Drop python 2 compatibility

.github/workflows/test-python.yml

@@ -29,11 +29,15 @@ jobs:
       matrix:
         os:
           - ubuntu-latest
-        python-version: [ '3.x', 'pypy-3.8', 'pypy-2.7' ]
-        # DISABLED: python-version: [ '3.x', '2.x' ]
-        # include:
-        #   - os: macos-latest
-        #     python-version: '3.x'
+          # - windows-latest
+          - macos-13
+        python-version:
+          - '3.12'
+          - '3.11'
+          - '3.10'
+          - '3.9'
+          - '3.8'
+          - 'pypy-3.8'
 
     steps:
       - uses: actions/checkout@v4
@@ -54,4 +58,4 @@ jobs:
 
       - name: run acceptance tests
         run: make acceptance
-        working-directory: python
+        working-directory: python

CHANGELOG.md

@@ -22,6 +22,9 @@ This document is formatted according to the principles of [Keep A CHANGELOG](htt
 - [.NET] Replace TinyJson to System.Text.Json
 - [.NET] Enable warnings as errors
 
+### Removed
+- [Python] Drop compatibility for python 2. Supported python versions are 3.8, 3.9, 3.10, 3.12
+
 ## [29.0.0] - 2024-08-12
 ### Added
 - (i18n) Added Gujarati translation for "Rule" ([#249](https://github.com/cucumber/gherkin/pull/249))

python/bin/gherkin

@@ -1,18 +1,13 @@
 #!/usr/bin/env sh
-# Use "make GHERKIN_PYTHON_VERSION=python2 ..." to use python2
 if [ -z "$GHERKIN_PYTHON_VERSION" ];
 then
-    if [ -x "$(command -v python)" ]
-    then
-        GHERKIN_PYTHON_VERSION=python
-    elif [ -x "$(command -v python3)" ]
+    if [ -x "$(command -v python3)" ]
     then
         GHERKIN_PYTHON_VERSION=python3
-    elif [ -x "$(command -v python2)" ]
+    elif [ -x "$(command -v python)" ]
     then
-        GHERKIN_PYTHON_VERSION=python2
-    else
-        echo "Neiter python, python3 or python2 found on PATH, exiting"
+        GHERKIN_PYTHON_VERSION=python
+        echo "Neither python3 or python found on PATH, exiting"
         exit 1
     fi
 fi

python/bin/gherkin-generate-tokens

@@ -1,18 +1,13 @@
 #!/usr/bin/env sh
-# Use "make GHERKIN_PYTHON_VERSION=python2 ..." to use python2
 if [ -z "$GHERKIN_PYTHON_VERSION" ];
 then
-    if [ -x "$(command -v python)" ]
-    then
-        GHERKIN_PYTHON_VERSION=python
-    elif [ -x "$(command -v python3)" ]
+    if [ -x "$(command -v python3)" ]
     then
         GHERKIN_PYTHON_VERSION=python3
-    elif [ -x "$(command -v python2)" ]
+    elif [ -x "$(command -v python)" ]
     then
-        GHERKIN_PYTHON_VERSION=python2
-    else
-        echo "Neiter python, python3 or python2 found on PATH, exiting"
+        GHERKIN_PYTHON_VERSION=python
+        echo "Neither python3 or python found on PATH, exiting"
         exit 1
     fi
 fi

python/bin/gherkin_generate_tokens.py

@@ -1,17 +1,11 @@
-import codecs
 import os
 import sys
-if sys.version_info < (3, 0):
-    import codecs
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.realpath(__file__))))
 from gherkin.token_scanner import TokenScanner
 from gherkin.token_formatter_builder import TokenFormatterBuilder
 from gherkin.parser import Parser
 
 files = sys.argv[1:]
-if sys.version_info < (3, 0) and os.name != 'nt':  # for Python2 unless on Windows native
-    UTF8Writer = codecs.getwriter('utf8')
-    sys.stdout = UTF8Writer(sys.stdout)
 parser = Parser(TokenFormatterBuilder())
 for file in files:
     scanner = TokenScanner(file)

python/gherkin-python.razor

@@ -41,24 +41,21 @@ RULE_TYPE = [
 ]
 
 
-class ParserContext(object):
+class ParserContext:
     def __init__(self, token_scanner, token_matcher, token_queue, errors):
         self.token_scanner = token_scanner
         self.token_matcher = token_matcher
         self.token_queue = token_queue
         self.errors = errors
 
