You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tika.apache.org by ta...@apache.org on 2015/05/29 16:36:24 UTC
svn commit: r1682489 [6/14] - in /tika/trunk:
tika-parsers/src/main/java/org/apache/tika/parser/html/
tika-parsers/src/main/java/org/apache/tika/parser/image/
tika-parsers/src/main/java/org/apache/tika/parser/image/xmp/
tika-parsers/src/main/java/org/a...
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=1682489&r1=1682488&r2=1682489&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 Fri May 29 14:36:21 2015
@@ -17,33 +17,32 @@ package org.apache.tika.parser.pdf;
* limitations under the License.
*/
-import org.apache.pdfbox.util.PDFTextStripper;
-
import java.io.IOException;
import java.io.InputStream;
import java.io.Serializable;
import java.util.Locale;
import java.util.Properties;
+import org.apache.pdfbox.util.PDFTextStripper;
+
/**
* Config for PDFParser.
- *
+ * <p/>
* This allows parameters to be set programmatically:
* <ol>
* <li>Calls to PDFParser, i.e. parser.getPDFParserConfig().setEnableAutoSpace() (as before)</li>
* <li>Constructor of PDFParser</li>
* <li>Passing to PDFParser through a ParseContext: context.set(PDFParserConfig.class, config);</li>
* </ol>
- *
+ * <p/>
* Parameters can also be set by modifying the PDFParserConfig.properties file,
* which lives in the expected places, in trunk:
* tika-parsers/src/main/resources/org/apache/tika/parser/pdf
- *
+ * <p/>
* Or, in tika-app-x.x.jar or tika-parsers-x.x.jar:
* org/apache/tika/parser/pdf
- *
*/
-public class PDFParserConfig implements Serializable{
+public class PDFParserConfig implements Serializable {
private static final long serialVersionUID = 6492570218190936986L;
@@ -63,7 +62,7 @@ public class PDFParserConfig implements
//True if we should use PDFBox's NonSequentialParser
private boolean useNonSequentialParser = false;
-
+
//True if acroform content should be extracted
private boolean extractAcroFormContent = true;
@@ -73,10 +72,10 @@ public class PDFParserConfig implements
//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;
-
+
//The space width-based tolerance value used to estimate where spaces in text should be added
private Float spacingTolerance;
@@ -90,7 +89,7 @@ public class PDFParserConfig implements
* Loads properties from InputStream and then tries to close InputStream.
* If there is an IOException, this silently swallows the exception
* and goes back to the default.
- *
+ *
* @param is
*/
public PDFParserConfig(InputStream is) {
@@ -109,7 +108,7 @@ public class PDFParserConfig implements
} catch (IOException e) {
} finally {
if (is != null) {
- try{
+ try {
is.close();
} catch (IOException e) {
//swallow
@@ -119,26 +118,26 @@ public class PDFParserConfig implements
setEnableAutoSpace(
getProp(props.getProperty("enableAutoSpace"), getEnableAutoSpace()));
setSuppressDuplicateOverlappingText(
- getProp(props.getProperty("suppressDuplicateOverlappingText"),
+ getProp(props.getProperty("suppressDuplicateOverlappingText"),
getSuppressDuplicateOverlappingText()));
setExtractAnnotationText(
- getProp(props.getProperty("extractAnnotationText"),
+ getProp(props.getProperty("extractAnnotationText"),
getExtractAnnotationText()));
setSortByPosition(
- getProp(props.getProperty("sortByPosition"),
+ getProp(props.getProperty("sortByPosition"),
getSortByPosition()));
setUseNonSequentialParser(
- getProp(props.getProperty("useNonSequentialParser"),
+ getProp(props.getProperty("useNonSequentialParser"),
getUseNonSequentialParser()));
setExtractAcroFormContent(
getProp(props.getProperty("extractAcroFormContent"),
- getExtractAcroFormContent()));
+ getExtractAcroFormContent()));
setExtractInlineImages(
getProp(props.getProperty("extractInlineImages"),
- getExtractInlineImages()));
+ getExtractInlineImages()));
setExtractUniqueInlineImagesOnly(
getProp(props.getProperty("extractUniqueInlineImagesOnly"),
- getExtractUniqueInlineImagesOnly()));
+ getExtractUniqueInlineImagesOnly()));
boolean checkExtractAccessPermission = getProp(props.getProperty("checkExtractAccessPermission"), false);
boolean allowExtractionForAccessibility = getProp(props.getProperty("allowExtractionForAccessibility"), true);
@@ -151,10 +150,10 @@ public class PDFParserConfig implements
accessChecker = new AccessChecker(allowExtractionForAccessibility);
}
}
-
+
/**
* Configures the given pdf2XHTML.
- *
+ *
* @param pdf2XHTML
*/
public void configure(PDF2XHTML pdf2XHTML) {
@@ -174,108 +173,118 @@ public class PDFParserConfig implements
pdf2XHTML.setSuppressDuplicateOverlappingText(getSuppressDuplicateOverlappingText());
}
-
+ /**
+ * @see #setExtractAcroFormContent(boolean)
+ */
+ public boolean getExtractAcroFormContent() {
+ return extractAcroFormContent;
+ }
+
/**
* If true (the default), extract content from AcroForms
* at the end of the document.
- *
+ *
* @param extractAcroFormContent
*/
public void setExtractAcroFormContent(boolean extractAcroFormContent) {
this.extractAcroFormContent = extractAcroFormContent;
-
+
}
- /** @see #setExtractAcroFormContent(boolean) */
- public boolean getExtractAcroFormContent() {
- return extractAcroFormContent;
+ /**
+ * @see #setExtractInlineImages(boolean)
+ */
+ public boolean getExtractInlineImages() {
+ return extractInlineImages;
}
/**
* 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,
+ * 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>
+ * <p/>
* The default is <code>false</code>.
- * <p>
+ * <p/>
* See also: {@see #setExtractUniqueInlineImagesOnly(boolean)};
- *
+ *
* @param extractInlineImages
*/
public void setExtractInlineImages(boolean extractInlineImages) {
- this.extractInlineImages = extractInlineImages;
+ this.extractInlineImages = extractInlineImages;
}
- /** @see #setExtractInlineImages(boolean) */
- public boolean getExtractInlineImages() {
- return extractInlineImages;
+ /**
+ * @see #setExtractUniqueInlineImagesOnly(boolean)
+ */
+ public boolean getExtractUniqueInlineImagesOnly() {
+ return extractUniqueInlineImagesOnly;
}
/**
* 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
+ * 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
+ * <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
+ * 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
+ * <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(boolean) */
+ /**
+ * @see #setEnableAutoSpace(boolean)
+ */
public boolean getEnableAutoSpace() {
return enableAutoSpace;
}
/**
- * If true (the default), the parser should estimate
- * where spaces should be inserted between words. For
- * many PDFs this is necessary as they do not include
- * explicit whitespace characters.
+ * If true (the default), the parser should estimate
+ * where spaces should be inserted between words. For
+ * many PDFs this is necessary as they do not include
+ * explicit whitespace characters.
