Fix issue jsvine#33 by checking decode_text arg type

t-t-t-t-t · May 11, 2017 · c9e45ce · c9e45ce
1 parent 8465877
commit c9e45ce
Show file tree

Hide file tree

Showing 3 changed files with 10 additions and 3 deletions.
diff --git a/pdfplumber/utils.py b/pdfplumber/utils.py
@@ -62,7 +62,7 @@ def decode_text(s):
     Decodes a PDFDocEncoding string to Unicode.
     Adds py3 compatability to pdfminer's version.
     """
-    if s.startswith(b'\xfe\xff'):
+    if type(s) == bytes and s.startswith(b'\xfe\xff'):
         return six.text_type(s[2:], 'utf-16be', 'ignore')
     else:
         ords = (ord(c) if type(c) == str else c for c in s)

diff --git a/tests/pdfs/issue-33-lorem-ipsum.pdf b/tests/pdfs/issue-33-lorem-ipsum.pdf
diff --git a/tests/test-issues.py b/tests/test-issues.py
@@ -77,10 +77,17 @@ def test_issue_14(self):
         pdf = pdfplumber.from_path(
             os.path.join(HERE, "pdfs/cupertino_usd_4-6-16.pdf")
         )
-        len(pdf.objects)
+        assert len(pdf.objects)
 
     def test_issue_21(self):
         pdf = pdfplumber.from_path(
             os.path.join(HERE, "pdfs/150109DSP-Milw-505-90D.pdf")
         )
-        len(pdf.objects)
+        assert len(pdf.objects)
+
+    def test_issue_33(self):
+        pdf = pdfplumber.from_path(
+            os.path.join(HERE, "pdfs/issue-33-lorem-ipsum.pdf")
+        )
+        assert len(pdf.metadata.keys())
+