CLOS-4334: Remove BOM file markers and introduce non-ASCII check action

.github/workflows/lint-cloudlinux.yml

@@ -0,0 +1,36 @@
+name: Lint (cloudlinux)
+
+# CloudLinux fork uses the `cloudlinux` branch as trunk; the inherited
+# upstream workflows (unit-tests.yml, codespell.yml, etc.) only fire on
+# `master` and so do not run on this fork. This narrow workflow guards
+# the cloudlinux branch against the most common regression we have hit:
+# stray non-ASCII bytes in Python source (em-dashes, smart quotes,
+# ellipsis from copy/paste, leftover BOMs).
+#
+# The check is just `make lint-non-ascii`, which delegates to
+# utils/check-non-ascii.py. That same target is also invoked as a
+# prerequisite of `make lint`, so local developers get the identical
+# rule when they run the full lint locally.
+#
+# Broader fixes (running the full upstream test+lint matrix on cloudlinux
+# PRs) are tracked separately.
+
+on:
+  pull_request:
+    branches: [cloudlinux]
+  push:
+    branches: [cloudlinux]
+
+permissions:
+  contents: read
+
+jobs:
+  lint-non-ascii:
+    name: Reject undeclared non-ASCII in Python source
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: make lint-non-ascii
+        run: make lint-non-ascii

Makefile

@@ -328,7 +328,17 @@ install-deps-fedora:
 		$(VENVNAME)/bin/pip install -I "git+https://github.com/oamg/leapp.git@refs/pull/$(REQ_LEAPP_PR)/head"; \
 	fi
 
-lint:
+# Reject undeclared non-ASCII bytes in Python source. Standalone target so
+# CI can call it without bringing up the full lint venv (the script is
+# pure stdlib python3) and so the rule is the single source of truth for
+# both `make lint` and the lint-cloudlinux GitHub Action.
+lint-non-ascii:
+	@echo "--- Checking for non-ASCII characters (Python 2.7 compat, PEP 263 aware) ---"
+	@SEARCH_PATH="$(TEST_PATHS)"; \
+	if [ -z "$${SEARCH_PATH}" ]; then echo "TEST_PATHS is empty; nothing to scan." >&2; exit 0; fi; \
+	python3 utils/check-non-ascii.py $${SEARCH_PATH}
+
+lint: lint-non-ascii
 	. $(VENVNAME)/bin/activate; \
 	echo "--- Linting ... ---" && \
 	SEARCH_PATH="$(TEST_PATHS)" && \

repos/system_upgrade/cloudlinux/actors/addcustomrepositories/libraries/addcustomrepositories.py

@@ -1,4 +1,4 @@
-import os
+import os
 import os.path
 import shutil
 import logging

repos/system_upgrade/cloudlinux/libraries/cl_repofileutils.py

@@ -1,4 +1,4 @@
-import os
+import os
 import os.path
 
 from leapp.libraries.stdlib import api

repos/system_upgrade/cloudlinux/libraries/detectcontrolpanel.py

@@ -1,4 +1,4 @@
-import os
+import os
 import os.path
 
 from leapp.libraries.stdlib import api

repos/system_upgrade/common/libraries/repomaputils.py

@@ -1,4 +1,4 @@
-import json
+import json
 from collections import defaultdict
 from leapp.models import PESIDRepositoryEntry, RepoMapEntry, RepositoriesMapping
 

utils/check-non-ascii.py

@@ -0,0 +1,91 @@
+#!/usr/bin/env python3
+"""Reject undeclared non-ASCII bytes in Python source.
+
+Python 2.7 source is ASCII-only unless the file declares its encoding via
+PEP 263 (a `# -*- coding: <name> -*-` or `# coding=<name>` comment on the
+first or second line). This script enforces that rule on the paths given
+as positional arguments: any *.py file that contains a byte > 0x7F and
+does NOT carry a PEP 263 declaration is reported and the script exits 1.
+
+Used both by `make lint-non-ascii` (and therefore `make lint`) and by
+the lint-cloudlinux GitHub Action so the two stay in sync.
+"""
+
+from __future__ import print_function
+
+import os
+import re
+import sys
+
+
+# PEP 263: an encoding declaration must appear on line 1 or 2 and match
+# the regex below. https://peps.python.org/pep-0263/
+_CODING_RE = re.compile(rb"^[ \t\f]*#.*?coding[=:][ \t]*([-_.a-zA-Z0-9]+)")
+
+
+def _file_has_encoding_declaration(data):
+    head = data.split(b"\n", 2)[:2]
+    return any(_CODING_RE.match(line) for line in head)
+
+
+def _scan_file(path):
+    """Return a list of (lineno, decoded_line) for lines with non-ASCII
+    bytes. Empty list means the file is clean OR has a PEP 263 declaration.
+    """
+    with open(path, "rb") as fp:
+        data = fp.read()
+    if not any(b > 0x7F for b in bytearray(data)):
+        return []
+    if _file_has_encoding_declaration(data):
+        return []
+    hits = []
+    for i, line in enumerate(data.splitlines(), start=1):
+        if any(b > 0x7F for b in bytearray(line)):
+            hits.append((i, line.decode("utf-8", "replace")))
+    return hits
+
+
+def _walk_paths(roots):
+    for root in roots:
+        if os.path.isfile(root):
+            if root.endswith(".py"):
+                yield root
+            continue
+        for dirpath, _dirs, files in os.walk(root):
+            for name in files:
+                if name.endswith(".py"):
+                    yield os.path.join(dirpath, name)
+
+
+def main(argv):
+    paths = argv[1:]
+    if not paths:
+        print("usage: {} <path> [path ...]".format(argv[0]), file=sys.stderr)
+        return 2
+    paths = [p for p in paths if os.path.exists(p)]
+    if not paths:
+        print("warning: no provided paths exist; nothing to scan", file=sys.stderr)
+        return 0
+
+    bad = 0
+    for path in _walk_paths(paths):
+        hits = _scan_file(path)
+        for lineno, text in hits:
+            print("{}:{}:{}".format(path, lineno, text))
+            bad += 1
+
+    if bad:
+        print(
+            "\nERROR: Non-ASCII bytes found in Python source without a PEP 263 "
+            "encoding declaration. Replace em-dashes (U+2014), smart quotes, "
+            "ellipsis, etc. with ASCII equivalents, or add "
+            "'# -*- coding: utf-8 -*-' on line 1 or 2 if the non-ASCII content "
+            "is intentional.",
+            file=sys.stderr,
+        )
+        return 1
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main(sys.argv))