*/
public void setEnableAutoSpace(boolean enableAutoSpace) {
this.enableAutoSpace = enableAutoSpace;
}
- /** @see #setSuppressDuplicateOverlappingText(boolean)*/
+ /**
+ * @see #setSuppressDuplicateOverlappingText(boolean)
+ */
public boolean getSuppressDuplicateOverlappingText() {
return suppressDuplicateOverlappingText;
}
/**
- * If true, the parser should try to remove duplicated
- * text over the same region. This is needed for some
- * PDFs that achieve bolding by re-writing the same
- * text in the same area. Note that this can
- * slow down extraction substantially (PDFBOX-956) and
- * sometimes remove characters that were not in fact
- * duplicated (PDFBOX-1155). By default this is disabled.
+ * If true, the parser should try to remove duplicated
+ * text over the same region. This is needed for some
+ * PDFs that achieve bolding by re-writing the same
+ * text in the same area. Note that this can
+ * slow down extraction substantially (PDFBOX-956) and
+ * sometimes remove characters that were not in fact
+ * duplicated (PDFBOX-1155). By default this is disabled.
*/
public void setSuppressDuplicateOverlappingText(
boolean suppressDuplicateOverlappingText) {
this.suppressDuplicateOverlappingText = suppressDuplicateOverlappingText;
}
- /** @see #setExtractAnnotationText(boolean)*/
+ /**
+ * @see #setExtractAnnotationText(boolean)
+ */
public boolean getExtractAnnotationText() {
return extractAnnotationText;
}
@@ -287,24 +296,29 @@ public class PDFParserConfig implements
public void setExtractAnnotationText(boolean extractAnnotationText) {
this.extractAnnotationText = extractAnnotationText;
}
- /** @see #setSortByPosition(boolean)*/
+
+ /**
+ * @see #setSortByPosition(boolean)
+ */
public boolean getSortByPosition() {
return sortByPosition;
}
/**
- * If true, sort text tokens by their x/y position
- * before extracting text. This may be necessary for
- * some PDFs (if the text tokens are not rendered "in
- * order"), while for other PDFs it can produce the
- * wrong result (for example if there are 2 columns,
- * the text will be interleaved). Default is false.
+ * If true, sort text tokens by their x/y position
+ * before extracting text. This may be necessary for
+ * some PDFs (if the text tokens are not rendered "in
+ * order"), while for other PDFs it can produce the
+ * wrong result (for example if there are 2 columns,
+ * the text will be interleaved). Default is false.
*/
public void setSortByPosition(boolean sortByPosition) {
this.sortByPosition = sortByPosition;
}
- /** @see #setUseNonSequentialParser(boolean)*/
+ /**
+ * @see #setUseNonSequentialParser(boolean)
+ */
public boolean getUseNonSequentialParser() {
return useNonSequentialParser;
}
@@ -312,18 +326,21 @@ public class PDFParserConfig implements
/**
* If true, uses PDFBox's non-sequential parser.
* The non-sequential parser should be much faster than the traditional
- * full doc parser. However, until PDFBOX-XXX is fixed,
+ * full doc parser. However, until PDFBOX-XXX is fixed,
* the non-sequential parser fails
* to extract some document metadata.
- * <p>
+ * <p/>
* Default is false (use the traditional parser)
+ *
* @param useNonSequentialParser
*/
public void setUseNonSequentialParser(boolean useNonSequentialParser) {
this.useNonSequentialParser = useNonSequentialParser;
}
- /** @see #setAverageCharTolerance(Float)*/
+ /**
+ * @see #setAverageCharTolerance(Float)
+ */
public Float getAverageCharTolerance() {
return averageCharTolerance;
}
@@ -335,7 +352,9 @@ public class PDFParserConfig implements
this.averageCharTolerance = averageCharTolerance;
}
- /** @see #setSpacingTolerance(Float)*/
+ /**
+ * @see #setSpacingTolerance(Float)
+ */
public Float getSpacingTolerance() {
return spacingTolerance;
}
@@ -347,16 +366,16 @@ public class PDFParserConfig implements
this.spacingTolerance = spacingTolerance;
}
- public void setAccessChecker(AccessChecker accessChecker) {
- this.accessChecker = accessChecker;
- }
-
public AccessChecker getAccessChecker() {
return accessChecker;
}
- private boolean getProp(String p, boolean defaultMissing){
- if (p == null){
+ public void setAccessChecker(AccessChecker accessChecker) {
+ this.accessChecker = accessChecker;
+ }
+
+ private boolean getProp(String p, boolean defaultMissing) {
+ if (p == null) {
return defaultMissing;
}
if (p.toLowerCase(Locale.ROOT).equals("true")) {
@@ -375,7 +394,7 @@ public class PDFParserConfig implements
result = prime
* result
+ ((averageCharTolerance == null) ? 0 : averageCharTolerance
- .hashCode());
+ .hashCode());
result = prime * result + (enableAutoSpace ? 1231 : 1237);
result = prime * result + (extractAcroFormContent ? 1231 : 1237);
result = prime * result + (extractAnnotationText ? 1231 : 1237);
Modified: tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/GroupState.java
URL: http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/GroupState.java?rev=1682489&r1=1682488&r2=1682489&view=diff
==============================================================================
--- tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/GroupState.java (original)
+++ tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/GroupState.java Fri May 29 14:36:21 2015
@@ -59,7 +59,7 @@ class GroupState {
list = other.list;
listLevel = other.listLevel;
fontCharset = other.fontCharset;
- depth = 1+other.depth;
+ depth = 1 + other.depth;
pictDepth = other.pictDepth > 0 ? other.pictDepth + 1 : 0;
//do not inherit object, sn, sv or sp
Modified: tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/ListDescriptor.java
URL: http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/ListDescriptor.java?rev=1682489&r1=1682488&r2=1682489&view=diff
==============================================================================
--- tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/ListDescriptor.java (original)
+++ tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/ListDescriptor.java Fri May 29 14:36:21 2015
@@ -29,8 +29,7 @@ public class ListDescriptor {
public boolean isStyle;
public int[] numberType = new int[9];
- public boolean isUnordered(int level)
- {
+ public boolean isUnordered(int level) {
return numberType[level] == NUMBER_TYPE_BULLET;
}
}
Modified: tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFEmbObjHandler.java
URL: http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFEmbObjHandler.java?rev=1682489&r1=1682488&r2=1682489&view=diff
==============================================================================
--- tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFEmbObjHandler.java (original)
+++ tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFEmbObjHandler.java Fri May 29 14:36:21 2015
@@ -14,143 +14,134 @@ package org.apache.tika.parser.rtf;
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
- */
-import java.io.ByteArrayOutputStream;
-import java.io.IOException;
-import java.io.InputStream;
-import java.util.concurrent.atomic.AtomicInteger;
-import org.apache.tika.config.TikaConfig;
-import org.apache.tika.detect.Detector;
-import org.apache.tika.exception.TikaException;
-import org.apache.tika.extractor.EmbeddedDocumentExtractor;
-import org.apache.tika.extractor.ParsingEmbeddedDocumentExtractor;
-import org.apache.tika.io.FilenameUtils;
-import org.apache.tika.io.TikaInputStream;
-import org.apache.tika.metadata.Metadata;
-import org.apache.tika.metadata.RTFMetadata;
-import org.apache.tika.mime.MediaType;
-import org.apache.tika.mime.MimeType;
-import org.apache.tika.mime.MimeTypeException;
-import org.apache.tika.mime.MimeTypes;
-import org.apache.tika.parser.ParseContext;
-import org.apache.tika.sax.EmbeddedContentHandler;
-import org.xml.sax.ContentHandler;
-import org.xml.sax.SAXException;
+ */
+
+import java.io.ByteArrayOutputStream;
+import java.io.IOException;
+import java.io.InputStream;
+import java.util.concurrent.atomic.AtomicInteger;
+
+import org.apache.tika.config.TikaConfig;
+import org.apache.tika.detect.Detector;
+import org.apache.tika.exception.TikaException;
+import org.apache.tika.extractor.EmbeddedDocumentExtractor;
+import org.apache.tika.extractor.ParsingEmbeddedDocumentExtractor;
+import org.apache.tika.io.FilenameUtils;
+import org.apache.tika.io.TikaInputStream;
+import org.apache.tika.metadata.Metadata;
+import org.apache.tika.metadata.RTFMetadata;
+import org.apache.tika.mime.MediaType;
+import org.apache.tika.mime.MimeType;
+import org.apache.tika.mime.MimeTypeException;
+import org.apache.tika.mime.MimeTypes;
+import org.apache.tika.parser.ParseContext;
+import org.apache.tika.sax.EmbeddedContentHandler;
+import org.xml.sax.ContentHandler;
+import org.xml.sax.SAXException;
/**
* This class buffers data from embedded objects and pictures.
