Author: tallison
Date: Tue May 27 19:33:07 2014
New Revision: 1597856
URL: http://svn.apache.org/r1597856
Log:
TIKA-1294 add ability to turn off image extraction from PDFs
Modified:
tika/trunk/CHANGES.txt
tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaCoreProperties.java
tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaMetadataKeys.java
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDF2XHTML.java
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParserConfig.java
tika/trunk/tika-parsers/src/main/resources/org/apache/tika/parser/pdf/PDFParser.properties
tika/trunk/tika-parsers/src/test/java/org/apache/tika/TikaTest.java
tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
Modified: tika/trunk/CHANGES.txt
URL:
http://svn.apache.org/viewvc/tika/trunk/CHANGES.txt?rev=1597856&r1=1597855&r2=1597856&view=diff
==============================================================================
--- tika/trunk/CHANGES.txt (original)
+++ tika/trunk/CHANGES.txt Tue May 27 19:33:07 2014
@@ -1,5 +1,8 @@
Release 1.6 - ??/??/2014
+ * Added ability to turn off image extraction from PDFs (TIKA-1294).
+ Users must now turn on this capability via the PDFParserConfig.
+
* Upgrade to PDFBox 1.8.5 (TIKA-1290, TIKA-1231, TIKA-1233)
* Zip Container Detection for DWFX and XPS formats, which are OPC
Modified:
tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaCoreProperties.java
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaCoreProperties.java?rev=1597856&r1=1597855&r2=1597856&view=diff
==============================================================================
---
tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaCoreProperties.java
(original)
+++
tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaCoreProperties.java
Tue May 27 19:33:07 2014
@@ -36,6 +36,22 @@ package org.apache.tika.metadata;
*/
@SuppressWarnings("deprecation")
public interface TikaCoreProperties {
+
+ /**
+ * A file might contain different types of embedded documents.
+ * The most common is the attachment.
+ * An "inline" embedded resource should be used for embedded image
+ * files that are used to render the page image (as in PDXObjImages in PDF
files).
+ * <p>
+ * Not all parsers have yet implemented this
+ *
+ */
+ public enum EmbeddedResourceType {
+ inline,
+ attachment
+ };
+
+
/**
* @see DublinCore#FORMAT
*/
@@ -246,5 +262,13 @@ public interface TikaCoreProperties {
@Deprecated
public static final Property TRANSITION_SUBJECT_TO_OO_SUBJECT =
Property.composite(OfficeOpenXMLCore.SUBJECT,
new Property[] { Property.internalText(Metadata.SUBJECT) });
+
+ /**
+ * See {@link #EMBEDDED_RESOURCE_TYPE}
+ */
+ public static final Property EMBEDDED_RESOURCE_TYPE =
+
Property.internalClosedChoise(TikaMetadataKeys.EMBEDDED_RESOURCE_TYPE,
+ new String[]{EmbeddedResourceType.attachment.toString(),
EmbeddedResourceType.inline.toString()});
+
}
Modified:
tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaMetadataKeys.java
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaMetadataKeys.java?rev=1597856&r1=1597855&r2=1597856&view=diff
==============================================================================
---
tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaMetadataKeys.java
(original)
+++
tika/trunk/tika-core/src/main/java/org/apache/tika/metadata/TikaMetadataKeys.java
Tue May 27 19:33:07 2014
@@ -26,4 +26,7 @@ public interface TikaMetadataKeys {
String PROTECTED = "protected";
String EMBEDDED_RELATIONSHIP_ID = "embeddedRelationshipId";
+
+ String EMBEDDED_RESOURCE_TYPE = "embeddedResourceType";
+
}
Modified:
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDF2XHTML.java
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDF2XHTML.java?rev=1597856&r1=1597855&r2=1597856&view=diff
==============================================================================
---
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDF2XHTML.java
(original)
+++
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDF2XHTML.java
Tue May 27 19:33:07 2014
@@ -22,10 +22,12 @@ import java.io.IOException;
import java.io.Writer;
import java.text.SimpleDateFormat;
import java.util.Calendar;
+import java.util.HashSet;
import java.util.Iterator;
import java.util.List;
import java.util.ListIterator;
import java.util.Map;
+import java.util.Set;
import java.util.TreeMap;
import org.apache.pdfbox.pdmodel.PDDocument;
@@ -63,6 +65,7 @@ import org.apache.tika.extractor.Parsing
import org.apache.tika.io.IOExceptionWithCause;
import org.apache.tika.io.TikaInputStream;
import org.apache.tika.metadata.Metadata;
+import org.apache.tika.metadata.TikaCoreProperties;
import org.apache.tika.parser.ParseContext;
import org.apache.tika.sax.EmbeddedContentHandler;
import org.apache.tika.sax.XHTMLContentHandler;
@@ -93,6 +96,14 @@ class PDF2XHTML extends PDFTextStripper
private boolean inParagraph = false;
/**
+ * This keeps track of the pdf object ids for inline
+ * images that have been processed. If {@link
PDFParserConfig#getExtractUniqueInlineImagesOnly()
+ * is true, this will be checked before extracting an embedded image.
