Skip to content

Commit

Permalink
upgraded pdfbox to 3.0.0
Browse files Browse the repository at this point in the history
  • Loading branch information
@Orbiter
Orbiter committed Oct 27, 2023
1 parent c10944b commit 5ba5fb5
Show file tree
Hide file tree
Showing 3 changed files with 46 additions and 43 deletions.
2 changes: 1 addition & 1 deletion ivy.xml
View file
Expand Up @@ -55,7 +55,7 @@
<dependency org="org.apache.lucene" name="lucene-queryparser" rev="8.11.2" conf="compile->master"/>
<dependency org="org.apache.lucene" name="lucene-spatial-extras" rev="8.11.2" conf="compile->master"/>
<dependency org="org.apache.lucene" name="lucene-suggest" rev="8.11.2"/>
<dependency org="org.apache.pdfbox" name="pdfbox" rev="2.0.29" />
<dependency org="org.apache.pdfbox" name="pdfbox" rev="3.0.0" />
<dependency org="org.apache.poi" name="poi" rev="3.17" />
<dependency org="org.apache.poi" name="poi-scratchpad" rev="3.17" />
<dependency org="org.apache.solr" name="solr-core" rev="8.11.2" conf="compile->master"/>
Expand Down
13 changes: 7 additions & 6 deletions source/net/yacy/cora/util/Html2Image.java
View file
Expand Up @@ -43,6 +43,7 @@
import javax.swing.text.html.HTMLEditorKit;
import javax.swing.text.html.ImageView;

import org.apache.pdfbox.Loader;
import org.apache.pdfbox.pdmodel.PDDocument;
import org.apache.pdfbox.rendering.ImageType;
import org.apache.pdfbox.rendering.PDFRenderer;
Expand Down Expand Up @@ -211,7 +212,7 @@ private static boolean convertAvailableInPath() {
* call termination. Beyond this limit the process is killed.
* @return true when the destination file was successfully written
*/
public static boolean writeWkhtmltopdf(String url, String proxy, String userAgent, final String acceptLanguage, final File destination, final long maxSeconds) {
public static boolean writeWkhtmltopdf(final String url, final String proxy, final String userAgent, final String acceptLanguage, final File destination, final long maxSeconds) {
boolean success = false;
for (final boolean ignoreErrors: new boolean[]{false, true}) {
success = writeWkhtmltopdfInternal(url, proxy, destination, userAgent, acceptLanguage, ignoreErrors, maxSeconds);
Expand Down Expand Up @@ -352,7 +353,7 @@ public static boolean pdf2image(final File pdf, final File image, final int widt

// convert pdf to jpg using internal pdfbox capability
if (convertCmd == null) {
try (final PDDocument pdoc = PDDocument.load(pdf);) {
try (final PDDocument pdoc = Loader.loadPDF(pdf);) {

final BufferedImage bi = new PDFRenderer(pdoc).renderImageWithDPI(0, density, ImageType.RGB);

Expand Down Expand Up @@ -432,7 +433,7 @@ public static boolean pdf2image(final File pdf, final File image, final int widt
* @param size
* @throws IOException
*/
public static void writeSwingImage(String url, Dimension size, File destination) throws IOException {
public static void writeSwingImage(final String url, final Dimension size, final File destination) throws IOException {

// set up a pane for rendering
final JEditorPane htmlPane = new JEditorPane();
Expand All @@ -453,7 +454,7 @@ public Document createDefaultDocument() {
public ViewFactory getViewFactory() {
return new HTMLFactory() {
@Override
public View create(Element elem) {
public View create(final Element elem) {
final View view = super.create(elem);
if (view instanceof ImageView) {
((ImageView) view).setLoadsSynchronously(true);
Expand All @@ -467,7 +468,7 @@ public View create(Element elem) {
htmlPane.setContentType("text/html");
htmlPane.addPropertyChangeListener(new PropertyChangeListener() {
@Override
public void propertyChange(PropertyChangeEvent evt) {
public void propertyChange(final PropertyChangeEvent evt) {
}
});

