1 files changed, 67 insertions, 0 deletions
diff --git a/annotation-summary.py b/annotation-summary.py
new file mode 100755
index 0000000..0b0fb5e
--- /dev/null
+++ b/annotation-summary.py
@@ -0,0 +1,67 @@
+#!/usr/bin/env python3
+
+import argparse
+import popplerqt5
+import pdfkit
+
+def extract(fn, show=False):
+    doc = popplerqt5.Poppler.Document.load(fn)
+    annotations = {}
+    for i in range(doc.numPages()):
+        annotations[i] = []
+        for annot in doc.page(i).annotations():
+            annotations[i].append(annot)
+    return annotations
+
+def generate_pdf(annotations, destination="annotations.pdf", pagesize="A4"):
+    options = {
+            'page-size': pagesize,
+            'encoding': "utf-8",
+            'margin-top': '0.75in',
+            'margin-right': '0.75in',
+            'margin-bottom': '0.75in',
+            'margin-left': '0.75in',
+            }
+
+    blocks = """
+        <!DOCTYPE html>
+        <html>
+        <head>
+            <meta charset="utf-8">
+        </head>
+        <body>
+    """
+
+    pages = list(annotations.keys())
+    pages.sort()
+    for page in pages:
+        for annotation in annotations[page]:
+            content = annotation.contents()
+            if not content:
+                continue
+            author = annotation.author()
+            if author != "":
+                header = f"<span>Page: {page + 1}</span> | <span>Author: {author}</span>"
+            else:
+                header = f"<span>Page: {page + 1}</span>"
+            blocks += f"""
+                <div style="border: 1px solid #300;">
+                    {header}<br/>
+                    <hr/>
+                    <div style="white-space: pre-wrap;">{content}</div>
+                </div>
+                <br/>
+            """
+
+    blocks += """
+            </body>
+        </html>
+    """
+    pdfkit.from_string(blocks, destination, options=options)
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('source')
+    args = parser.parse_args()
+    annotations = extract(args.source, show=True)
+    generate_pdf(annotations)