- *
* <p/>
- *
+ * <p/>
+ * <p/>
* When the parser has finished an object or picture and called
* {@link #handleCompletedObject()}, this will write the object
* to the {@link #handler}.
- *
* <p/>
- *
+ * <p/>
+ * <p/>
* This (in combination with TextExtractor) expects basically a flat parse. It will pull out
* all pict whether they are tied to objdata or are intended
* to be standalone.
- *
+ * <p/>
* <p/>
* This tries to pull metadata around a pict that is encoded
* with {sp {sn} {sv}} types of data. This information
* sometimes contains the name and even full file path of the original file.
- *
- */ class RTFEmbObjHandler {
-
+ */
+class RTFEmbObjHandler {
+
private static final String EMPTY_STRING = "";
-
- private enum EMB_STATE {
- PICT, //recording pict data
- OBJDATA, //recording objdata
- NADA
- };
-
+ private final ContentHandler handler;
+
+
+ private final ParseContext context;
+ private final ByteArrayOutputStream os;
//high hex cached for writing hexpair chars (data)
private int hi = -1;
-
private int thumbCount = 0;
//don't need atomic, do need mutable
private AtomicInteger unknownFilenameCount = new AtomicInteger();
-
private boolean inObject = false;
-
private String sv = EMPTY_STRING;
private String sn = EMPTY_STRING;
-
private StringBuilder sb = new StringBuilder();
-
- private final ContentHandler handler;
private Metadata metadata;
- private final ParseContext context;
-
- private final ByteArrayOutputStream os;
private EMB_STATE state = EMB_STATE.NADA;
-
protected RTFEmbObjHandler(ContentHandler handler, Metadata metadata, ParseContext context) {
this.handler = handler;
this.context = context;
os = new ByteArrayOutputStream();
}
+
protected void startPict() {
state = EMB_STATE.PICT;
metadata = new Metadata();
}
-
+
protected void startObjData() {
state = EMB_STATE.OBJDATA;
metadata = new Metadata();
}
-
+
protected void startSN() {
sb.setLength(0);
sb.append(RTFMetadata.RTF_PICT_META_PREFIX);
}
-
+
protected void endSN() {
sn = sb.toString();
}
-
+
protected void startSV() {
sb.setLength(0);
}
-
+
protected void endSV() {
sv = sb.toString();
}
-
+
//end metadata pair
protected void endSP() {
metadata.add(sn, sv);
}
-
- protected void setInObject(boolean v) {
- inObject = v;
- }
-
+
protected boolean getInObject() {
return inObject;
}
-
+
+ protected void setInObject(boolean v) {
+ inObject = v;
+ }
+
protected void writeMetadataChar(char c) {
sb.append(c);
}
-
+
protected void writeHexChar(int b) throws IOException, TikaException {
//if not hexchar, ignore
//white space is common
if (TextExtractor.isHexChar(b)) {
if (hi == -1) {
- hi = 16*TextExtractor.hexValue(b);
+ hi = 16 * TextExtractor.hexValue(b);
} else {
- long sum = hi+TextExtractor.hexValue(b);
+ long sum = hi + TextExtractor.hexValue(b);
if (sum > Integer.MAX_VALUE || sum < 0) {
throw new IOException("hex char to byte overflow");
}
-
- os.write((int)sum);
-
+
+ os.write((int) sum);
+
hi = -1;
}
return;
@@ -159,80 +150,80 @@ import org.xml.sax.SAXException;
throw new TikaException("hit end of stream before finishing byte pair");
}
}
-
-
+
protected void writeBytes(InputStream is, int len) throws IOException, TikaException {
if (len < 0 || len > RTFParser.getMaxBytesForEmbeddedObject()) {
throw new IOException("length of bytes to read out of bounds: " + len);
}
-
+
byte[] bytes = new byte[len];
int bytesRead = is.read(bytes);
if (bytesRead < len) {
throw new TikaException("unexpected end of file: need " + len +
- " bytes of binary data, found " + (len-bytesRead));
+ " bytes of binary data, found " + (len - bytesRead));
}
os.write(bytes);
}
-
+
/**
* Call this when the objdata/pict has completed
+ *
* @throws IOException
* @throws SAXException
* @throws TikaException
*/
protected void handleCompletedObject() throws IOException, SAXException, TikaException {
- EmbeddedDocumentExtractor embeddedExtractor = context.get(EmbeddedDocumentExtractor.class);
-
- if (embeddedExtractor == null) {
- embeddedExtractor = new ParsingEmbeddedDocumentExtractor(context);
- }
-
- byte[] bytes = os.toByteArray();
- if (state == EMB_STATE.OBJDATA) {
- RTFObjDataParser objParser = new RTFObjDataParser();
- try{
- byte[] objBytes = objParser.parse(bytes, metadata, unknownFilenameCount);
- extractObj(objBytes, handler, embeddedExtractor, metadata);
- } catch (IOException e) {
- //swallow. If anything goes wrong, ignore.
- }
- } else if (state == EMB_STATE.PICT) {
- String filePath = metadata.get(RTFMetadata.RTF_PICT_META_PREFIX+"wzDescription");
- if (filePath != null && filePath.length() > 0){
- metadata.set(Metadata.EMBEDDED_RELATIONSHIP_ID, filePath);
- metadata.set(Metadata.RESOURCE_NAME_KEY, FilenameUtils.getName(filePath));
- }
- metadata.set(RTFMetadata.THUMBNAIL, Boolean.toString(inObject));
- extractObj(bytes, handler, embeddedExtractor, metadata);
-
- } else if (state == EMB_STATE.NADA) {
- //swallow...no start for pict or embed?!
