Use charset-normalizer instead of chardet (#744)

* Use charset-normalizer instead of chardet

* Ignore charset_normalizer type stub

* Add CHANGELOG.md

CHANGELOG.md

@@ -19,6 +19,10 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 
 - Exporting images without any specific encoding ([#737](https://github.com/pdfminer/pdfminer.six/pull/737))
 
+### Changed
+
+- Using charset-normalizer instead of chardet for less restrictive license ([#744](https://github.com/pdfminer/pdfminer.six/pull/744))
+
 ## [20220319]
 
 ### Added

mypy.ini

@@ -23,8 +23,11 @@ ignore_missing_imports = True
 [mypy-pytest.*]
 ignore_missing_imports = True
 
-[mypy-setuptools]
+[mypy-setuptools.*]
 ignore_missing_imports = True
 
-[mypy-nox]
+[mypy-nox.*]
+ignore_missing_imports = True
+
+[mypy-charset_normalizer.*]
 ignore_missing_imports = True

pdfminer/utils.py

@@ -28,7 +28,7 @@ from typing import (
 if TYPE_CHECKING:
     from .layout import LTComponent
 
-import chardet  # For str encoding detection
+import charset_normalizer  # For str encoding detection
 
 # from sys import maxint as INF doesn't work anymore under Python3, but PDF
 # still uses 32 bits ints
@@ -75,7 +75,7 @@ def make_compat_bytes(in_str: str) -> bytes:
 def make_compat_str(o: object) -> str:
     """Converts everything to string, if bytes guessing the encoding."""
     if isinstance(o, bytes):
-        enc = chardet.detect(o)
+        enc = charset_normalizer.detect(o)
         try:
             return o.decode(enc["encoding"])
         except UnicodeDecodeError:

setup.py

@@ -17,8 +17,8 @@ setup(
     packages=["pdfminer"],
     package_data={"pdfminer": ["cmap/*.pickle.gz", "py.typed"]},
     install_requires=[
-        'chardet ; python_version > "3.0"',
-        "cryptography",
+        "charset-normalizer~=2.0.0",
+        "cryptography~=36.0.0",
     ],
     extras_require={
         "dev": ["pytest", "nox", "black", "mypy == 0.931"],