 
-class @(Model.ParserClassName)(object):
+class @(Model.ParserClassName):
     def __init__(self, ast_builder=None):
         self.ast_builder = ast_builder if ast_builder is not None else AstBuilder()
         self.stop_at_first_error = False
 
     def parse(self, token_scanner_or_str, token_matcher=None):
-        if sys.version_info < (3, 0):
-            token_scanner = TokenScanner(token_scanner_or_str) if isinstance(token_scanner_or_str, basestring) else token_scanner_or_str
-        else:
-            token_scanner = TokenScanner(token_scanner_or_str) if isinstance(token_scanner_or_str, str) else token_scanner_or_str
+        token_scanner = TokenScanner(token_scanner_or_str) if isinstance(token_scanner_or_str, str) else token_scanner_or_str
         self.ast_builder.reset()
         if token_matcher is None:
             token_matcher = TokenMatcher()

python/gherkin/__main__.py

@@ -1,14 +1,6 @@
 import os
 from optparse import OptionParser
 import sys
-if sys.version_info < (3, 0):
-    string_type = basestring
-    if os.name != 'nt':
-        import codecs
-        UTF8Writer = codecs.getwriter('utf8')
-        sys.stdout = UTF8Writer(sys.stdout)
-else:
-    string_type = str
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.realpath(__file__))))
 import json

python/gherkin/ast_builder.py

@@ -2,7 +2,7 @@
 from .errors import AstBuilderException
 from .stream.id_generator import IdGenerator
 
-class AstBuilder(object):
+class AstBuilder:
     def __init__(self, id_generator=None):
         self.id_generator = id_generator
         if self.id_generator is None:

python/gherkin/ast_node.py

@@ -1,7 +1,7 @@
 from collections import defaultdict
 
 
-class AstNode(object):
+class AstNode:
 
     def __init__(self, rule_type):
         self.rule_type = rule_type

python/gherkin/dialect.py

@@ -7,11 +7,11 @@
     os.path.dirname(__file__),
     'gherkin-languages.json')
 
-with io.open(DIALECT_FILE_PATH, 'r', encoding='utf-8') as file:
+with open(DIALECT_FILE_PATH, encoding='utf-8') as file:
     DIALECTS = json.load(file)
 
 
-class Dialect(object):
+class Dialect:
 
     @classmethod
     def for_name(cls, name):

python/gherkin/errors.py

@@ -5,14 +5,14 @@ class ParserError(Exception):
 class ParserException(ParserError):
     def __init__(self, message, location):
         self.location = location
-        super(ParserException, self).__init__('(' + str(location['line']) + ':' +
+        super().__init__('(' + str(location['line']) + ':' +
                                               str(location['column'] if 'column' in
                                                   location else 0) + '): ' + message)
 
 
 class NoSuchLanguageException(ParserException):
     def __init__(self, language, location):
-        super(NoSuchLanguageException, self).__init__('Language not supported: ' + language,
+        super().__init__('Language not supported: ' + language,
                                                       location)
 
 
@@ -23,7 +23,7 @@ class AstBuilderException(ParserException):
 class UnexpectedEOFException(ParserException):
     def __init__(self, received_token, expected_token_types, state_comment):
         message = 'unexpected end of file, expected: ' + ', '.join(expected_token_types)
-        super(UnexpectedEOFException, self).__init__(message, received_token.location)
+        super().__init__(message, received_token.location)
 
 
 class UnexpectedTokenException(ParserException):
@@ -34,12 +34,12 @@ def __init__(self, received_token, expected_token_types, state_comment):
         location = (received_token.location if column else
                     {'line': received_token.location['line'],
                      'column': received_token.line.indent + 1})
-        super(UnexpectedTokenException, self).__init__(message, location)
+        super().__init__(message, location)
 
 
 class CompositeParserException(ParserError):
     def __init__(self, errors):
         self.errors = errors
-        super(CompositeParserException, self).__init__("Parser errors:\n" +
+        super().__init__("Parser errors:\n" +
                                                        '\n'.join([error.args[0] for error in
                                                                   errors]))

python/gherkin/gherkin_line.py

@@ -2,7 +2,7 @@
 from .errors import ParserException
 
 
-class GherkinLine(object):
+class GherkinLine:
     def __init__(self, line_text, line_number):
         self._line_text = line_text
         self._line_number = line_number

python/gherkin/inout.py

@@ -1,11 +1,10 @@
-from __future__ import print_function
 import json
 from .parser import Parser
 from .token_scanner import TokenScanner
 from .pickles.compiler import compile
 from .errors import ParserException, CompositeParserException
 
-class Inout(object):
+class Inout:
     def __init__(self, print_source, print_ast, print_pickles):
         self.print_source = print_source
         self.print_ast = print_ast

python/gherkin/parser.py

@@ -43,24 +43,21 @@
 ]
 
 
-class ParserContext(object):
+class ParserContext:
     def __init__(self, token_scanner, token_matcher, token_queue, errors):
         self.token_scanner = token_scanner
         self.token_matcher = token_matcher
         self.token_queue = token_queue
         self.errors = errors
 