+ */
+ private Set<String> processedInlineImages = new HashSet<String>();
+
+
+ /**
* Converts the given PDF document (and related metadata) to a stream
* of XHTML SAX events sent to the given content handler.
*
@@ -279,14 +290,27 @@ class PDF2XHTML extends PDFTextStripper
}
private void extractImages(PDResources resources) throws SAXException {
- if (resources == null) {
+ if (resources == null || config.getExtractInlineImages() == false) {
return;
}
- for (PDXObject object : resources.getXObjects().values()) {
+ for (Map.Entry<String, PDXObject> entry :
resources.getXObjects().entrySet()) {
+
+ PDXObject object = entry.getValue();
if (object instanceof PDXObjectForm) {
extractImages(((PDXObjectForm) object).getResources());
} else if (object instanceof PDXObjectImage) {
+
+ //Do we only want to process unique COSObject ids?
+ //If so, have we already processed this one?
+ if (config.getExtractUniqueInlineImagesOnly() == true) {
+ String cosObjectId = entry.getKey();
+ if (processedInlineImages.contains(cosObjectId)){
+ continue;
+ }
+ processedInlineImages.add(cosObjectId);
+ }
+
PDXObjectImage image = (PDXObjectImage) object;
Metadata metadata = new Metadata();
@@ -297,6 +321,8 @@ class PDF2XHTML extends PDFTextStripper
} else if (image instanceof PDPixelMap) {
metadata.set(Metadata.CONTENT_TYPE, "image/png");
}
+ metadata.set(TikaCoreProperties.EMBEDDED_RESOURCE_TYPE,
+
TikaCoreProperties.EmbeddedResourceType.inline.toString());
EmbeddedDocumentExtractor extractor =
getEmbeddedDocumentExtractor();
@@ -454,6 +480,8 @@ class PDF2XHTML extends PDFTextStripper
metadata.set(Metadata.RESOURCE_NAME_KEY, ent.getKey());
metadata.set(Metadata.CONTENT_TYPE, file.getSubtype());
metadata.set(Metadata.CONTENT_LENGTH,
Long.toString(file.getSize()));
+ metadata.set(TikaCoreProperties.EMBEDDED_RESOURCE_TYPE,
+
TikaCoreProperties.EmbeddedResourceType.attachment.toString());
if (extractor.shouldParseEmbedded(metadata)) {
TikaInputStream stream =
TikaInputStream.get(file.createInputStream());
Modified:
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java?rev=1597856&r1=1597855&r2=1597856&view=diff
==============================================================================
---
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java
(original)
+++
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java
Tue May 27 19:33:07 2014
@@ -63,6 +63,13 @@ import org.xml.sax.SAXException;
* the PDF contains any embedded documents (for example as part of a PDF
* package) then this parser will use the {@link EmbeddedDocumentExtractor}
* to handle them.
+ * <p>
+ * As of Tika 1.6, it is possible to extract inline images with
+ * the {@link EmbeddedDocumentExtractor} as if they were regular
+ * attachments. By default, this feature is turned off because of
+ * the potentially enormous number and size of inline images. To
+ * turn this feature on, see
+ * {@link PDFParserConfig#setExtractInlineImages(boolean)}.
*/
public class PDFParser extends AbstractParser {
Modified:
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParserConfig.java
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParserConfig.java?rev=1597856&r1=1597855&r2=1597856&view=diff
==============================================================================
---
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParserConfig.java
(original)
+++
tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParserConfig.java
Tue May 27 19:33:07 2014
@@ -1,11 +1,5 @@
package org.apache.tika.parser.pdf;
-import java.io.IOException;
-import java.io.InputStream;
-import java.io.Serializable;
-import java.util.Properties;
-
-import org.apache.pdfbox.util.PDFTextStripper;
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
@@ -23,6 +17,13 @@ import org.apache.pdfbox.util.PDFTextStr
* limitations under the License.
