Resolve un-resolved metadata fields & decode_text

pdfplumber/pdf.py

@@ -1,5 +1,6 @@
 from pdfplumber.container import Container
 from pdfplumber.page import Page
+from pdfplumber.utils import decode_text
 
 from pdfminer.pdfparser import PDFParser
 from pdfminer.pdfdocument import PDFDocument
@@ -22,6 +23,10 @@ class PDF(Container):
         self.metadata = {}
         for info in self.doc.info:
             self.metadata.update(info)
+        for k, v in self.metadata.items():
+            if hasattr(v, "resolve"):
+                v = v.resolve()
+            self.metadata[k] = decode_text(v)
         self.device = PDFPageAggregator(rsrcmgr, laparams=self.laparams)
         self.interpreter = PDFPageInterpreter(rsrcmgr, self.device)
         atexit.register(self.close)

pdfplumber/utils.py

@@ -1,6 +1,19 @@
 from pdfplumber import helpers
 from operator import itemgetter
 import itertools
+from pdfminer.utils import PDFDocEncoding
+import six
+
+def decode_text(s):
+    """
+    Decodes a PDFDocEncoding string to Unicode.
+    Adds py3 compatability to pdfminer's version.
+    """
+    if 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)
+        return ''.join(PDFDocEncoding[o] for o in ords)
 
 def is_dataframe(collection):
     cls = collection.__class__