fix github issue pdfminer#566.

.gitignore

@@ -18,4 +18,6 @@ tests/*.txt
 .idea/
 .tox/
 Pipfile
-Pipfile.lock
+Pipfile.lock
+.noseids
+.vscode/

pdfminer/cmapdb.py

@@ -338,7 +338,7 @@ def do_keyword(self, pos, token):
         if token is self.KEYWORD_ENDCIDRANGE:
             objs = [obj for (__, obj) in self.popall()]
             for (s, e, cid) in choplist(3, objs):
-                if (not isinstance(s, str) or not isinstance(e, str) or
+                if (not isinstance(s, bytes) or not isinstance(e, bytes) or
                    not isinstance(cid, int) or len(s) != len(e)):
                     continue
                 sprefix = s[:-4]
@@ -352,7 +352,7 @@ def do_keyword(self, pos, token):
                 vlen = len(svar)
                 for i in range(e1-s1+1):
                     x = sprefix+struct.pack('>L', s1+i)[-vlen:]
-                    self.cmap.add_code2cid(x, cid+i)
+                    self.cmap.add_cid2unichr(cid+i, x)
             return
 
         if token is self.KEYWORD_BEGINCIDCHAR:
@@ -361,8 +361,8 @@ def do_keyword(self, pos, token):
         if token is self.KEYWORD_ENDCIDCHAR:
             objs = [obj for (__, obj) in self.popall()]
             for (cid, code) in choplist(2, objs):
-                if isinstance(code, str) and isinstance(cid, str):
-                    self.cmap.add_code2cid(code, nunpack(cid))
+                if isinstance(code, bytes) and isinstance(cid, bytes):
+                    self.cmap.add_cid2unichr(cid, code)
             return
 
         if token is self.KEYWORD_BEGINBFRANGE:

pdfminer/pdffont.py

@@ -403,9 +403,9 @@ def __init__(self, name, fp):
         return
 
     def create_unicode_map(self):
-        if 'cmap' not in self.tables:
+        if b'cmap' not in self.tables:
             raise TrueTypeFont.CMapNotFound
-        (base_offset, length) = self.tables['cmap']
+        (base_offset, length) = self.tables[b'cmap']
         fp = self.fp
         fp.seek(base_offset)
         (version, nsubtables) = struct.unpack('>HH', fp.read(4))

tests/test_hcheng.py

@@ -0,0 +1,25 @@
+from pdfminer.settings import STRICT
+import sys
+print(sys.path)
+
+from pdfminer import high_level
+
+STRICT = True
+
+# pdf_file = '/home/huan_cheng/workspace/algrithm/contractsimilarity/data/pdf/B3.pdf'
+# pdf_file = '/home/huan_cheng/Documents/pdf_files/A0095607-010169.pdf'
+pdf_file = '/tmp/aa.pdf'
+
+text = high_level.extract_text(
+    pdf_file, 
+    password='', 
+    # page_numbers=[13], 
+    # page_numbers=[0], 
+    maxpages=0, 
+    caching=True, 
+    codec='utf-8', 
+    laparams=None
+)
+print(text)
+
+print("-------> Done!")