- }
- reset();
+ EmbeddedDocumentExtractor embeddedExtractor = context.get(EmbeddedDocumentExtractor.class);
+
+ if (embeddedExtractor == null) {
+ embeddedExtractor = new ParsingEmbeddedDocumentExtractor(context);
+ }
+
+ byte[] bytes = os.toByteArray();
+ if (state == EMB_STATE.OBJDATA) {
+ RTFObjDataParser objParser = new RTFObjDataParser();
+ try {
+ byte[] objBytes = objParser.parse(bytes, metadata, unknownFilenameCount);
+ extractObj(objBytes, handler, embeddedExtractor, metadata);
+ } catch (IOException e) {
+ //swallow. If anything goes wrong, ignore.
+ }
+ } else if (state == EMB_STATE.PICT) {
+ String filePath = metadata.get(RTFMetadata.RTF_PICT_META_PREFIX + "wzDescription");
+ if (filePath != null && filePath.length() > 0) {
+ metadata.set(Metadata.EMBEDDED_RELATIONSHIP_ID, filePath);
+ metadata.set(Metadata.RESOURCE_NAME_KEY, FilenameUtils.getName(filePath));
+ }
+ metadata.set(RTFMetadata.THUMBNAIL, Boolean.toString(inObject));
+ extractObj(bytes, handler, embeddedExtractor, metadata);
+
+ } else if (state == EMB_STATE.NADA) {
+ //swallow...no start for pict or embed?!
+ }
+ reset();
}
-
+
private void extractObj(byte[] bytes, ContentHandler handler,
- EmbeddedDocumentExtractor embeddedExtractor, Metadata metadata)
- throws SAXException, IOException, TikaException {
-
+ EmbeddedDocumentExtractor embeddedExtractor, Metadata metadata)
+ throws SAXException, IOException, TikaException {
+
if (bytes == null) {
return;
}
-
+
metadata.set(Metadata.CONTENT_LENGTH, Integer.toString(bytes.length));
-
+
if (embeddedExtractor.shouldParseEmbedded(metadata)) {
TikaInputStream stream = TikaInputStream.get(bytes);
if (metadata.get(Metadata.RESOURCE_NAME_KEY) == null) {
String extension = getExtension(stream, metadata);
stream.reset();
if (inObject && state == EMB_STATE.PICT) {
- metadata.set(Metadata.RESOURCE_NAME_KEY, "thumbnail_"+thumbCount++ + extension);
+ metadata.set(Metadata.RESOURCE_NAME_KEY, "thumbnail_" + thumbCount++ + extension);
metadata.set(RTFMetadata.THUMBNAIL, "true");
} else {
- metadata.set(Metadata.RESOURCE_NAME_KEY, "file_"+unknownFilenameCount.getAndIncrement() +
-extension);
+ metadata.set(Metadata.RESOURCE_NAME_KEY, "file_" + unknownFilenameCount.getAndIncrement() +
+ extension);
}
}
try {
@@ -245,7 +236,7 @@ extension);
}
}
}
-
+
private String getExtension(TikaInputStream is, Metadata metadata) {
String cType = metadata.get(Metadata.CONTENT_TYPE);
TikaConfig config = getConfig();
@@ -260,12 +251,12 @@ extension);
} catch (IOException e) {
//swallow
} catch (MimeTypeException e) {
-
+
}
}
return ".bin";
}
-
+
private TikaConfig getConfig() {
TikaConfig config = context.get(TikaConfig.class);
if (config == null) {
@@ -273,7 +264,7 @@ extension);
}
return config;
}
-
+
/**
* reset state after each object.
* Do not reset unknown file number.
@@ -287,4 +278,10 @@ extension);
sn = EMPTY_STRING;
sb.setLength(0);
}
+
+ private enum EMB_STATE {
+ PICT, //recording pict data
+ OBJDATA, //recording objdata
+ NADA
+ }
}
Modified: tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFObjDataParser.java
URL: http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFObjDataParser.java?rev=1682489&r1=1682488&r2=1682489&view=diff
==============================================================================
--- tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFObjDataParser.java (original)
+++ tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFObjDataParser.java Fri May 29 14:36:21 2015
@@ -43,37 +43,35 @@ import org.apache.tika.parser.microsoft.
/**
* Many thanks to Simon Mourier for:
- * http://stackoverflow.com/questions/14779647/extract-embedded-image-object-in-rtf
+ * http://stackoverflow.com/questions/14779647/extract-embedded-image-object-in-rtf
* and for granting permission to use his code in Tika.
- *
*/
class RTFObjDataParser {
private final static int[] INT_LE_POWS = new int[]{
- 1, 256, 65536, 16777216
+ 1, 256, 65536, 16777216
};
private final static String WIN_ASCII = "WINDOWS-1252";
/**
* Parses the embedded object/pict string
- *
+ *
* @param bytes actual bytes (already converted from the
* hex pair string stored in the embedded object data into actual bytes or read
* as raw binary bytes)
* @return a SimpleRTFEmbObj or null
* @throws IOException if there are any surprise surprises during parsing
*/
-
+
/**
- *
* @param bytes
- * @param metadata incoming metadata
- * @param unknownFilenameCount
+ * @param metadata incoming metadata
+ * @param unknownFilenameCount
* @return byte[] for contents of obj data
* @throws IOException
*/
- protected byte[] parse(byte[] bytes, Metadata metadata, AtomicInteger unknownFilenameCount)
+ protected byte[] parse(byte[] bytes, Metadata metadata, AtomicInteger unknownFilenameCount)
throws IOException {
ByteArrayInputStream is = new ByteArrayInputStream(bytes);
long version = readUInt(is);
@@ -84,9 +82,9 @@ class RTFObjDataParser {
if (formatId != 2L) {
return null;
}
- String className = readLengthPrefixedAnsiString(is).trim();
+ String className = readLengthPrefixedAnsiString(is).trim();
String topicName = readLengthPrefixedAnsiString(is).trim();
- String itemName = readLengthPrefixedAnsiString(is).trim();
+ String itemName = readLengthPrefixedAnsiString(is).trim();
if (className != null && className.length() > 0) {
metadata.add(RTFMetadata.EMB_CLASS, className);
@@ -103,19 +101,19 @@ class RTFObjDataParser {
//readBytes tests for reading too many bytes
byte[] embObjBytes = readBytes(is, dataSz);
- if (className.toLowerCase(Locale.ROOT).equals("package")){
+ if (className.toLowerCase(Locale.ROOT).equals("package")) {
return handlePackage(embObjBytes, metadata);
} else if (className.toLowerCase(Locale.ROOT).equals("pbrush")) {
//simple bitmap bytes
return embObjBytes;
} else {
