reading-pdf-annotations.md

Reading PDF Annotations

PDF 2.0 defines the following annotation types:

• Text
• Link
• FreeText
• Line
• Square
• Circle
• Polygon
• PolyLine
• Highlight
• Underline
• Squiggly
• StrikeOut
• Caret
• Stamp
• Ink
• Popup
• FileAttachment
• Sound
• Movie
• Screen
• Widget
• PrinterMark
• TrapNet
• Watermark
• 3D
• Redact
• Projection
• RichMedia

In general, annotations can be read like this:

from pypdf import PdfReader

reader = PdfReader("annotated.pdf")

for page in reader.pages:
    if "/Annots" in page:
        for annot in page["/Annots"]:
            obj = annot.get_object()
            annotation = {"subtype": obj["/Subtype"], "location": obj["/Rect"]}
            print(annotation)

Examples of reading three of the most common annotations:

Text

from pypdf import PdfReader

reader = PdfReader("example.pdf")

for page in reader.pages:
    if "/Annots" in page:
        for annot in page["/Annots"]:
            subtype = annot.get_object()["/Subtype"]
            if subtype == "/Text":
                print(annot.get_object()["/Contents"])

Highlights

from pypdf import PdfReader

reader = PdfReader("example.pdf")

for page in reader.pages:
    if "/Annots" in page:
        for annot in page["/Annots"]:
            subtype = annot.get_object()["/Subtype"]
            if subtype == "/Highlight":
                coords = annot.get_object()["/QuadPoints"]
                x1, y1, x2, y2, x3, y3, x4, y4 = coords

Attachments

from pypdf import PdfReader

reader = PdfReader("example.pdf")

attachments = {}
for page in reader.pages:
    if "/Annots" in page:
        for annotation in page["/Annots"]:
            subtype = annot.get_object()["/Subtype"]
            if subtype == "/FileAttachment":
                fileobj = annotobj["/FS"]
                attachments[fileobj["/F"]] = fileobj["/EF"]["/F"].get_data()