Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/apache/pdfbox/llms.txt

Use this file to discover all available pages before exploring further.

PDFBox provides several classes for extracting text from PDF documents. PDFTextStripper handles full-document and page-range extraction, PDFTextStripperByArea lets you extract text from a specific rectangular region, and TextPosition gives you per-glyph coordinates and font metrics.

​
Basic text extraction

The simplest approach is to create a PDFTextStripper, optionally configure it, then call getText(). The example below is adapted from ExtractTextSimple.java:
ExtractTextSimple.java
import org.apache.pdfbox.Loader;
import org.apache.pdfbox.pdmodel.PDDocument;
import org.apache.pdfbox.pdmodel.encryption.AccessPermission;
import org.apache.pdfbox.text.PDFTextStripper;

try (PDDocument document = Loader.loadPDF(new File(args[0])))
{
    AccessPermission ap = document.getCurrentAccessPermission();
    if (!ap.canExtractContent())
    {
        throw new IOException("You do not have permission to extract text");
    }

    PDFTextStripper stripper = new PDFTextStripper();

    // This example uses sorting, but in some cases it is more useful to switch it off,
    // e.g. in some files with columns where the PDF content stream respects the
    // column order.
    stripper.setSortByPosition(true);

    String text = stripper.getText(document);
    System.out.println(text);
}
setSortByPosition(true) re-orders characters by their x/y coordinates before building the string, which is useful when text is rendered out of reading order. Turn it off if the PDF’s content stream already encodes the correct reading order (for example, multi-column layouts where the stream follows column order).

​
Extracting text from a specific page range

Set startPage and endPage (both 1-based) before calling getText(). To process each page individually:
ExtractTextSimple.java
PDFTextStripper stripper = new PDFTextStripper();
stripper.setSortByPosition(true);

for (int p = 1; p <= document.getNumberOfPages(); ++p)
{
    // Set the page interval to extract. If you don't, then all pages would be extracted.
    stripper.setStartPage(p);
    stripper.setEndPage(p);

    // let the magic happen
    String text = stripper.getText(document);

    System.out.println("page " + p + ":");
    System.out.println(text.trim());
}
To extract a range of pages (for example, pages 3–5), set setStartPage(3) and setEndPage(5) once and call getText() a single time.

​
Extracting text by region

PDFTextStripperByArea extracts text from a named rectangular region. Coordinates use Java screen coordinates where y = 0 is the top of the page (not the PDF default where y = 0 is the bottom).
ExtractTextByArea.java
import java.awt.Rectangle;
import org.apache.pdfbox.Loader;
import org.apache.pdfbox.pdmodel.PDDocument;
import org.apache.pdfbox.pdmodel.PDPage;
import org.apache.pdfbox.text.PDFTextStripperByArea;

try (PDDocument document = Loader.loadPDF(new File(args[0])))
{
    PDFTextStripperByArea stripper = new PDFTextStripperByArea();
    stripper.setSortByPosition(true);

    // Rectangle(x, y, width, height) in Java screen coordinates
    Rectangle rect = new Rectangle(10, 280, 275, 60);
    stripper.addRegion("class1", rect);

    PDPage firstPage = document.getPage(0);
    stripper.extractRegions(firstPage);

    System.out.println("Text in the area:" + rect);
    System.out.println(stripper.getTextForRegion("class1"));
}
You can add multiple named regions before calling extractRegions(), then retrieve each with getTextForRegion(name).
PDFTextStripperByArea uses Java y-coordinates (y = 0 at top), while PDFBox page coordinates have y = 0 at the bottom. Subtract your PDF y-coordinate from the page height to convert.

​
Getting text positions

Subclass PDFTextStripper and override writeString() to access individual TextPosition objects with character-level coordinates and font information. DrawPrintTextLocations.java in the examples module demonstrates the full pattern:
DrawPrintTextLocations.java
import org.apache.pdfbox.text.PDFTextStripper;
import org.apache.pdfbox.text.TextPosition;
import java.util.List;

public class MyTextLocations extends PDFTextStripper
{
    public MyTextLocations() throws IOException
    {
        super();
    }

    @Override
    protected void writeString(String string, List<TextPosition> textPositions)
            throws IOException
    {
        for (TextPosition text : textPositions)
        {
            System.out.printf(
                "String[%s] at (%f, %f) font size: %f pt%n",
                text.getUnicode(),
                text.getX(),
                text.getY(),
                text.getFontSizeInPt()
            );
        }
        super.writeString(string, textPositions);
    }
}
Key TextPosition methods:
MethodDescription
getX()X coordinate in user space
getY()Y coordinate in user space
getWidth()Width of the character
getFontSizeInPt()Rendered font size in points
getUnicode()Unicode string for this glyph
getFont()The PDFont used to render this character

​
Handling permissions

PDFs can restrict content extraction. Always check AccessPermission.canExtractContent() before attempting extraction:
Permission check
import org.apache.pdfbox.pdmodel.encryption.AccessPermission;

AccessPermission ap = document.getCurrentAccessPermission();
if (!ap.canExtractContent())
{
    throw new IOException("You do not have permission to extract text");
}
Loader.loadPDF(file, password) accepts a password string to open encrypted documents before checking permissions.

​
Common issues and tips

Some PDFs encode text using custom glyph mappings without proper Unicode ToUnicode entries. PDFBox can only decode what the PDF provides. Try opening the file in Adobe Reader and copying text manually — if that also fails, the PDF likely does not contain extractable text.If Adobe Reader can copy text but PDFBox cannot, file a bug report with a minimal reproducer.
By default, PDFTextStripper follows the content stream order, which may not match reading order. Enable setSortByPosition(true) to sort characters by their x/y coordinates.For multi-column documents, position-based sorting can merge columns incorrectly. Try disabling setSortByPosition and rely on stream order instead.
PDFBox infers word spacing from glyph positions. If the inter-glyph distance is smaller than the threshold, spaces are omitted. There is no public API to tune this threshold directly; subclassing PDFTextStripper and overriding writeLine() gives you lower-level control.
Scanned documents are images, not text. PDFBox cannot extract text from image-only PDFs. You need an OCR engine such as Tesseract to first convert the scanned image to a text layer before PDFBox can process it.
Loading a large PDF fully into memory via Loader.loadPDF(File) can be slow. Use Loader.loadPDF(RandomAccessRead) backed by a memory-mapped file for better performance. You can also limit the page range with setStartPage / setEndPage to avoid processing the entire document.

Build docs developers (and LLMs) love

Get started for freeTalk to us