[symilar] Migrate from getopt to argparse (#9731)

Co-authored-by: Roger Sheu <[email protected]>

Author: Pierre-Sassoulas

doc/whatsnew/fragments/9731.user_action

@@ -0,0 +1,6 @@
+We migrated ``symilar`` to argparse, from getopt, so the error and help output changed
+(for the better). We exit with 2 instead of sometime 1, sometime 2. The error output
+is not captured by the runner anymore. It's not possible to use a value for the
+boolean options anymore (``--ignore-comments 1`` should become ``--ignore-comments``).
+
+Refs #9731

pylint/checkers/symilar.py

@@ -39,7 +39,6 @@
 import warnings
 from collections import defaultdict
 from collections.abc import Callable, Generator, Iterable, Sequence
-from getopt import GetoptError, getopt
 from io import BufferedIOBase, BufferedReader, BytesIO
 from itertools import chain
 from typing import TYPE_CHECKING, NamedTuple, NewType, NoReturn, TextIO, Union
@@ -750,15 +749,20 @@ class SimilaritiesChecker(BaseRawFileChecker, Symilar):
 
     name = "similarities"
     msgs = MSGS
-    # for available dict keys/values see the optik parser 'add_option' method
+    MIN_SIMILARITY_HELP = "Minimum lines number of a similarity."
+    IGNORE_COMMENTS_HELP = "Comments are removed from the similarity computation"
+    IGNORE_DOCSTRINGS_HELP = "Docstrings are removed from the similarity computation"
+    IGNORE_IMPORTS_HELP = "Imports are removed from the similarity computation"
+    IGNORE_SIGNATURES_HELP = "Signatures are removed from the similarity computation"
+    # for available dict keys/values see the option parser 'add_option' method
     options: Options = (
         (
             "min-similarity-lines",
             {
                 "default": DEFAULT_MIN_SIMILARITY_LINE,
                 "type": "int",
                 "metavar": "<int>",
-                "help": "Minimum lines number of a similarity.",
+                "help": MIN_SIMILARITY_HELP,
             },
         ),
         (
@@ -767,7 +771,7 @@ class SimilaritiesChecker(BaseRawFileChecker, Symilar):
                 "default": True,
                 "type": "yn",
                 "metavar": "<y or n>",
-                "help": "Comments are removed from the similarity computation",
+                "help": IGNORE_COMMENTS_HELP,
             },
         ),
         (
@@ -776,7 +780,7 @@ class SimilaritiesChecker(BaseRawFileChecker, Symilar):
                 "default": True,
                 "type": "yn",
                 "metavar": "<y or n>",
-                "help": "Docstrings are removed from the similarity computation",
+                "help": IGNORE_DOCSTRINGS_HELP,
             },
         ),
         (
@@ -785,7 +789,7 @@ class SimilaritiesChecker(BaseRawFileChecker, Symilar):
                 "default": True,
                 "type": "yn",
                 "metavar": "<y or n>",
-                "help": "Imports are removed from the similarity computation",
+                "help": IGNORE_IMPORTS_HELP,
             },
         ),
         (
@@ -794,7 +798,7 @@ class SimilaritiesChecker(BaseRawFileChecker, Symilar):
                 "default": True,
                 "type": "yn",
                 "metavar": "<y or n>",
-                "help": "Signatures are removed from the similarity computation",
+                "help": IGNORE_SIGNATURES_HELP,
             },
         ),
     )
@@ -876,67 +880,53 @@ def register(linter: PyLinter) -> None:
     linter.register_checker(SimilaritiesChecker(linter))
 
