.java
@@ -319,7 +319,7 @@ public List<Clause> getClauses() {
      * @return a SAXParser
      * @throws TikaException
      */
-    public static SAXParser acquireSAXParser()
+    private static SAXParser acquireSAXParser()
             throws TikaException {
         while (true) {
             SAXParser parser = null;
@@ -343,7 +343,7 @@ public static SAXParser acquireSAXParser()
      *
      * @param parser parser to return
      */
-    public static void releaseParser(SAXParser parser) {
+    private static void releaseParser(SAXParser parser) {
         try {
             parser.reset();
         } catch (UnsupportedOperationException e) {

diff --git a/tika-core/src/main/java/org/apache/tika/parser/NetworkParser.java b/tika-core/src/main/java/org/apache/tika/parser/NetworkParser.java
@@ -37,6 +37,7 @@
 import org.apache.tika.sax.OfflineContentHandler;
 import org.apache.tika.sax.TaggedContentHandler;
 import org.apache.tika.sax.TeeContentHandler;
+import org.apache.tika.utils.XMLReaderUtils;
 import org.xml.sax.Attributes;
 import org.xml.sax.ContentHandler;
 import org.xml.sax.SAXException;
@@ -126,12 +127,10 @@ public void parse(
 
             TaggedContentHandler tagged = new TaggedContentHandler(
                     new OfflineContentHandler(handler));
-            SAXParser parser = null;
             try {
-                parser = context.acquireSAXParser();
-                parser.parse(
+                XMLReaderUtils.parseSAX(
                         stream, new TeeContentHandler(
-                                tagged, new MetaHandler(metadata)));
+                                tagged, new MetaHandler(metadata)), context);
             } catch (SAXException e) {
                 tagged.throwIfCauseOf(e);
                 throw new TikaException(
@@ -140,7 +139,6 @@ stream, new TeeContentHandler(
                 throw new TikaException(
                         "Unable to read network parser output", e);
             } finally {
-                context.releaseParser(parser);
                 try {
                     thread.join(1000);
                 } catch (InterruptedException e) {

diff --git a/tika-core/src/main/java/org/apache/tika/parser/ParseContext.java b/tika-core/src/main/java/org/apache/tika/parser/ParseContext.java
@@ -23,14 +23,9 @@
 import javax.xml.parsers.SAXParser;
 import javax.xml.parsers.SAXParserFactory;
 import javax.xml.stream.XMLInputFactory;
-import javax.xml.stream.XMLResolver;
-import javax.xml.stream.XMLStreamException;
 import javax.xml.transform.Transformer;
-import javax.xml.transform.TransformerConfigurationException;
-import javax.xml.transform.TransformerFactory;
-import javax.xml.transform.TransformerFactoryConfigurationError;
 
-import java.io.IOException;
+import java.io.InputStream;
 import java.io.Serializable;
 import java.util.HashMap;
 import java.util.Map;
@@ -40,6 +35,7 @@
 import org.xml.sax.SAXNotRecognizedException;
 import org.xml.sax.SAXNotSupportedException;
 import org.xml.sax.XMLReader;
+import org.xml.sax.helpers.DefaultHandler;
 
 /**
  * Parse context. Used to pass context information to Tika parsers.
@@ -121,7 +117,9 @@ public XMLReader getXMLReader() throws TikaException {
     /**
      * Returns the SAX parser specified in this parsing context. If a parser
      * is not explicitly specified, then one is created using the specified
-     * or the default SAX parser factory.
+     * or the default SAX parser factory. Consider using
+     * {@link XMLReaderUtils#parseSAX(InputStream, DefaultHandler, ParseContext)}
+     * for more efficient reuse of SAXParsers.
      *
      * @see #getSAXParserFactory()
      * @since Apache Tika 0.8
@@ -137,39 +135,6 @@ public SAXParser getSAXParser() throws TikaException {
         }
     }
 
-    /**
-     * Returns the SAX parser specified in this parsing context. If a parser
-     * is not explicitly specified, then one is acquired from the pool.
-     * <p>
-     * Make sure to {@link #releaseParser(SAXParser)} as the
-     * first call in a <code>finally</code> block every time
-     * you call this!
-     * </p>
-     *
-     * @return SAXParser
-     * @throws TikaException
-     */
-    public SAXParser acquireSAXParser() throws TikaException {
-        if (context.containsKey(SAXParser.class)) {
-            return get(SAXParser.class);
-        }
-        return XMLReaderUtils.acquireSAXParser();
-    }
-
-    /**
-     * If the context already has a SAXParser, this is a no-op.
-     * Otherwise, this returns the parser to the pool
-     *
-     * @param parser
-     * @throws TikaException
-     */
-    public void releaseParser(SAXParser parser) {
-        if (context.containsKey(SAXParser.class)) {
-            return;
-        }
-        XMLReaderUtils.releaseParser(parser);
-    }
-
     /**
      * Returns the SAX parser factory specified in this parsing context.
      * If a factory is not explicitly specified, then a default factory
@@ -228,6 +193,8 @@ private DocumentBuilderFactory getDocumentBuilderFactory() {
      * instance is created and returned. The builder instance is
      * configured to apply an {@link XMLReaderUtils#IGNORING_SAX_ENTITY_RESOLVER},
      * and it sets the ErrorHandler to <code>null</code>.
+     * Consider using {@link XMLReaderUtils#buildDOM(InputStream, ParseContext)}
+     * instead for more efficient reuse of document builders.
      *
      * @since Apache Tika 1.13
      * @return DOM Builder