*/
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.Serializable;
+import java.util.Properties;
+
+import org.apache.pdfbox.util.PDFTextStripper;
+
/**
* Config for PDFParser.
*
@@ -64,6 +65,13 @@ public class PDFParserConfig implements
//True if acroform content should be extracted
private boolean extractAcroFormContent = true;
+
+ //True if inline PDXImage objects should be extracted
+ private boolean extractInlineImages = false;
+
+ //True if inline images (as identified by their object id within
+ //a pdf file) should only be extracted once.
+ private boolean extractUniqueInlineImagesOnly = true;
//The character width-based tolerance value used to estimate where spaces
in text should be added
private Float averageCharTolerance;
@@ -122,6 +130,12 @@ public class PDFParserConfig implements
setExtractAcroFormContent(
getProp(props.getProperty("extractAcroFormContent"),
getExtractAcroFormContent()));
+ setExtractInlineImages(
+ getProp(props.getProperty("extractInlineImages"),
+ getExtractInlineImages()));
+ setExtractUniqueInlineImagesOnly(
+ getProp(props.getProperty("extractUniqueInlineImagesOnly"),
+ getExtractUniqueInlineImagesOnly()));
}
/**
@@ -163,6 +177,56 @@ public class PDFParserConfig implements
return extractAcroFormContent;
}
+ /**
+ * If true, extract inline embedded OBXImages.
+ * <b>Beware:</b> some PDF documents of modest size (~4MB) can contain
+ * thousands of embedded images totaling > 2.5 GB. Also, at least as of
PDFBox 1.8.5,
+ * there can be surprisingly large memory consumption and/or out of memory
errors.
+ * Set to <code>true</code> with caution.
+ * <p>
+ * The default is <code>false</code>.
+ * <p>
+ * See also: {@see #setExtractUniqueInlineImagesOnly(boolean)};
+ *
+ * @param extractInlineImages
+ */
+ public void setExtractInlineImages(boolean extractInlineImages) {
+ this.extractInlineImages = extractInlineImages;
+ }
+
+ /** @see #setExtractInlineImages(boolean) */
+ public boolean getExtractInlineImages() {
+ return extractInlineImages;
+ }
+
+ /**
+ * Multiple pages within a PDF file might refer to the same underlying
image.
+ * If {@link #extractUniqueInlineImagesOnly} is set to <code>false</code>,
the
+ * parser will call the EmbeddedExtractor each time the image appears on a
page.
+ * This might be desired for some use cases. However, to avoid
duplication of
+ * extracted images, set this to <code>true</code>. The default is
<code>true</code>.
+ * <p>
+ * Note that uniqueness is determined only by the underlying PDF COSObject
id, not by
+ * file hash or similar equality metric.
+ * If the PDF actually contains multiple copies of the same image
+ * -- all with different object ids -- then all images will be extracted.
+ * <p>
+ * For this parameter to have any effect, {@link #extractInlineImages}
must be
+ * set to <code>true</code>.
