Optimize handling of long lines for checkers like 'missing-final-newline' (#5925)

* Fix parsing of long lines when ``missing-final-newline`` is enabled

* Adapt fa31b6b6 to be backward-compatible

Fixes #5724

Also address comments from the PR PyCQA/pylint#5786

Co-authored-by: Daniël van Noord <13665637+DanielNoord@users.noreply.github.com>
Co-authored-by: Pierre Sassoulas <pierre.sassoulas@gmail.com>

6 files changed, 37 insertions, 11 deletions

diff --git a/CONTRIBUTORS.txt b/CONTRIBUTORS.txt
index 01c49f1b7..989f9297d 100644
--- a/CONTRIBUTORS.txt
+++ b/CONTRIBUTORS.txt
@@ -355,3 +355,4 @@ contributors:
 - Daniel Brookman
 - Téo Bouvard
 - Konrad Weihmann
+- Sergey B Kirpichev
diff --git a/ChangeLog b/ChangeLog
index bcebfa447..99ad7a237 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -201,6 +201,10 @@ Release date: TBA
 
   Closes #5569
 
+* Optimize parsing of long lines when ``missing-final-newline`` is enabled.
+
+  Closes #5724
+
 * Fix false positives for ``used-before-assignment`` from using named
   expressions in a ternary operator test and using that expression as
   a call argument.
diff --git a/doc/whatsnew/2.13.rst b/doc/whatsnew/2.13.rst
index 629798f42..d0696f1db 100644
--- a/doc/whatsnew/2.13.rst
+++ b/doc/whatsnew/2.13.rst
@@ -261,6 +261,10 @@ Other Changes
   Closes #5771
   Ref PyCQA/astroid#1382
 
+* Optimize parsing of long lines when ``missing-final-newline`` is enabled.
+
+  Closes #5724
+
 * Fix ``unnecessary_dict_index_lookup`` false positive when deleting a dictionary's entry.
 
   Closes #4716
diff --git a/pylint/utils/pragma_parser.py b/pylint/utils/pragma_parser.py
index 0bf25de7c..e09cba7c9 100644
--- a/pylint/utils/pragma_parser.py
+++ b/pylint/utils/pragma_parser.py
@@ -9,17 +9,18 @@ from typing import Generator, List, Optional
 # so that an option can be continued with the reasons
 # why it is active or disabled.
 OPTION_RGX = r"""
-    \s*                # Any number of whitespace
-    \#?                # One or zero hash
-    .*                 # Anything (as much as possible)
-    (\s*               # Beginning of first matched group and any number of whitespaces
-    \#                 # Beginning of comment
-    .*?                # Anything (as little as possible)
-    \bpylint:          # pylint word and column
-    \s*                # Any number of whitespaces
-    ([^;#\n]+))        # Anything except semicolon or hash or newline (it is the second matched group)
-                       # and end of the first matched group
-    [;#]{0,1}"""  # From 0 to 1 repetition of semicolon or hash
+    (?:^\s*\#.*|\s*|               # Comment line, or whitespaces,
+       \s*\#.*(?=\#.*?\bpylint:))  # or a beginning of an inline comment
+                                   # followed by "pylint:" pragma
+    (\#                            # Beginning of comment
+    .*?                            # Anything (as little as possible)
+    \bpylint:                      # pylint word and column
+    \s*                            # Any number of whitespaces
+    ([^;#\n]+))                    # Anything except semicolon or hash or
+                                   # newline (it is the second matched group)
+                                   # and end of the first matched group
+    [;#]{0,1}                      # From 0 to 1 repetition of semicolon or hash
+"""
 OPTION_PO = re.compile(OPTION_RGX, re.VERBOSE)
 
 
diff --git a/tests/checkers/unittest_refactoring.py b/tests/checkers/unittest_refactoring.py
index a2694200b..760eb2fe8 100644
--- a/tests/checkers/unittest_refactoring.py
+++ b/tests/checkers/unittest_refactoring.py
@@ -32,3 +32,18 @@ def test_process_tokens() -> None:
         with pytest.raises(SystemExit) as cm:
             Run([os.path.join(REGR_DATA, "very_long_line.py")], reporter=TextReporter())
         assert cm.value.code == 0
+
+
+@pytest.mark.skipif(not hasattr(signal, "setitimer"), reason="Assumes POSIX signals")
+def test_issue_5724() -> None:
+    """Regression test for parsing of pylint disable pragma's."""
+    with timeout(25.0):
+        with pytest.raises(SystemExit) as cm:
+            Run(
+                [
+                    os.path.join(REGR_DATA, "issue_5724.py"),
+                    "--enable=missing-final-newline",
+                ],
+                reporter=TextReporter(),
+            )
+        assert cm.value.code == 0
diff --git a/tests/regrtest_data/issue_5724.py b/tests/regrtest_data/issue_5724.py
new file mode 100644
index 000000000..dad184c43
--- /dev/null
+++ b/tests/regrtest_data/issue_5724.py
@@ -0,0 +1 @@
+a = "a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #a #"