 
-def usage(status: int = 0) -> NoReturn:
-    """Display command line usage information."""
-    print("finds copy pasted blocks in a set of files")
-    print()
-    print(
-        "Usage: symilar [-d|--duplicates min_duplicated_lines] \
-[-i|--ignore-comments] [--ignore-docstrings] [--ignore-imports] [--ignore-signatures] file1..."
-    )
-    sys.exit(status)
-
-
 def Run(argv: Sequence[str] | None = None) -> NoReturn:
     """Standalone command line access point."""
-    if argv is None:
-        argv = sys.argv[1:]
-
-    s_opts = "hd:i:"
-    l_opts = [
-        "help",
-        "duplicates=",
-        "ignore-comments",
-        "ignore-imports",
-        "ignore-docstrings",
-        "ignore-signatures",
-    ]
-    min_lines = DEFAULT_MIN_SIMILARITY_LINE
-    ignore_comments = False
-    ignore_docstrings = False
-    ignore_imports = False
-    ignore_signatures = False
-    try:
-        opts, args = getopt(list(argv), s_opts, l_opts)
-    except GetoptError as e:
-        print(e)
-        usage(2)
-    for opt, val in opts:
-        if opt in {"-d", "--duplicates"}:
-            try:
-                min_lines = int(val)
-            except ValueError as e:
-                print(e)
-                usage(2)
-        elif opt in {"-h", "--help"}:
-            usage()
-        elif opt in {"-i", "--ignore-comments"}:
-            ignore_comments = True
-        elif opt in {"--ignore-docstrings"}:
-            ignore_docstrings = True
-        elif opt in {"--ignore-imports"}:
-            ignore_imports = True
-        elif opt in {"--ignore-signatures"}:
-            ignore_signatures = True
-    if not args:
-        usage(1)
-    sim = Symilar(
-        min_lines, ignore_comments, ignore_docstrings, ignore_imports, ignore_signatures
+    parser = argparse.ArgumentParser(
+        prog="symilar", description="Finds copy pasted blocks in a set of files."
+    )
+    parser.add_argument("files", nargs="+")
+    parser.add_argument(
+        "-d",
+        "--duplicates",
+        type=int,
+        default=DEFAULT_MIN_SIMILARITY_LINE,
+        help=SimilaritiesChecker.MIN_SIMILARITY_HELP,
+    )
+    parser.add_argument(
+        "-i",
+        "--ignore-comments",
+        action="store_true",
+        help=SimilaritiesChecker.IGNORE_COMMENTS_HELP,
+    )
+    parser.add_argument(
+        "--ignore-docstrings",
+        action="store_true",
+        help=SimilaritiesChecker.IGNORE_DOCSTRINGS_HELP,
+    )
+    parser.add_argument(
+        "--ignore-imports",
+        action="store_true",
+        help=SimilaritiesChecker.IGNORE_IMPORTS_HELP,
+    )
+    parser.add_argument(
+        "--ignore-signatures",
+        action="store_true",
+        help=SimilaritiesChecker.IGNORE_SIGNATURES_HELP,
+    )
+    parsed_args = parser.parse_args(args=argv)
+    similar_runner = Symilar(
+        min_lines=parsed_args.duplicates,
+        ignore_comments=parsed_args.ignore_comments,
+        ignore_docstrings=parsed_args.ignore_docstrings,
+        ignore_imports=parsed_args.ignore_imports,
+        ignore_signatures=parsed_args.ignore_signatures,
     )
-    for filename in args:
+    for filename in parsed_args.files:
         with open(filename, encoding="utf-8") as stream:
-            sim.append_stream(filename, stream)
-    sim.run()
+            similar_runner.append_stream(filename, stream)
+    similar_runner.run()
+    # the sys exit must be kept because of the unit tests that rely on it
     sys.exit(0)
 
 

tests/checkers/unittest_symilar.py

@@ -7,6 +7,7 @@
 from pathlib import Path
 
 import pytest
+from _pytest.capture import CaptureFixture
 
 from pylint.checkers import symilar
 from pylint.lint import PyLinter
@@ -364,13 +365,16 @@ def test_help() -> None:
             pytest.fail("not system exit")
 
 
-def test_no_args() -> None:
+def test_no_args(capsys: CaptureFixture) -> None:
     output = StringIO()
     with redirect_stdout(output):
         try:
             symilar.Run([])
         except SystemExit as ex:
-            assert ex.code == 1
+            assert ex.code == 2
+            out, err = capsys.readouterr()
+            assert not out
+            assert "the following arguments are required: files" in err
         else:
             pytest.fail("not system exit")
 
@@ -494,30 +498,30 @@ def test_set_duplicate_lines_to_zero() -> None:
     assert output.getvalue() == ""
 
 
-@pytest.mark.parametrize("v", ["d"])
-def test_bad_equal_short_form_option(v: str) -> None:
+def test_equal_short_form_option() -> None:
     """Regression test for https://github.com/pylint-dev/pylint/issues/9343"""
     output = StringIO()
     with redirect_stdout(output), pytest.raises(SystemExit) as ex:
-        symilar.Run([f"-{v}=0", SIMILAR1, SIMILAR2])
-    assert ex.value.code == 2
-    assert "invalid literal for int() with base 10: '=0'" in output.getvalue()
+        symilar.Run(["-d=2", SIMILAR1, SIMILAR2])
+    assert ex.value.code == 0
+    assert "similar lines in" in output.getvalue()
 
 
-@pytest.mark.parametrize("v", ["i", "d"])
-def test_space_short_form_option(v: str) -> None:
+def test_space_short_form_option() -> None:
     """Regression test for https://github.com/pylint-dev/pylint/issues/9343"""
     output = StringIO()
     with redirect_stdout(output), pytest.raises(SystemExit) as ex:
-        symilar.Run([f"-{v} 2", SIMILAR1, SIMILAR2])
+        symilar.Run(["-d 2", SIMILAR1, SIMILAR2])
     assert ex.value.code == 0
     assert "similar lines in" in output.getvalue()
 
 
-def test_bad_short_form_option() -> None:
+def test_bad_short_form_option(capsys: CaptureFixture) -> None:
     """Regression test for https://github.com/pylint-dev/pylint/issues/9343"""
     output = StringIO()
     with redirect_stdout(output), pytest.raises(SystemExit) as ex:
         symilar.Run(["-j=0", SIMILAR1, SIMILAR2])
+    out, err = capsys.readouterr()
     assert ex.value.code == 2
-    assert "option -j not recognized" in output.getvalue()
+    assert not out
+    assert "unrecognized arguments: -j=0" in err