+ *
+ * @param extractUniqueInlineImagesOnly
+ */
+ public void setExtractUniqueInlineImagesOnly(boolean
extractUniqueInlineImagesOnly) {
+ this.extractUniqueInlineImagesOnly = extractUniqueInlineImagesOnly;
+
+ }
+
+ /** @see #setExtractUniqueInlineImagesOnly(boolean) */
+ public boolean getExtractUniqueInlineImagesOnly() {
+ return extractUniqueInlineImagesOnly;
+ }
+
+
/** @see #setEnableAutoSpace. */
public boolean getEnableAutoSpace() {
return enableAutoSpace;
@@ -286,10 +350,19 @@ public class PDFParserConfig implements
public int hashCode() {
final int prime = 31;
int result = 1;
+ result = prime
+ * result
+ + ((averageCharTolerance == null) ? 0 : averageCharTolerance
+ .hashCode());
result = prime * result + (enableAutoSpace ? 1231 : 1237);
result = prime * result + (extractAcroFormContent ? 1231 : 1237);
result = prime * result + (extractAnnotationText ? 1231 : 1237);
+ result = prime * result + (extractInlineImages ? 1231 : 1237);
+ result = prime * result + (extractUniqueInlineImagesOnly ? 1231 :
1237);
result = prime * result + (sortByPosition ? 1231 : 1237);
+ result = prime
+ * result
+ + ((spacingTolerance == null) ? 0 :
spacingTolerance.hashCode());
result = prime * result
+ (suppressDuplicateOverlappingText ? 1231 : 1237);
result = prime * result + (useNonSequentialParser ? 1231 : 1237);
@@ -305,14 +378,28 @@ public class PDFParserConfig implements
if (getClass() != obj.getClass())
return false;
PDFParserConfig other = (PDFParserConfig) obj;
+ if (averageCharTolerance == null) {
+ if (other.averageCharTolerance != null)
+ return false;
+ } else if (!averageCharTolerance.equals(other.averageCharTolerance))
+ return false;
if (enableAutoSpace != other.enableAutoSpace)
return false;
if (extractAcroFormContent != other.extractAcroFormContent)
return false;
if (extractAnnotationText != other.extractAnnotationText)
return false;
+ if (extractInlineImages != other.extractInlineImages)
+ return false;
+ if (extractUniqueInlineImagesOnly !=
other.extractUniqueInlineImagesOnly)
+ return false;
if (sortByPosition != other.sortByPosition)
return false;
+ if (spacingTolerance == null) {
+ if (other.spacingTolerance != null)
+ return false;
+ } else if (!spacingTolerance.equals(other.spacingTolerance))
+ return false;
if (suppressDuplicateOverlappingText !=
other.suppressDuplicateOverlappingText)
return false;
if (useNonSequentialParser != other.useNonSequentialParser)
@@ -327,6 +414,11 @@ public class PDFParserConfig implements
+ suppressDuplicateOverlappingText + ", extractAnnotationText="
+ extractAnnotationText + ", sortByPosition=" + sortByPosition
+ ", useNonSequentialParser=" + useNonSequentialParser
- + ", extractAcroFormContent=" + extractAcroFormContent + "]";
+ + ", extractAcroFormContent=" + extractAcroFormContent
+ + ", extractInlineImages=" + extractInlineImages
+ + ", extractUniqueInlineImagesOnly="
+ + extractUniqueInlineImagesOnly + ", averageCharTolerance="
+ + averageCharTolerance + ", spacingTolerance="
+ + spacingTolerance + "]";
}
}
Modified:
tika/trunk/tika-parsers/src/main/resources/org/apache/tika/parser/pdf/PDFParser.properties
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/resources/org/apache/tika/parser/pdf/PDFParser.properties?rev=1597856&r1=1597855&r2=1597856&view=diff
==============================================================================
---
tika/trunk/tika-parsers/src/main/resources/org/apache/tika/parser/pdf/PDFParser.properties
(original)
+++
tika/trunk/tika-parsers/src/main/resources/org/apache/tika/parser/pdf/PDFParser.properties
Tue May 27 19:33:07 2014
@@ -19,3 +19,5 @@ sortByPosition false
suppressDuplicateOverlappingText false
useNonSequentialParser false
extractAcroFormContent true
+extractInlineImages false
+extractUniqueInlineImagesOnly true
Modified: tika/trunk/tika-parsers/src/test/java/org/apache/tika/TikaTest.java
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/test/java/org/apache/tika/TikaTest.java?rev=1597856&r1=1597855&r2=1597856&view=diff
