setup refactor v1.5:

* drop usage of charade - no longer maintained
* refactor setup for simplicity

Author: mindw

README.rst

@@ -113,9 +113,8 @@ functionality:
 
 - ``genshi`` has a treewalker (but not builder); and
 
-- ``charade`` can be used as a fallback when character encoding cannot
-  be determined; ``chardet``, from which it was forked, can also be used
-  on Python 2.
+- ``chardet`` can be used as a fallback when character encoding cannot
+  be determined.
 
 - ``ordereddict`` can be used under Python 2.6
   (``collections.OrderedDict`` is used instead on later versions) to

debug-info.py

@@ -12,7 +12,7 @@
     "maxsize": sys.maxsize
 }
 
-search_modules = ["charade", "chardet", "datrie", "genshi", "html5lib", "lxml", "six"]
+search_modules = ["chardet", "datrie", "genshi", "html5lib", "lxml", "six"]
 found_modules = []
 
 for m in search_modules:

html5lib/inputstream.py

@@ -471,10 +471,7 @@ def detectEncoding(self, parseMeta=True, chardet=True):
         if encoding is None and chardet:
             confidence = "tentative"
             try:
-                try:
-                    from charade.universaldetector import UniversalDetector
-                except ImportError:
-                    from chardet.universaldetector import UniversalDetector
+                from chardet.universaldetector import UniversalDetector
                 buffers = []
                 detector = UniversalDetector()
                 while not detector.done:
@@ -490,7 +487,7 @@ def detectEncoding(self, parseMeta=True, chardet=True):
             except ImportError:
                 pass
         # If all else fails use the default encoding
-        if encoding is None:
+        if not encoding:
             confidence = "tentative"
             encoding = lookupEncoding(self.defaultEncoding)
 

html5lib/tests/test_encoding.py

@@ -40,12 +40,9 @@ def test_encoding():
             yield (runPreScanEncodingTest, test[b'data'], test[b'encoding'])
 
 try:
-    try:
-        import charade  # flake8: noqa
-    except ImportError:
-        import chardet  # flake8: noqa
+    import chardet  # flake8: noqa
 except ImportError:
-    print("charade/chardet not found, skipping chardet tests")
+    print("chardet not found, skipping chardet tests")
 else:
     def test_chardet():
         with open(os.path.join(test_dir, "encoding" , "chardet", "test_big5.txt"), "rb") as fp:

requirements-optional.txt

@@ -4,9 +4,9 @@
 # streams.
 genshi
 
-# charade can be used as a fallback in case we are unable to determine
+# chardet can be used as a fallback in case we are unable to determine
 # the encoding of a document.
-charade
+chardet>2.2
 
 # lxml is supported with its own treebuilder ("lxml") and otherwise
 # uses the standard ElementTree support

setup.py

@@ -1,8 +1,6 @@
-import ast
-import os
 import codecs
-
-from setuptools import setup
+from os.path import join, dirname
+from setuptools import setup, find_packages
 
 
 classifiers=[
@@ -22,27 +20,19 @@
     'Topic :: Text Processing :: Markup :: HTML'
     ]
 
-packages = ['html5lib'] + ['html5lib.'+name
-                           for name in os.listdir(os.path.join('html5lib'))
-                           if os.path.isdir(os.path.join('html5lib', name)) and
-                           not name.startswith('.') and name != 'tests']
-
-current_dir = os.path.dirname(__file__)
-with codecs.open(os.path.join(current_dir, 'README.rst'), 'r', 'utf8') as readme_file:
-    with codecs.open(os.path.join(current_dir, 'CHANGES.rst'), 'r', 'utf8') as changes_file:
+here = dirname(__file__)
+with codecs.open(join(here, 'README.rst'), 'r', 'utf8') as readme_file:
+    with codecs.open(join(here, 'CHANGES.rst'), 'r', 'utf8') as changes_file:
         long_description = readme_file.read() + '\n' + changes_file.read()
 
 version = None
-with open(os.path.join("html5lib", "__init__.py"), "rb") as init_file:
-    t = ast.parse(init_file.read(), filename="__init__.py", mode="exec")
-    assert isinstance(t, ast.Module)
-    assignments = filter(lambda x: isinstance(x, ast.Assign), t.body)
-    for a in assignments:
-        if (len(a.targets) == 1 and
-              isinstance(a.targets[0], ast.Name) and
-              a.targets[0].id == "__version__" and
-              isinstance(a.value, ast.Str)):
-            version = a.value.s
+with open(join(here, 'html5lib', '__init__.py')) as fp:
+    for line in fp:
+        _locals = {}
+        if line.startswith('__version__'):
+            exec(line, None, _locals)
+            version = _locals['__version__']
+            break
 
 setup(name='html5lib',
       version=version,
@@ -53,7 +43,7 @@
       classifiers=classifiers,
       maintainer='James Graham',
       maintainer_email='james@hoppipolla.co.uk',
-      packages=packages,
+      packages=find_packages(exclude=["*.tests", "*.tests.*", "tests.*", "tests"]),
       install_requires=[
           'six',
           'webencodings',
@@ -70,13 +60,13 @@
 
           # Standard extras, will be installed when the extra is requested.
           "genshi": ["genshi"],
-          "charade": ["charade"],
+          "chardet": ["chardet>=2.2"],
 
           # The all extra combines a standard extra which will be used anytime
           # the all extra is requested, and it extends it with a conditional
           # extra that will be installed whenever the condition matches and the
           # all extra is requested.
-          "all": ["genshi", "charade"],
+          "all": ["genshi", "chardet>=2.2"],
           "all:python_implementation == 'CPython'": ["datrie", "lxml"],
       },
       )