Remove Python 2 compatibility (#1348)

* Remove Python 2 compatibility

* remove 2/3 shims in pygments.util
* update setup.py metadata

* Remove unneeded object inheritance.

* Remove unneeded future imports.

.github/workflows/build.yaml

@@ -9,7 +9,7 @@ jobs:
     strategy:
       max-parallel: 4
       matrix:
-        python-version: [2.7, 3.5, 3.6, 3.7, 3.8, pypy2, pypy3]
+        python-version: [3.5, 3.6, 3.7, 3.8, pypy3]
 
     steps:
     - uses: actions/checkout@v1

CHANGES

@@ -11,6 +11,9 @@ Version 2.6
 -----------
 (not released yet)
 
+- Running Pygments on Python 2.x is no longer supported.
+  (The Python 2 lexer still exists.)
+
 - Added lexers:
 
   * Linux kernel logs (PR#1310)

pygments/__init__.py

@@ -26,8 +26,7 @@
     :license: BSD, see LICENSE for details.
 """
 import sys
-
-from pygments.util import StringIO, BytesIO
+from io import StringIO, BytesIO
 
 __version__ = '2.5.2'
 __docformat__ = 'restructuredtext'

pygments/cmdline.py

@@ -9,16 +9,15 @@
     :license: BSD, see LICENSE for details.
 """
 
-from __future__ import print_function
-
 import os
 import sys
 import getopt
 from textwrap import dedent
 
 from pygments import __version__, highlight
 from pygments.util import ClassNotFound, OptionError, docstring_headline, \
-    guess_decode, guess_decode_from_terminal, terminal_encoding
+    guess_decode, guess_decode_from_terminal, terminal_encoding, \
+    UnclosingTextIOWrapper
 from pygments.lexers import get_all_lexers, get_lexer_by_name, guess_lexer, \
     load_lexer_from_file, get_lexer_for_filename, find_lexer_class_for_filename
 from pygments.lexers.special import TextLexer
@@ -397,11 +396,7 @@ def main_inner(popts, args, usage):
     elif '-s' not in opts:  # treat stdin as full file (-s support is later)
         # read code from terminal, always in binary mode since we want to
         # decode ourselves and be tolerant with it
-        if sys.version_info > (3,):
-            # Python 3: we have to use .buffer to get a binary stream
-            code = sys.stdin.buffer.read()
-        else:
-            code = sys.stdin.read()
+        code = sys.stdin.buffer.read()  # use .buffer to get a binary stream
         if not inencoding:
             code, inencoding = guess_decode_from_terminal(code, sys.stdin)
             # else the lexer will do the decoding
@@ -466,11 +461,7 @@ def main_inner(popts, args, usage):
                 fmter = Terminal256Formatter(**parsed_opts)
             else:
                 fmter = TerminalFormatter(**parsed_opts)
-        if sys.version_info > (3,):
-            # Python 3: we have to use .buffer to get a binary stream
-            outfile = sys.stdout.buffer
-        else:
-            outfile = sys.stdout
+        outfile = sys.stdout.buffer
 
     # determine output encoding if not explicitly selected
     if not outencoding:
@@ -485,10 +476,8 @@ def main_inner(popts, args, usage):
     if not outfn and sys.platform in ('win32', 'cygwin') and \
        fmter.name in ('Terminal', 'Terminal256'):  # pragma: no cover
         # unfortunately colorama doesn't support binary streams on Py3
-        if sys.version_info > (3,):
-            from pygments.util import UnclosingTextIOWrapper
-            outfile = UnclosingTextIOWrapper(outfile, encoding=fmter.encoding)
-            fmter.encoding = None
+        outfile = UnclosingTextIOWrapper(outfile, encoding=fmter.encoding)
+        fmter.encoding = None
         try:
             import colorama.initialise
         except ImportError:
@@ -515,11 +504,7 @@ def main_inner(popts, args, usage):
         # line by line processing of stdin (eg: for 'tail -f')...
         try:
             while 1:
-                if sys.version_info > (3,):
-                    # Python 3: we have to use .buffer to get a binary stream
-                    line = sys.stdin.buffer.readline()
-                else:
-                    line = sys.stdin.readline()
+                line = sys.stdin.buffer.readline()
                 if not line:
                     break
                 if not inencoding:

pygments/filter.py

@@ -40,7 +40,7 @@ def lowercase(self, lexer, stream, options):
     })
 
 
-class Filter(object):
+class Filter:
     """
     Default filter. Subclass this class or use the `simplefilter`
     decorator to create own filters.

pygments/filters/__init__.py

@@ -16,7 +16,7 @@
     string_to_tokentype
 from pygments.filter import Filter
 from pygments.util import get_list_opt, get_int_opt, get_bool_opt, \
-     get_choice_opt, ClassNotFound, OptionError, text_type, string_types
+    get_choice_opt, ClassNotFound, OptionError
 from pygments.plugin import find_plugin_filters
 
