Merge pull request #519 from jsvine/issue/463

Handle utf-16-encoded annotations (#463)

CHANGELOG.md

@@ -12,6 +12,7 @@ All notable changes to this project will be documented in this file. The format
 
 ### Fixed
 - Fix slowdown in `.extract_words(...)`/`WordExtractor.iter_chars_to_words(...)` on very long words, caused by repeatedly re-calculating bounding box. ([#483](https://github.com/jsvine/pdfplumber/discussions/483))
+- Handle `UnicodeDecodeError` when trying to decode utf-16-encoded annotations ([#463](https://github.com/jsvine/pdfplumber/issues/463)) [h/t @tungph]
 
 ### Development Changes
 - Add `CONTRIBUTING.md` ([#428](https://github.com/jsvine/pdfplumber/pull/428))

pdfplumber/page.py

@@ -124,7 +124,10 @@ def parse(annot):
             }
             for k, v in extras.items():
                 if v is not None:
-                    extras[k] = v.decode("utf-8")
+                    try:
+                        extras[k] = v.decode("utf-8")
+                    except UnicodeDecodeError:
+                        extras[k] = v.decode("utf-16")
 
             parsed = {
                 "page_number": self.page_number,

tests/test_issues.py

@@ -187,3 +187,12 @@ def test_issue_386(self):
         with pdfplumber.open(path) as pdf:
             chars = (char for char in pdf.chars)
             pdfplumber.utils.extract_text(chars)
+
+    def test_issue_463(self):
+        """
+        Extracting annotations should not raise UnicodeDecodeError on utf-16 text
+        """
+        path = os.path.join(HERE, "pdfs/issue-463-example.pdf")
+        with pdfplumber.open(path) as pdf:
+            annots = pdf.annots
+            annots[0]["contents"] == "日本語"