==============================================================================
--- tika/trunk/tika-parsers/src/test/java/org/apache/tika/TikaTest.java
(original)
+++ tika/trunk/tika-parsers/src/test/java/org/apache/tika/TikaTest.java Tue May
27 19:33:07 2014
@@ -232,7 +232,11 @@ public abstract class TikaTest {
public List<Metadata> getAllMetadata() {
return metadatas;
- }
+ }
+
+ public void clear() {
+ metadatas.clear();
+ }
}
}
Modified:
tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java?rev=1597856&r1=1597855&r2=1597856&view=diff
==============================================================================
---
tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
(original)
+++
tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
Tue May 27 19:33:07 2014
@@ -34,6 +34,7 @@ import java.util.Set;
import org.apache.tika.TikaTest;
import org.apache.tika.extractor.ContainerExtractor;
+import org.apache.tika.extractor.DocumentSelector;
import org.apache.tika.extractor.ParserContainerExtractor;
import org.apache.tika.io.TikaInputStream;
import org.apache.tika.metadata.Metadata;
@@ -640,28 +641,35 @@ public class PDFParserTest extends TikaT
//"regressiveness" exists only in Unit10.doc not in the container pdf
document
assertTrue(xml.contains("regressiveness"));
- TrackingHandler tracker = new TrackingHandler();
+ RecursiveMetadataParser p = new RecursiveMetadataParser(new
AutoDetectParser(), false);
TikaInputStream tis = null;
- ContainerExtractor ex = new ParserContainerExtractor();
+ ParseContext context = new ParseContext();
+ PDFParserConfig config = new PDFParserConfig();
+ config.setExtractInlineImages(true);
+ config.setExtractUniqueInlineImagesOnly(false);
+ context.set(org.apache.tika.parser.pdf.PDFParserConfig.class, config);
+ context.set(org.apache.tika.parser.Parser.class, p);
+
try {
tis= TikaInputStream.get(
-
getResourceAsStream("/test-documents/testPDF_childAttachments.pdf"));
- ex.extract(tis, ex, tracker);
+
getResourceAsStream("/test-documents/testPDF_childAttachments.pdf"));
+ p.parse(tis, new BodyContentHandler(-1), new Metadata(), context);
} finally {
if (tis != null) {
tis.close();
}
}
- assertEquals(4, tracker.filenames.size());
- assertEquals(4, tracker.mediaTypes.size());
- assertNull(tracker.filenames.get(0));
- assertNull(tracker.filenames.get(1));
- assertEquals("Press Quality(1).joboptions", tracker.filenames.get(2));
- assertEquals("Unit10.doc", tracker.filenames.get(3));
- assertEquals(MediaType.image("jpeg"), tracker.mediaTypes.get(0));
- assertEquals(MediaType.image("tiff"), tracker.mediaTypes.get(1));
- assertEquals(TYPE_TEXT, tracker.mediaTypes.get(2));
- assertEquals(TYPE_DOC, tracker.mediaTypes.get(3));
+
+ List<Metadata> metadatas = p.getAllMetadata();
+ assertEquals(5, metadatas.size());
+ assertNull(metadatas.get(0).get(Metadata.RESOURCE_NAME_KEY));
+ assertNull(metadatas.get(1).get(Metadata.RESOURCE_NAME_KEY));
+ assertEquals("Press Quality(1).joboptions",
metadatas.get(2).get(Metadata.RESOURCE_NAME_KEY));
+ assertEquals("Unit10.doc",
metadatas.get(3).get(Metadata.RESOURCE_NAME_KEY));
+ assertEquals(MediaType.image("jpeg").toString(),
metadatas.get(0).get(Metadata.CONTENT_TYPE));
+ assertEquals(MediaType.image("tiff").toString(),
metadatas.get(1).get(Metadata.CONTENT_TYPE));
+ assertEquals("text/plain; charset=ISO-8859-1",
metadatas.get(2).get(Metadata.CONTENT_TYPE));
+ assertEquals(TYPE_DOC.toString(),
metadatas.get(3).get(Metadata.CONTENT_TYPE));
}
public void testVersions() throws Exception {
@@ -839,4 +847,144 @@ public class PDFParserTest extends TikaT
}
}
}
+
+ @Test
+ public void testInlineSelector() throws Exception {
+
+ PDFParserConfig config = new PDFParserConfig();
+ config.setExtractInlineImages(true);
+ config.setExtractUniqueInlineImagesOnly(false);
+
+ Parser defaultParser = new AutoDetectParser();
+
+ RecursiveMetadataParser p = new RecursiveMetadataParser(defaultParser,
false);