 
@@ -113,7 +113,7 @@ def __init__(self, **options):
         Filter.__init__(self, **options)
         case = get_choice_opt(options, 'case',
                               ['lower', 'upper', 'capitalize'], 'lower')
-        self.convert = getattr(text_type, case)
+        self.convert = getattr(str, case)
 
     def filter(self, lexer, stream):
         for ttype, value in stream:
@@ -233,7 +233,7 @@ def __init__(self, **options):
                               ('tabs',     u'»'),
                               ('newlines', u'¶')]:
             opt = options.get(name, False)
-            if isinstance(opt, string_types) and len(opt) == 1:
+            if isinstance(opt, str) and len(opt) == 1:
                 setattr(self, name, opt)
             else:
                 setattr(self, name, (opt and default or ''))
@@ -250,6 +250,7 @@ def filter(self, lexer, stream):
             tabs = self.tabs or u'\t'
             newlines = self.newlines or u'\n'
             regex = re.compile(r'\s')
+
             def replacefunc(wschar):
                 if wschar == ' ':
                     return spaces
@@ -302,7 +303,7 @@ def gobble(self, value, left):
 
     def filter(self, lexer, stream):
         n = self.n
-        left = n # How many characters left to gobble.
+        left = n  # How many characters left to gobble.
         for ttype, value in stream:
             # Remove ``left`` tokens from first line, ``n`` from all others.
             parts = value.split('\n')

pygments/formatter.py

@@ -11,19 +11,19 @@
 
 import codecs
 
-from pygments.util import get_bool_opt, string_types
+from pygments.util import get_bool_opt
 from pygments.styles import get_style_by_name
 
 __all__ = ['Formatter']
 
 
 def _lookup_style(style):
-    if isinstance(style, string_types):
+    if isinstance(style, str):
         return get_style_by_name(style)
     return style
 
 
-class Formatter(object):
+class Formatter:
     """
     Converts a token stream to text.
 

pygments/formatters/__init__.py

@@ -17,7 +17,7 @@
 
 from pygments.formatters._mapping import FORMATTERS
 from pygments.plugin import find_plugin_formatters
-from pygments.util import ClassNotFound, itervalues
+from pygments.util import ClassNotFound
 
 __all__ = ['get_formatter_by_name', 'get_formatter_for_filename',
            'get_all_formatters', 'load_formatter_from_file'] + list(FORMATTERS)
@@ -45,7 +45,7 @@ def _load_formatters(module_name):
 def get_all_formatters():
     """Return a generator for all formatter classes."""
     # NB: this returns formatter classes, not info like get_all_lexers().
-    for info in itervalues(FORMATTERS):
+    for info in FORMATTERS.values():
         if info[1] not in _formatter_cache:
             _load_formatters(info[0])
         yield _formatter_cache[info[1]]
@@ -58,7 +58,7 @@ def find_formatter_class(alias):
 
     Returns None if not found.
     """
-    for module_name, name, aliases, _, _ in itervalues(FORMATTERS):
+    for module_name, name, aliases, _, _ in FORMATTERS.values():
         if alias in aliases:
             if name not in _formatter_cache:
                 _load_formatters(module_name)
@@ -121,7 +121,7 @@ def get_formatter_for_filename(fn, **options):
     Raises ClassNotFound if not found.
     """
     fn = basename(fn)
-    for modname, name, _, filenames, _ in itervalues(FORMATTERS):
+    for modname, name, _, filenames, _ in FORMATTERS.values():
         for filename in filenames:
             if _fn_matches(fn, filename):
                 if name not in _formatter_cache:

pygments/formatters/_mapping.py

@@ -13,8 +13,6 @@
     :license: BSD, see LICENSE for details.
 """
 
-from __future__ import print_function
-
 FORMATTERS = {
     'BBCodeFormatter': ('pygments.formatters.bbcode', 'BBCode', ('bbcode', 'bb'), (), 'Format tokens with BBcodes. These formatting codes are used by many bulletin boards, so you can highlight your sourcecode with pygments before posting it there.'),
     'BmpImageFormatter': ('pygments.formatters.img', 'img_bmp', ('bmp', 'bitmap'), ('*.bmp',), 'Create a bitmap image from source code. This uses the Python Imaging Library to generate a pixmap from the source code.'),

pygments/formatters/html.py

@@ -9,16 +9,14 @@
     :license: BSD, see LICENSE for details.
 """
 
