Fix `TypeError` by Ignoring null characters in PSBaseParser (#768)

* Ignore null characters in PSBaseParser

Beforehand, null characters were encoded as PSKeyword tokens. This caused
issue #617, as pdfdevice.py would attempt to decode the null character
PSKeyword, when it expects a byte string, as opposed to a PSKeyword, causing
pdfminer.six to crash.

As null characters are superfluous within PSBaseParser, ignore them.

* Update CHANGELOG.md

Co-authored-by: Pieter Marsman <pietermarsman@gmail.com>

CHANGELOG.md

@@ -11,6 +11,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 - `ValueError` when trying to decrypt empty metadata values ([#766](https://github.com/pdfminer/pdfminer.six/issues/766))
 - Sphinx errors during building of documentation ([#760](https://github.com/pdfminer/pdfminer.six/pull/760))
 - `TypeError` when getting default width of font ([#720](https://github.com/pdfminer/pdfminer.six/issues/720))
+- `TypeError` in cmapdb.py when parsing null characters ([#768](https://github.com/pdfminer/pdfminer.six/pull/768))
 
 ### Deprecated
 

pdfminer/psparser.py

@@ -334,6 +334,8 @@ class PSBaseParser:
             self._curtoken = b""
             self._parse1 = self._parse_wclose
             return j + 1
+        elif c == b"\x00":
+            return j + 1
         else:
             self._add_token(KWD(c))
             return j + 1