ByteArrayInputStream embIs = new ByteArrayInputStream(embObjBytes);
- if (NPOIFSFileSystem.hasPOIFSHeader(embIs)){
- try{
+ if (NPOIFSFileSystem.hasPOIFSHeader(embIs)) {
+ try {
return handleEmbeddedPOIFS(embIs, metadata, unknownFilenameCount);
} catch (IOException e) {
//swallow
- }
+ }
}
}
return embObjBytes;
@@ -124,8 +122,8 @@ class RTFObjDataParser {
//will throw IOException if not actually POIFS
//can return null byte[]
- private byte[] handleEmbeddedPOIFS(InputStream is, Metadata metadata,
- AtomicInteger unknownFilenameCount)
+ private byte[] handleEmbeddedPOIFS(InputStream is, Metadata metadata,
+ AtomicInteger unknownFilenameCount)
throws IOException {
NPOIFSFileSystem fs = null;
@@ -140,7 +138,7 @@ class RTFObjDataParser {
return ret;
}
- if (root.hasEntry("Package")){
+ if (root.hasEntry("Package")) {
Entry ooxml = root.getEntry("Package");
TikaInputStream stream = TikaInputStream.get(new DocumentInputStream((DocumentEntry) ooxml));
@@ -163,9 +161,9 @@ class RTFObjDataParser {
DocumentEntry contentsEntry;
try {
- contentsEntry = (DocumentEntry)root.getEntry("CONTENTS");
+ contentsEntry = (DocumentEntry) root.getEntry("CONTENTS");
} catch (FileNotFoundException ioe) {
- contentsEntry = (DocumentEntry)root.getEntry("Contents");
+ contentsEntry = (DocumentEntry) root.getEntry("Contents");
}
DocumentInputStream inp = null;
@@ -184,7 +182,7 @@ class RTFObjDataParser {
is.reset();
IOUtils.copy(is, out);
ret = out.toByteArray();
- metadata.set(Metadata.RESOURCE_NAME_KEY, "file_"+unknownFilenameCount.getAndIncrement() + "."+type.getExtension());
+ metadata.set(Metadata.RESOURCE_NAME_KEY, "file_" + unknownFilenameCount.getAndIncrement() + "." + type.getExtension());
metadata.set(Metadata.CONTENT_TYPE, type.getType().toString());
}
}
@@ -197,12 +195,11 @@ class RTFObjDataParser {
}
-
/**
- * can return null if there is a linked object
+ * can return null if there is a linked object
* instead of an embedded file
*/
- private byte[] handlePackage(byte[] pkgBytes, Metadata metadata) throws IOException {
+ private byte[] handlePackage(byte[] pkgBytes, Metadata metadata) throws IOException {
//now parse the package header
ByteArrayInputStream is = new ByteArrayInputStream(pkgBytes);
readUShort(is);
@@ -231,16 +228,16 @@ class RTFObjDataParser {
try {
long unicodeLen = readUInt(is);
- for (int i = 0; i < unicodeLen; i++){
+ for (int i = 0; i < unicodeLen; i++) {
int lo = is.read();
int hi = is.read();
- int sum = lo+256*hi;
- if (hi == -1 || lo == -1){
+ int sum = lo + 256 * hi;
+ if (hi == -1 || lo == -1) {
//stream ran out; empty SB and stop
unicodeFilePath.setLength(0);
break;
}
- unicodeFilePath.append((char)sum);
+ unicodeFilePath.append((char) sum);
}
} catch (IOException e) {
//swallow; the unicode file path is optional and might not happen
@@ -248,7 +245,7 @@ class RTFObjDataParser {
}
String fileNameToUse = "";
String pathToUse = "";
- if (unicodeFilePath.length() > 0){
+ if (unicodeFilePath.length() > 0) {
String p = unicodeFilePath.toString();
fileNameToUse = p;
pathToUse = p;
@@ -265,21 +262,21 @@ class RTFObjDataParser {
private int readUShort(InputStream is) throws IOException {
int lo = is.read();
- int hi = is.read()*256;
+ int hi = is.read() * 256;
if (lo == -1 || hi == -1) {
throw new IOException("Hit end of stream before reading little endian unsigned short.");
}
- return hi+lo;
+ return hi + lo;
}
private long readUInt(InputStream is) throws IOException {
long sum = 0;
- for (int i = 0; i < 4; i++){
+ for (int i = 0; i < 4; i++) {
int v = is.read();
if (v == -1) {
throw new IOException("Hit end of stream before finishing little endian unsigned int.");
}
- sum += v*(long)INT_LE_POWS[i];
+ sum += v * (long) INT_LE_POWS[i];
}
return sum;
}
@@ -288,7 +285,7 @@ class RTFObjDataParser {
StringBuilder sb = new StringBuilder();
int c = is.read();
while (c > 0) {
- sb.append((char)c);
+ sb.append((char) c);
c = is.read();
}
if (c == -1) {
@@ -319,13 +316,13 @@ class RTFObjDataParser {
return bytes;
}
-
+
private byte[] initByteArray(long len) throws IOException {
if (len < 0 || len > RTFParser.getMaxBytesForEmbeddedObject()) {
throw new IOException("Requested length for reading bytes is out of bounds: " + len);
}
- return new byte[(int)len];
-
+ return new byte[(int) len];
+
}
}
Modified: tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFParser.java
URL: http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFParser.java?rev=1682489&r1=1682488&r2=1682489&view=diff
==============================================================================
--- tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFParser.java (original)
+++ tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/RTFParser.java Fri May 29 14:36:21 2015
@@ -43,17 +43,21 @@ public class RTFParser extends AbstractP
private static final Set<MediaType> SUPPORTED_TYPES =
Collections.singleton(MediaType.application("rtf"));
-
- public Set<MediaType> getSupportedTypes(ParseContext context) {
- return SUPPORTED_TYPES;
- }
-
/**
* maximum number of bytes per embedded object/pict (default: 20MB)
*/
private static int EMB_OBJ_MAX_BYTES = 20 * 1024 * 1024; //20MB
/**
+ * See {@link #setMaxBytesForEmbeddedObject(int)}.
+ *
+ * @return maximum number of bytes allowed for an embedded object.
+ */
+ public static int getMaxBytesForEmbeddedObject() {
+ return EMB_OBJ_MAX_BYTES;
+ }
+
+ /**
* Bytes for embedded objects are currently cached in memory.
* If something goes wrong during the parsing of an embedded object,
* it is possible that a read length may be crazily too long
@@ -66,19 +70,14 @@ public class RTFParser extends AbstractP
EMB_OBJ_MAX_BYTES = max;
}
- /**
- * See {@link #setMaxBytesForEmbeddedObject(int)}.
- *
- * @return maximum number of bytes allowed for an embedded object.