-from __future__ import print_function
-
 import os
 import sys
 import os.path
+from io import StringIO
 
 from pygments.formatter import Formatter
 from pygments.token import Token, Text, STANDARD_TYPES
-from pygments.util import get_bool_opt, get_int_opt, get_list_opt, \
-    StringIO, string_types, iteritems
+from pygments.util import get_bool_opt, get_int_opt, get_list_opt
 
 try:
     import ctags
@@ -41,12 +39,14 @@ def escape_html(text, table=_escape_html_table):
     """Escape &, <, > as well as single and double quotes for HTML."""
     return text.translate(table)
 
+
 def webify(color):
     if color.startswith('calc') or color.startswith('var'):
         return color
     else:
         return '#' + color
 
+
 def _get_ttype_class(ttype):
     fname = STANDARD_TYPES.get(ttype)
     if fname:
@@ -497,7 +497,7 @@ def get_style_defs(self, arg=None):
         """
         if arg is None:
             arg = ('cssclass' in self.options and '.'+self.cssclass or '')
-        if isinstance(arg, string_types):
+        if isinstance(arg, str):
             args = [arg]
         else:
             args = list(arg)
@@ -511,7 +511,7 @@ def prefix(cls):
             return ', '.join(tmp)
 
         styles = [(level, ttype, cls, style)
-                  for cls, (style, ttype, level) in iteritems(self.class2style)
+                  for cls, (style, ttype, level) in self.class2style.items()
                   if cls and style]
         styles.sort()
         lines = ['%s { %s } /* %s */' % (prefix(cls), style, repr(ttype)[6:])

pygments/formatters/img.py

@@ -14,7 +14,7 @@
 
 from pygments.formatter import Formatter
 from pygments.util import get_bool_opt, get_int_opt, get_list_opt, \
-    get_choice_opt, xrange
+    get_choice_opt
 
 import subprocess
 
@@ -59,7 +59,7 @@ class FontNotFound(Exception):
     """When there are no usable fonts specified"""
 
 
-class FontManager(object):
+class FontManager:
     """
     Manages a set of fonts: normal, italic, bold, etc...
     """
@@ -504,7 +504,7 @@ def _draw_line_numbers(self):
         """
         if not self.line_numbers:
             return
-        for p in xrange(self.maxlineno):
+        for p in range(self.maxlineno):
             n = p + self.line_number_start
             if (n % self.line_number_step) == 0:
                 self._draw_linenumber(p, n)

pygments/formatters/latex.py

@@ -9,13 +9,12 @@
     :license: BSD, see LICENSE for details.
 """
 
-from __future__ import division
+from io import StringIO
 
 from pygments.formatter import Formatter
 from pygments.lexer import Lexer
 from pygments.token import Token, STANDARD_TYPES
-from pygments.util import get_bool_opt, get_int_opt, StringIO, xrange, \
-    iteritems
+from pygments.util import get_bool_opt, get_int_opt
 
 
 __all__ = ['LatexFormatter']
@@ -322,7 +321,7 @@ def get_style_defs(self, arg=''):
         """
         cp = self.commandprefix
         styles = []
-        for name, definition in iteritems(self.cmd2def):
+        for name, definition in self.cmd2def.items():
             styles.append(r'\expandafter\def\csname %s@tok@%s\endcsname{%s}' %
                           (cp, name, definition))
         return STYLE_TEMPLATE % {'cp': self.commandprefix,
@@ -354,7 +353,7 @@ def format_unencoded(self, tokensource, outfile):
                 if self.texcomments:
                     # Try to guess comment starting lexeme and escape it ...
                     start = value[0:1]
-                    for i in xrange(1, len(value)):
+                    for i in range(1, len(value)):
                         if start[0] != value[i]:
                             break
                         start += value[i]

pygments/formatters/terminal.py

@@ -92,12 +92,6 @@ def __init__(self, **options):
         self._lineno = 0
 
     def format(self, tokensource, outfile):
-        # hack: if the output is a terminal and has an encoding set,
-        # use that to avoid unicode encode problems
-        if not self.encoding and hasattr(outfile, "encoding") and \
-           hasattr(outfile, "isatty") and outfile.isatty() and \
-           sys.version_info < (3,):
-            self.encoding = outfile.encoding
         return Formatter.format(self, tokensource, outfile)
 
     def _write_lineno(self, outfile):

pygments/formatters/terminal256.py

@@ -239,12 +239,6 @@ def _setup_styles(self):
                                              escape.reset_string())
 
     def format(self, tokensource, outfile):
-        # hack: if the output is a terminal and has an encoding set,
-        # use that to avoid unicode encode problems
-        if not self.encoding and hasattr(outfile, "encoding") and \
-           hasattr(outfile, "isatty") and outfile.isatty() and \
-           sys.version_info < (3,):
-            self.encoding = outfile.encoding
         return Formatter.format(self, tokensource, outfile)
 
     def format_unencoded(self, tokensource, outfile):