Merge pull request #162 from babenek/doc

WPSOffice doc format detection

Author: h2non

filetype/types/document.py

@@ -116,8 +116,12 @@ def match(self, buf):
                 return True
             if (
                 len(buf) > 2142
-                and b"\x00\x0A\x00\x00\x00MSWordDoc\x00\x10\x00\x00\x00Word.Document.8\x00\xF49\xB2q"
-                in buf[2075:2142]
+                and (
+                    b"\x00\x0A\x00\x00\x00MSWordDoc\x00\x10\x00\x00\x00Word.Document.8\x00\xF49\xB2q"
+                    in buf[2075:2142]
+                    or b"W\0o\0r\0d\0D\0o\0c\0u\0m\0e\0n\0t\0"
+                    in buf[0x580:0x598]
+                )
             ):
                 return True
 

tox.ini

@@ -26,7 +26,7 @@ commands = pytest \
 basepython = python3
 deps =
      flake8
-commands = flake8 {toxinidir} --extend-exclude tests,docs,build,dist,venv --extend-ignore=E501
+commands = flake8 {toxinidir} --extend-exclude tests,docs,build,dist,venv,.venv --extend-ignore=E501
 
 [testenv:doc]
 basepython = python3