- */
- public static int getMaxBytesForEmbeddedObject() {
- return EMB_OBJ_MAX_BYTES;
+ public Set<MediaType> getSupportedTypes(ParseContext context) {
+ return SUPPORTED_TYPES;
}
public void parse(
InputStream stream, ContentHandler handler,
Metadata metadata, ParseContext context)
- throws IOException, SAXException, TikaException {
+ throws IOException, SAXException, TikaException {
TaggedInputStream tagged = new TaggedInputStream(stream);
try {
XHTMLContentHandler xhtmlHandler = new XHTMLContentHandler(handler, metadata);
Modified: tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/TextExtractor.java
URL: http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/TextExtractor.java?rev=1682489&r1=1682488&r2=1682489&view=diff
==============================================================================
--- tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/TextExtractor.java (original)
+++ tika/trunk/tika-parsers/src/main/java/org/apache/tika/parser/rtf/TextExtractor.java Fri May 29 14:36:21 2015
@@ -54,15 +54,6 @@ import org.xml.sax.SAXException;
final class TextExtractor {
private static final Charset ASCII = Charset.forName("US-ASCII");
-
- private static Charset getCharset(String name) {
- try {
- return CharsetUtils.forName(name);
- } catch (Exception e) {
- return ASCII;
- }
- }
-
private static final Charset WINDOWS_1252 = getCharset("WINDOWS-1252");
private static final Charset MAC_ROMAN = getCharset("MacRoman");
private static final Charset SHIFT_JIS = getCharset("Shift_JIS");
@@ -131,100 +122,17 @@ final class TextExtractor {
private static final Charset BIG5 = getCharset("Big5");
private static final Charset GB2312 = getCharset("GB2312");
private static final Charset MS949 = getCharset("ms949");
-
- private int written = 0;
- // Hold pending bytes (encoded in the current charset)
- // for text output:
- private byte[] pendingBytes = new byte[16];
- private int pendingByteCount;
- private ByteBuffer pendingByteBuffer = ByteBuffer.wrap(pendingBytes);
-
- // Holds pending chars for text output
- private char[] pendingChars = new char[10];
- private int pendingCharCount;
-
- // Holds chars for a still-being-tokenized control word
- private byte[] pendingControl = new byte[10];
- private int pendingControlCount;
-
- // Used when we decode bytes -> chars using CharsetDecoder:
- private final char[] outputArray = new char[128];
- private final CharBuffer outputBuffer = CharBuffer.wrap(outputArray);
-
- // Reused when possible:
- private CharsetDecoder decoder;
- private Charset lastCharset;
-
- private Charset globalCharset = WINDOWS_1252;
- private int globalDefaultFont = -1;
- private int curFontID = -1;
-
- // Holds the font table from this RTF doc, mapping
- // the font number (from \fN control word) to the
- // corresponding charset:
- private final Map<Integer, Charset> fontToCharset =
- new HashMap<Integer, Charset>();
-
- // Group stack: when we open a new group, we push
- // the previous group state onto the stack; when we
- // close the group, we restore it
- private final LinkedList<GroupState> groupStates = new LinkedList<GroupState>();
-
- // Current group state; in theory this initial
- // GroupState is unused because the RTF doc should
- // immediately open the top group (start with {):
- private GroupState groupState = new GroupState();
-
- private boolean inHeader = true;
- private int fontTableState;
- private int fontTableDepth;
-
- // Non null if we are processing metadata (title,
- // keywords, etc.) inside the info group:
- private Property nextMetaData;
- private boolean inParagraph;
-
- // Non-zero if we are processing inside a field destination:
- private int fieldState;
-
- // Non-zero list index
- private int pendingListEnd;
- private Map<Integer, ListDescriptor> listTable = new HashMap<Integer, ListDescriptor>();
- private Map<Integer, ListDescriptor> listOverrideTable = new HashMap<Integer, ListDescriptor>();
- private Map<Integer, ListDescriptor> currentListTable;
- private ListDescriptor currentList;
- private int listTableLevel = -1;
- private boolean ignoreLists;
-
- // Non-null if we've seen the url for a HYPERLINK but not yet
- // its text:
- private String pendingURL;
-
- private final StringBuilder pendingBuffer = new StringBuilder();
-
- // Used to process the sub-groups inside the upr
- // group:
- private int uprState = -1;
-
- private final XHTMLContentHandler out;
- private final Metadata metadata;
- private final RTFEmbObjHandler embObjHandler;
-
- // Used when extracting CREATION date:
- private int year, month, day, hour, minute;
-
- // How many next ansi chars we should skip; this
- // is 0 except when we are still in the "ansi
- // shadow" after seeing a unicode escape, at which
- // point it's set to the last ucN skip we had seen:
- int ansiSkip = 0;
-
// The RTF doc has a "font table" that assigns ords
// (f0, f1, f2, etc.) to fonts and charsets, using the
// \fcharsetN control word. This mapping maps from the
// N to corresponding Java charset:
private static final Map<Integer, Charset> FCHARSET_MAP =
new HashMap<Integer, Charset>();
+ // The RTF may specify the \ansicpgN charset in the
+ // header; this maps the N to the corresponding Java
+ // character set:
+ private static final Map<Integer, Charset> ANSICPG_MAP =
+ new HashMap<Integer, Charset>();
static {
FCHARSET_MAP.put(0, WINDOWS_1252); // ANSI
@@ -267,15 +175,10 @@ final class TextExtractor {
FCHARSET_MAP.put(255, CP850); // OEM
}
- // The RTF may specify the \ansicpgN charset in the
- // header; this maps the N to the corresponding Java
- // character set:
- private static final Map<Integer, Charset> ANSICPG_MAP =
- new HashMap<Integer, Charset>();
static {
ANSICPG_MAP.put(437, CP4372); // US IBM
ANSICPG_MAP.put(708, ISO_8859_6); // Arabic (ASMO 708)
-
+
ANSICPG_MAP.put(709, WINDOWS_709); // Arabic (ASMO 449+, BCON V4)
ANSICPG_MAP.put(710, WINDOWS_710); // Arabic (transparent Arabic)
ANSICPG_MAP.put(710, WINDOWS_711); // Arabic (Nafitha Enhanced)
@@ -331,30 +234,99 @@ final class TextExtractor {
ANSICPG_MAP.put(57011, WINDOWS_57011); // Punjabi
}
+ // Used when we decode bytes -> chars using CharsetDecoder:
+ private final char[] outputArray = new char[128];
+ private final CharBuffer outputBuffer = CharBuffer.wrap(outputArray);
+ // Holds the font table from this RTF doc, mapping
+ // the font number (from \fN control word) to the
+ // corresponding charset:
+ private final Map<Integer, Charset> fontToCharset =
+ new HashMap<Integer, Charset>();
+ // Group stack: when we open a new group, we push
+ // the previous group state onto the stack; when we
+ // close the group, we restore it
+ private final LinkedList<GroupState> groupStates = new LinkedList<GroupState>();
+ private final StringBuilder pendingBuffer = new StringBuilder();
+ private final XHTMLContentHandler out;