 
-class Parser(object):
+class Parser:
     def __init__(self, ast_builder=None):
         self.ast_builder = ast_builder if ast_builder is not None else AstBuilder()
         self.stop_at_first_error = False
 
     def parse(self, token_scanner_or_str, token_matcher=None):
-        if sys.version_info < (3, 0):
-            token_scanner = TokenScanner(token_scanner_or_str) if isinstance(token_scanner_or_str, basestring) else token_scanner_or_str
-        else:
-            token_scanner = TokenScanner(token_scanner_or_str) if isinstance(token_scanner_or_str, str) else token_scanner_or_str
+        token_scanner = TokenScanner(token_scanner_or_str) if isinstance(token_scanner_or_str, str) else token_scanner_or_str
         self.ast_builder.reset()
         if token_matcher is None:
             token_matcher = TokenMatcher()

python/gherkin/pickles/compiler.py

@@ -1,10 +1,8 @@
 import re
-
-from ..count_symbols import count_symbols
 from ..stream.id_generator import IdGenerator
 
 
-class Compiler(object):
+class Compiler:
     def __init__(self, id_generator=None):
         self.id_generator = id_generator
         if self.id_generator is None:
@@ -158,7 +156,7 @@ def _interpolate(self, name, variable_cells, value_cells):
             # For the case of trailing backslash, re-escaping backslashes are needed
             reescaped_value = re.sub(r'\\', r'\\\\', value_cell['value'])
             name = re.sub(
-                u'<{0[value]}>'.format(variable_cell),
+                '<{0[value]}>'.format(variable_cell),
                 reescaped_value,
                 name
             )

python/gherkin/stream/gherkin_events.py

@@ -46,8 +46,6 @@ def enum(self, source_event):
                         'pickle': pickle
                     }
         except CompositeParserException as e:
-            for event in create_errors(e.errors, uri):
-                yield event
+            yield from create_errors(e.errors, uri)
         except ParserError as e:
-            for event in create_errors([e], uri):
-                yield event
+            yield from create_errors([e], uri)

python/gherkin/stream/id_generator.py

@@ -1,4 +1,4 @@
-class IdGenerator(object):
+class IdGenerator:
     def __init__(self):
         self._id_counter = 0
 

python/gherkin/stream/source_events.py

@@ -1,10 +1,8 @@
-import io
-
 def source_event(path):
     event = {
         'source': {
             'uri': path,
-            'data': io.open(path, 'r', encoding='utf8', newline='').read(),
+            'data': open(path, encoding='utf8', newline='').read(),
             'mediaType': 'text/x.cucumber.gherkin+plain'
         }
     }

python/gherkin/token.py

@@ -1,4 +1,4 @@
-class Token(object):
+class Token:
     def __init__(self, gherkin_line, location):
         self.line = gherkin_line
         self.location = location

python/gherkin/token_matcher.py

@@ -3,19 +3,8 @@
 from .dialect import Dialect
 from .errors import NoSuchLanguageException
 
-# Source: https://stackoverflow.com/a/8348914
-try:
-    import textwrap
-    textwrap.indent
-except AttributeError:  # undefined function (wasn't added until Python 3.3)
-    def indent(text, amount, ch=' '):
-        padding = amount * ch
-        return ''.join(padding+line for line in text.splitlines(True))
-else:
-    def indent(text, amount, ch=' '):
-        return textwrap.indent(text, amount * ch)
-
-class TokenMatcher(object):
+
+class TokenMatcher:
     LANGUAGE_RE = re.compile(r"^\s*#\s*language\s*:\s*([a-zA-Z\-_]+)\s*$")
 
     def __init__(self, dialect_name='en'):

python/gherkin/token_matcher_markdown.py

@@ -7,10 +7,10 @@
 class GherkinInMarkdownTokenMatcher(TokenMatcher):
 
     def __init__(self, dialect_name='en'):
-        super(GherkinInMarkdownTokenMatcher, self).__init__(dialect_name)
+        super().__init__(dialect_name)
 
     def reset(self):
-        super(GherkinInMarkdownTokenMatcher, self).reset()
+        super().reset()
         self.matched_feature_line=False
 
     def match_FeatureLine(self, token):
@@ -142,7 +142,7 @@ def _default_docstring_content_type():
     def _match_title_line(self, prefix, keywords, keywordSuffix, token, token_type):
 
         keywords_or_list="|".join(map(lambda x: re.escape(x), keywords))
-        match = re.search(u'{}({}){}(.*)'.format(prefix, keywords_or_list, keywordSuffix), token.line.get_line_text())
+        match = re.search(f'{prefix}({keywords_or_list}){keywordSuffix}(.*)', token.line.get_line_text())
         indent = token.line.indent
 
         if(match):