Expand Down Expand Up @@ -501,7 +502,7 @@ public void propertyChange(PropertyChangeEvent evt) {
* </li>
* </ol>
*/
public static void main(String[] args) {
public static void main(final String[] args) {
final String usageMessage = "Usage : java " + Html2Image.class.getName()
+ " <url> <target-file[.pdf|.jpg|.png]> [wkhtmltopdf|swing]";
int exitStatus = 0;
Expand Down
74 changes: 38 additions & 36 deletions source/net/yacy/document/parser/pdfParser.java
View file
Expand Up @@ -39,8 +39,10 @@
import java.util.HashSet;
import java.util.List;

import org.apache.pdfbox.Loader;
import org.apache.pdfbox.cos.COSName;
import org.apache.pdfbox.io.MemoryUsageSetting;
import org.apache.pdfbox.io.RandomAccessRead;
import org.apache.pdfbox.io.RandomAccessReadBuffer;
import org.apache.pdfbox.pdmodel.PDDocument;
import org.apache.pdfbox.pdmodel.PDDocumentInformation;
import org.apache.pdfbox.pdmodel.PDPage;
Expand Down Expand Up @@ -69,7 +71,7 @@ public class pdfParser extends AbstractParser implements Parser {

public static boolean individualPages = false;
public static String individualPagePropertyname = "page";

public pdfParser() {
super("Acrobat Portable Document Parser");
this.SUPPORTED_EXTENSIONS.add("pdf");
Expand All @@ -86,7 +88,7 @@ public Document[] parse(
final DigestURL location,
final String mimeType,
final String charset,
final VocabularyScraper scraper,
final VocabularyScraper scraper,
final int timezoneOffset,
final InputStream source) throws Parser.Failure, InterruptedException {

Expand All @@ -98,8 +100,8 @@ public Document[] parse(
PDDocument pdfDoc;
try {
Thread.currentThread().setPriority(Thread.MIN_PRIORITY); // the pdfparser is a big pain
MemoryUsageSetting mus = MemoryUsageSetting.setupMixed(200*1024*1024);
pdfDoc = PDDocument.load(source, mus);
final RandomAccessRead readBuffer = new RandomAccessReadBuffer(source);
pdfDoc = Loader.loadPDF(readBuffer);
} catch (final IOException e) {
throw new Parser.Failure(e.getMessage(), location);
} finally {
Expand Down Expand Up @@ -141,34 +143,34 @@ public Document[] parse(
if (docKeywordStr != null) {
docKeywords = docKeywordStr.split(" |,");
}

Document[] result = null;
try {
// get the links
final List<Collection<AnchorURL>> pdflinks = extractPdfLinks(pdfDoc);

// get the fulltext (either per document or for each page)
final PDFTextStripper stripper = new PDFTextStripper(/*StandardCharsets.UTF_8.name()*/);

if (individualPages) {
// this is a hack which stores individual pages of the source pdf into individual index documents
// the new documents will get a virtual link with a post argument page=X appended to the original url

// collect text
int pagecount = pdfDoc.getNumberOfPages();
String[] pages = new String[pagecount];
final int pagecount = pdfDoc.getNumberOfPages();
final String[] pages = new String[pagecount];
for (int page = 1; page <= pagecount; page++) {
stripper.setStartPage(page);
stripper.setEndPage(page);
pages[page - 1] = stripper.getText(pdfDoc);
//System.out.println("PAGE " + page + ": " + pages[page - 1]);
}

// create individual documents for each page
assert pages.length == pdflinks.size() : "pages.length = " + pages.length + ", pdflinks.length = " + pdflinks.size();
result = new Document[Math.min(pages.length, pdflinks.size())];
String loc = location.toNormalform(true);
for (int page = 0; page < result.length; page++) {
final String loc = location.toNormalform(true);
for (int page = 0; page < result.length; page++) {
result[page] = new Document(
new AnchorURL(loc + (loc.indexOf('?') > 0 ? '&' : '?') + individualPagePropertyname + '=' + (page + 1)), // these are virtual new pages; we cannot combine them with '#' as that would be removed when computing the urlhash
mimeType,
Expand Down Expand Up @@ -216,9 +218,9 @@ public void run() {
contentBytes = writer.getBytes(); // get final text before closing writer
writer.close(); // free writer resources
}
Collection<AnchorURL> pdflinksCombined = new HashSet<AnchorURL>();
for (Collection<AnchorURL> pdflinksx: pdflinks) if (pdflinksx != null) pdflinksCombined.addAll(pdflinksx);

final Collection<AnchorURL> pdflinksCombined = new HashSet<>();
for (final Collection<AnchorURL> pdflinksx: pdflinks) if (pdflinksx != null) pdflinksCombined.addAll(pdflinksx);
result = new Document[]{new Document(
location,
mimeType,
Expand All @@ -238,7 +240,7 @@ public void run() {
null,
false,
docDate)};
}
}
} catch (final Throwable e) {
//throw new Parser.Failure(e.getMessage(), location);
} finally {
Expand All @@ -248,7 +250,7 @@ public void run() {
// clear cached resources in pdfbox.
pdfDoc = null;
clearPdfBoxCaches();