+ private final Metadata metadata;
+ private final RTFEmbObjHandler embObjHandler;
+ // How many next ansi chars we should skip; this
+ // is 0 except when we are still in the "ansi
+ // shadow" after seeing a unicode escape, at which
+ // point it's set to the last ucN skip we had seen:
+ int ansiSkip = 0;
+ private int written = 0;
+ // Hold pending bytes (encoded in the current charset)
+ // for text output:
+ private byte[] pendingBytes = new byte[16];
+ private int pendingByteCount;
+ private ByteBuffer pendingByteBuffer = ByteBuffer.wrap(pendingBytes);
+ // Holds pending chars for text output
+ private char[] pendingChars = new char[10];
+ private int pendingCharCount;
+ // Holds chars for a still-being-tokenized control word
+ private byte[] pendingControl = new byte[10];
+ private int pendingControlCount;
+ // Reused when possible:
+ private CharsetDecoder decoder;
+ private Charset lastCharset;
+ private Charset globalCharset = WINDOWS_1252;
+ private int globalDefaultFont = -1;
+ private int curFontID = -1;
+ // Current group state; in theory this initial
+ // GroupState is unused because the RTF doc should
+ // immediately open the top group (start with {):
+ private GroupState groupState = new GroupState();
+ private boolean inHeader = true;
+ private int fontTableState;
+ private int fontTableDepth;
+ // Non null if we are processing metadata (title,
+ // keywords, etc.) inside the info group:
+ private Property nextMetaData;
+ private boolean inParagraph;
+ // Non-zero if we are processing inside a field destination:
+ private int fieldState;
+ // Non-zero list index
+ private int pendingListEnd;
+ private Map<Integer, ListDescriptor> listTable = new HashMap<Integer, ListDescriptor>();
+ private Map<Integer, ListDescriptor> listOverrideTable = new HashMap<Integer, ListDescriptor>();
+ private Map<Integer, ListDescriptor> currentListTable;
+ private ListDescriptor currentList;
+ private int listTableLevel = -1;
+ private boolean ignoreLists;
+ // Non-null if we've seen the url for a HYPERLINK but not yet
+ // its text:
+ private String pendingURL;
+ // Used to process the sub-groups inside the upr
+ // group:
+ private int uprState = -1;
+ // Used when extracting CREATION date:
+ private int year, month, day, hour, minute;
+
public TextExtractor(XHTMLContentHandler out, Metadata metadata,
- RTFEmbObjHandler embObjHandler) {
+ RTFEmbObjHandler embObjHandler) {
this.metadata = metadata;
this.out = out;
this.embObjHandler = embObjHandler;
}
- public boolean isIgnoringLists() {
- return ignoreLists;
- }
-
- public void setIgnoreLists(boolean ignore) {
- this.ignoreLists = ignore;
+ private static Charset getCharset(String name) {
+ try {
+ return CharsetUtils.forName(name);
+ } catch (Exception e) {
+ return ASCII;
+ }
}
protected static boolean isHexChar(int ch) {
return (ch >= '0' && ch <= '9') ||
- (ch >= 'a' && ch <= 'f') ||
- (ch >= 'A' && ch <= 'F');
+ (ch >= 'a' && ch <= 'f') ||
+ (ch >= 'A' && ch <= 'F');
}
private static boolean isAlpha(int ch) {
return (ch >= 'a' && ch <= 'z') ||
- (ch >= 'A' && ch <= 'Z');
+ (ch >= 'A' && ch <= 'Z');
}
private static boolean isDigit(int ch) {
@@ -372,6 +344,14 @@ final class TextExtractor {
}
}
+ public boolean isIgnoringLists() {
+ return ignoreLists;
+ }
+
+ public void setIgnoreLists(boolean ignore) {
+ this.ignoreLists = ignore;
+ }
+
// Push pending bytes or pending chars:
private void pushText() throws IOException, SAXException, TikaException {
if (pendingByteCount != 0) {
@@ -391,27 +371,27 @@ final class TextExtractor {
pushChars();
}
if (groupState.pictDepth > 0) {
- embObjHandler.writeMetadataChar((char)b);
+ embObjHandler.writeMetadataChar((char) b);
} else {
// Save the byte in pending buffer:
if (pendingByteCount == pendingBytes.length) {
// Gradual but exponential growth:
- final byte[] newArray = new byte[(int) (pendingBytes.length*1.25)];
+ final byte[] newArray = new byte[(int) (pendingBytes.length * 1.25)];
System.arraycopy(pendingBytes, 0, newArray, 0, pendingBytes.length);
pendingBytes = newArray;
pendingByteBuffer = ByteBuffer.wrap(pendingBytes);
}
pendingBytes[pendingByteCount++] = (byte) b;
- }
+ }
}
- // Buffers a byte as part of a control word:
+ // Buffers a byte as part of a control word:
private void addControl(int b) {
assert isAlpha(b);
// Save the byte in pending buffer:
if (pendingControlCount == pendingControl.length) {
// Gradual but exponential growth:
- final byte[] newArray = new byte[(int) (pendingControl.length*1.25)];
+ final byte[] newArray = new byte[(int) (pendingControl.length * 1.25)];
System.arraycopy(pendingControl, 0, newArray, 0, pendingControl.length);
pendingControl = newArray;
}
@@ -431,7 +411,7 @@ final class TextExtractor {
} else {
if (pendingCharCount == pendingChars.length) {
// Gradual but exponential growth:
- final char[] newArray = new char[(int) (pendingChars.length*1.25)];
+ final char[] newArray = new char[(int) (pendingChars.length * 1.25)];
System.arraycopy(pendingChars, 0, newArray, 0, pendingChars.length);
pendingChars = newArray;
}
@@ -458,7 +438,7 @@ final class TextExtractor {
// };
extract(new PushbackInputStream(in, 2));
}
-
+
private void extract(PushbackInputStream in) throws IOException, SAXException, TikaException {
out.startDocument();
@@ -471,7 +451,7 @@ final class TextExtractor {
} else if (b == '{') {
pushText();
processGroupStart(in);
- } else if (b == '}') {
+ } else if (b == '}') {
pushText();
processGroupEnd();
if (groupStates.isEmpty()) {
@@ -479,9 +459,9 @@ final class TextExtractor {
break;
}
} else if (groupState.objdata == true ||
- groupState.pictDepth == 1) {
+ groupState.pictDepth == 1) {
embObjHandler.writeHexChar(b);
- } else if (b != '\r' && b != '\n'
+ } else if (b != '\r' && b != '\n'
&& (!groupState.ignore || nextMetaData != null ||
groupState.sn == true || groupState.sv == true)) {
// Linefeed and carriage return are not
@@ -497,7 +477,7 @@ final class TextExtractor {
endParagraph(false);
out.endDocument();
}
-
+
private void parseControlToken(PushbackInputStream in) throws IOException, SAXException, TikaException {
int b = in.read();
if (b == '\'') {
@@ -505,16 +485,16 @@ final class TextExtractor {
parseHexChar(in);
} else if (isAlpha(b)) {
// control word
- parseControlWord((char)b, in);
+ parseControlWord((char) b, in);
} else if (b == '{' || b == '}' || b == '\\' || b == '\r' || b == '\n') {
// escaped char
addOutputByte(b);
} else if (b != -1) {
// control symbol, eg \* or \~
- processControlSymbol((char)b);
+ processControlSymbol((char) b);
}
}
-
+
private void parseHexChar(PushbackInputStream in) throws IOException, SAXException, TikaException {
int hex1 = in.read();
if (!isHexChar(hex1)) {
@@ -522,7 +502,7 @@ final class TextExtractor {
in.unread(hex1);
return;
}
-
+
int hex2 = in.read();
if (!isHexChar(hex2)) {
// TODO: log a warning here, somehow?