+ ParseContext context = new ParseContext();
+ context.set(org.apache.tika.parser.pdf.PDFParserConfig.class, config);
+ context.set(org.apache.tika.parser.Parser.class, p);
+ Metadata metadata = new Metadata();
+ ContentHandler handler = new BodyContentHandler(-1);
+ String path = "/test-documents/testPDF_childAttachments.pdf";
+ InputStream stream =
TikaInputStream.get(this.getClass().getResource(path));
+
+ p.parse(stream, handler, metadata, context);
+
+ List<Metadata> metadatas = p.getAllMetadata();
+ int inline = 0;
+ int attach = 0;
+ for (Metadata m : metadatas) {
+ String v = m.get(TikaCoreProperties.EMBEDDED_RESOURCE_TYPE);
+ if (v != null) {
+ if
(v.equals(TikaCoreProperties.EmbeddedResourceType.inline.toString())){
+ inline++;
+ } else if
(v.equals(TikaCoreProperties.EmbeddedResourceType.attachment.toString())){
+ attach++;
+ }
+ }
+ }
+ assertEquals(2, inline);
+ assertEquals(2, attach);
+
+ stream.close();
+ p.clear();
+
+ //now try turning off inline
+ stream = TikaInputStream.get(this.getClass().getResource(path));
+
+ context.set(org.apache.tika.extractor.DocumentSelector.class, new
AvoidInlineSelector());
+ inline = 0;
+ attach = 0;
+ handler = new BodyContentHandler(-1);
+ metadata = new Metadata();
+ p.parse(stream, handler, metadata, context);
+
+ metadatas = p.getAllMetadata();
+ for (Metadata m : metadatas) {
+ String v = m.get(TikaCoreProperties.EMBEDDED_RESOURCE_TYPE);
+ if (v != null) {
+ if
(v.equals(TikaCoreProperties.EmbeddedResourceType.inline.toString())){
+ inline++;
+ } else if
(v.equals(TikaCoreProperties.EmbeddedResourceType.attachment.toString())){
+ attach++;
+ }
+ }
+ }
+ assertEquals(0, inline);
+ assertEquals(2, attach);
+
+ }
+
+
+ @Test
+ public void testInlineConfig() throws Exception {
+
+ Parser defaultParser = new AutoDetectParser();
+ RecursiveMetadataParser p = new RecursiveMetadataParser(defaultParser,
false);
+ ParseContext context = new ParseContext();
+ context.set(org.apache.tika.parser.Parser.class, p);
+ Metadata metadata = new Metadata();
+ ContentHandler handler = new BodyContentHandler(-1);
+ String path = "/test-documents/testPDF_childAttachments.pdf";
+ InputStream stream =
TikaInputStream.get(this.getClass().getResource(path));
+
+ p.parse(stream, handler, metadata, context);
+
+ List<Metadata> metadatas = p.getAllMetadata();
+ int inline = 0;
+ int attach = 0;
+ for (Metadata m : metadatas) {
+ String v = m.get(TikaCoreProperties.EMBEDDED_RESOURCE_TYPE);
+ if (v != null) {
+ if
(v.equals(TikaCoreProperties.EmbeddedResourceType.inline.toString())){
+ inline++;
+ } else if
(v.equals(TikaCoreProperties.EmbeddedResourceType.attachment.toString())){
+ attach++;
+ }
+ }
+ }
+ assertEquals(0, inline);
+ assertEquals(2, attach);
+
+ stream.close();
+ p.clear();
+
+ //now try turning off inline
+ stream = TikaInputStream.get(this.getClass().getResource(path));
+ PDFParserConfig config = new PDFParserConfig();
+ config.setExtractInlineImages(true);
+ config.setExtractUniqueInlineImagesOnly(false);
+
+ context.set(org.apache.tika.parser.pdf.PDFParserConfig.class, config);
+ inline = 0;
+ attach = 0;
+ handler = new BodyContentHandler(-1);
+ metadata = new Metadata();
+ p.parse(stream, handler, metadata, context);
+
+ metadatas = p.getAllMetadata();
+ for (Metadata m : metadatas) {
+ String v = m.get(TikaCoreProperties.EMBEDDED_RESOURCE_TYPE);
+ if (v != null) {
+ if
(v.equals(TikaCoreProperties.EmbeddedResourceType.inline.toString())){
+ inline++;
+ } else if
(v.equals(TikaCoreProperties.EmbeddedResourceType.attachment.toString())){
+ attach++;
+ }
+ }
+ }
+ assertEquals(2, inline);
+ assertEquals(2, attach);
+ }
+
+
+ private class AvoidInlineSelector implements DocumentSelector {
+
+ @Override
+ public boolean select(Metadata metadata) {
+ String v = metadata.get(TikaCoreProperties.EMBEDDED_RESOURCE_TYPE);
+ if (v != null &&
v.equals(TikaCoreProperties.EmbeddedResourceType.inline.toString())){
+ return false;
+ }
+ return true;
+ }
+ }
}