return result;
}

Expand All @@ -258,25 +260,25 @@ public void run() {
* @return all detected links
*/
private List<Collection<AnchorURL>> extractPdfLinks(final PDDocument pdf) {
List<Collection<AnchorURL>> linkCollections = new ArrayList<>(pdf.getNumberOfPages());
for (PDPage page : pdf.getPages()) {
final Collection<AnchorURL> pdflinks = new ArrayList<AnchorURL>();
final List<Collection<AnchorURL>> linkCollections = new ArrayList<>(pdf.getNumberOfPages());
for (final PDPage page : pdf.getPages()) {
final Collection<AnchorURL> pdflinks = new ArrayList<>();
try {
List<PDAnnotation> annotations = page.getAnnotations();
final List<PDAnnotation> annotations = page.getAnnotations();
if (annotations != null) {
for (PDAnnotation pdfannotation : annotations) {
for (final PDAnnotation pdfannotation : annotations) {
if (pdfannotation instanceof PDAnnotationLink) {
PDAction link = ((PDAnnotationLink)pdfannotation).getAction();
final PDAction link = ((PDAnnotationLink)pdfannotation).getAction();
if (link != null && link instanceof PDActionURI) {
PDActionURI pdflinkuri = (PDActionURI) link;
String uristr = pdflinkuri.getURI();
AnchorURL url = new AnchorURL(uristr);
final PDActionURI pdflinkuri = (PDActionURI) link;
final String uristr = pdflinkuri.getURI();
final AnchorURL url = new AnchorURL(uristr);
pdflinks.add(url);
}
}
}
}
} catch (IOException ex) {}
} catch (final IOException ex) {}
linkCollections.add(pdflinks);
}
return linkCollections;
Expand All @@ -292,17 +294,17 @@ public static void clearPdfBoxCaches() {
* situation is now from far better, but one (unnecessary?) cache structure in
* the COSName class still needs to be explicitely cleared.
*/

// History of related issues :
// http://markmail.org/thread/quk5odee4hbsauhu
// https://issues.apache.org/jira/browse/PDFBOX-313
// https://issues.apache.org/jira/browse/PDFBOX-313
// https://issues.apache.org/jira/browse/PDFBOX-351
// https://issues.apache.org/jira/browse/PDFBOX-441
// https://issues.apache.org/jira/browse/PDFBOX-2200
// https://issues.apache.org/jira/browse/PDFBOX-2149

COSName.clearResources();

/*
* Prior to PDFBox 2.0.0, clearResources() function had to be called on the
* org.apache.pdfbox.pdmodel.font.PDFont class and its children. After version
Expand All @@ -327,7 +329,7 @@ public static void main(final String[] args) {
// parse
final AbstractParser parser = new pdfParser();
Document document = null;
FileInputStream inStream = null;
FileInputStream inStream = null;
try {
inStream = new FileInputStream(pdfFile);
document = Document.mergeDocuments(null, "application/pdf", parser.parse(null, "application/pdf", null, new VocabularyScraper(), 0, inStream));
Expand All @@ -345,7 +347,7 @@ public static void main(final String[] args) {
if(inStream != null) {
try {
inStream.close();
} catch(IOException e) {
} catch(final IOException e) {
System.err.println("Could not close input stream on file " + pdfFile);
}
}
Expand All @@ -359,7 +361,7 @@ public static void main(final String[] args) {
System.out.println("\t!!!Parsing without result!!!");
} else {
System.out.println("\tParsed text with " + document.getTextLength() + " chars of text and " + document.getAnchors().size() + " anchors");
InputStream textStream = document.getTextStream();
final InputStream textStream = document.getTextStream();
try {
// write file
FileUtils.copy(textStream, new File("parsedPdf.txt"));
Expand All @@ -372,7 +374,7 @@ public static void main(final String[] args) {
/* textStream can be a FileInputStream : we must close it to ensure releasing system resource */
textStream.close();
}
} catch (IOException e) {
} catch (final IOException e) {
ConcurrentLog.warn("PDFPARSER", "Could not close text input stream");
}
}
Expand Down

0 comments on commit 5ba5fb5

Please sign in to comment.