@@ -531,7 +511,7 @@ final class TextExtractor {
in.unread(hex2);
return;
}
-
+
if (ansiSkip != 0) {
// Skip this ansi char since we are
// still in the shadow of a unicode
@@ -539,19 +519,19 @@ final class TextExtractor {
ansiSkip--;
} else {
// Unescape:
- addOutputByte(16*hexValue(hex1) + hexValue(hex2));
+ addOutputByte(16 * hexValue(hex1) + hexValue(hex2));
}
}
private void parseControlWord(int firstChar, PushbackInputStream in) throws IOException, SAXException, TikaException {
addControl(firstChar);
-
+
int b = in.read();
while (isAlpha(b)) {
addControl(b);
b = in.read();
}
-
+
boolean hasParam = false;
boolean negParam = false;
if (b == '-') {
@@ -567,14 +547,14 @@ final class TextExtractor {
hasParam = true;
b = in.read();
}
-
+
// space is consumed as part of the
// control word, but is not added to the
// control word
if (b != ' ') {
in.unread(b);
}
-
+
if (hasParam) {
if (negParam) {
param = -param;
@@ -583,7 +563,7 @@ final class TextExtractor {
} else {
processControlWord();
}
-
+
pendingControlCount = 0;
}
@@ -602,7 +582,7 @@ final class TextExtractor {
}
if (inList() && pendingListEnd != groupState.list) {
startList(groupState.list);
- }
+ }
if (inList()) {
out.startElement("li");
} else {
@@ -738,7 +718,7 @@ final class TextExtractor {
if (pendingControlCount != s.length()) {
return false;
}
- for(int idx=0;idx<pendingControlCount;idx++) {
+ for (int idx = 0; idx < pendingControlCount; idx++) {
assert isAlpha(s.charAt(idx));
if (((byte) s.charAt(idx)) != pendingControl[idx]) {
return false;
@@ -748,26 +728,26 @@ final class TextExtractor {
}
private void processControlSymbol(char ch) throws IOException, SAXException, TikaException {
- switch(ch) {
- case '~':
- // Non-breaking space -> unicode NON-BREAKING SPACE
- addOutputChar('\u00a0');
- break;
- case '*':
- // Ignorable destination (control words defined after
- // the 1987 RTF spec). These are already handled by
- // processGroupStart()
- break;
- case '-':
- // Optional hyphen -> unicode SOFT HYPHEN
- addOutputChar('\u00ad');
- break;
- case '_':
- // Non-breaking hyphen -> unicode NON-BREAKING HYPHEN
- addOutputChar('\u2011');
- break;
- default:
- break;
+ switch (ch) {
+ case '~':
+ // Non-breaking space -> unicode NON-BREAKING SPACE
+ addOutputChar('\u00a0');
+ break;
+ case '*':
+ // Ignorable destination (control words defined after
+ // the 1987 RTF spec). These are already handled by
+ // processGroupStart()
+ break;
+ case '-':
+ // Optional hyphen -> unicode SOFT HYPHEN
+ addOutputChar('\u00ad');
+ break;
+ case '_':
+ // Non-breaking hyphen -> unicode NON-BREAKING HYPHEN
+ addOutputChar('\u2011');
+ break;
+ default:
+ break;
}
}
@@ -890,8 +870,8 @@ final class TextExtractor {
currentList.templateID = param;
} else if (equals("levelnfc") || equals("levelnfcn")) {
//sanity check to make sure list information isn't corrupt
- if (listTableLevel > -1 &&
- listTableLevel < currentList.numberType.length ) {
+ if (listTableLevel > -1 &&
+ listTableLevel < currentList.numberType.length) {
currentList.numberType[listTableLevel] = param;
}
}
@@ -963,7 +943,7 @@ final class TextExtractor {
} else if (equals("bin")) {
if (param >= 0) {
if (groupState.pictDepth == 1) {
- try{
+ try {
embObjHandler.writeBytes(in, param);
} catch (IOException e) {
//param was out of bounds or something went wrong during writing.
@@ -977,7 +957,7 @@ final class TextExtractor {
while (bytesToRead > 0) {
int r = in.read(tmpArray, 0, Math.min(bytesToRead, tmpArray.length));
if (r < 0) {
- throw new TikaException("unexpected end of file: need " + param + " bytes of binary data, found " + (param-bytesToRead));
+ throw new TikaException("unexpected end of file: need " + param + " bytes of binary data, found " + (param - bytesToRead));
}
bytesToRead -= r;
}
@@ -1005,6 +985,7 @@ final class TextExtractor {
/**
* Emits the end tag of a list. Uses {@link #isUnorderedList(int)} to determine the list
* type for the given <code>listID</code>.
+ *
* @param listID The ID of the list.
* @throws IOException
* @throws SAXException
@@ -1019,6 +1000,7 @@ final class TextExtractor {
/**
* Emits the start tag of a list. Uses {@link #isUnorderedList(int)} to determine the list
* type for the given <code>listID</code>.
+ *
* @param listID The ID of the list.
* @throws IOException
* @throws SAXException
@@ -1051,11 +1033,11 @@ final class TextExtractor {
if (inHeader) {
if (equals("ansi")) {
globalCharset = WINDOWS_1252;
- } else if (equals("pca")) {
+ } else if (equals("pca")) {
globalCharset = CP850;
- } else if (equals("pc")) {
+ } else if (equals("pc")) {
globalCharset = CP437;
- } else if (equals("mac")) {
+ } else if (equals("mac")) {
globalCharset = MAC_ROMAN;
}
@@ -1319,13 +1301,13 @@ final class TextExtractor {
// Make new GroupState
groupState = new GroupState(groupState);
- assert groupStates.size() == groupState.depth: "size=" + groupStates.size() + " depth=" + groupState.depth;
+ assert groupStates.size() == groupState.depth : "size=" + groupStates.size() + " depth=" + groupState.depth;
if (uprState == 0) {
uprState = 1;
groupState.ignore = true;
}
-
+
// Check for ignorable groups. Note that
// sometimes we un-ignore within this group, eg
// when handling upr escape.
@@ -1335,7 +1317,7 @@ final class TextExtractor {
if (b3 == '*') {
groupState.ignore = true;
}
- in.unread(b3);
+ in.unread(b3);
}
in.unread(b2);
}
@@ -1346,7 +1328,7 @@ final class TextExtractor {
if (nextMetaData != null) {
if (nextMetaData == TikaCoreProperties.CREATED) {
Calendar cal = Calendar.getInstance(TimeZone.getDefault(), Locale.ROOT);
- cal.set(year, month-1, day, hour, minute, 0);
+ cal.set(year, month - 1, day, hour, minute, 0);
metadata.set(nextMetaData, cal.getTime());
} else if (nextMetaData.isMultiValuePermitted()) {
metadata.add(nextMetaData, pendingBuffer.toString());
@@ -1360,7 +1342,7 @@ final class TextExtractor {
assert groupState.depth > 0;
ansiSkip = 0;
-
+
if (groupState.objdata == true) {
embObjHandler.handleCompletedObject();
groupState.objdata = false;
@@ -1391,7 +1373,7 @@ final class TextExtractor {
// bold changed:
if (groupState.italic) {
if (!outerGroupState.italic ||
- groupState.bold != outerGroupState.bold) {
+ groupState.bold != outerGroupState.bold) {
end("i");
groupState.italic = false;
}
@@ -1425,9 +1407,9 @@ final class TextExtractor {
final boolean isLocalLink = s.contains("\\l ");
int idx = s.indexOf('"');
if (idx != -1) {
- int idx2 = s.indexOf('"', 1+idx);
+ int idx2 = s.indexOf('"', 1 + idx);
if (idx2 != -1) {
- s = s.substring(1+idx, idx2);
+ s = s.substring(1 + idx, idx2);
}
}
pendingURL = (isLocalLink ? "#" : "") + s;