sarj-python-lint 0.82.1__tar.gz → 0.82.3__tar.gz

{sarj_python_lint-0.82.1 → sarj_python_lint-0.82.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: sarj-python-lint
-Version: 0.82.1
+Version: 0.82.3
 Summary: Custom Python lint rules — AST-based, pre-commit-friendly, hypermodern defaults
 Project-URL: Homepage, https://code-standards.sarj.ai/rules/python/
 Project-URL: Documentation, https://code-standards.sarj.ai/rules/python/

{sarj_python_lint-0.82.1 → sarj_python_lint-0.82.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "sarj-python-lint"
-version = "0.82.1"
+version = "0.82.3"
 description = "Custom Python lint rules — AST-based, pre-commit-friendly, hypermodern defaults"
 readme = "README.md"
 authors = [{ name = "sarj-ai" }]

{sarj_python_lint-0.82.1 → sarj_python_lint-0.82.3}/src/sarj_python_lint/__main__.py

@@ -14,7 +14,7 @@ import typer
 from sarj_python_lint import __version__
 from sarj_python_lint._analysis_session import AnalysisSession
 from sarj_python_lint._filesystem import atomic_write_text
-from sarj_python_lint.rule_base import Diagnostic, ProjectRule, Severity, is_suppressed
+from sarj_python_lint.rule_base import Diagnostic, ProjectRule, Rule, Severity, is_suppressed
 from sarj_python_lint.rules import REGISTRY
 from sarj_python_lint.rules._paths import clear_path_caches
 from sarj_python_lint.rules._project_index import ProjectIndexSet
@@ -59,17 +59,23 @@ def _expand_paths(paths: list[Path]) -> list[Path]:
             except OSError:
                 pass
             continue
-        for child in p.rglob("*.py"):
-            if not child.is_file():
-                continue
-            if any(part in SKIP_DIR_NAMES for part in child.parts):
-                continue
-            try:
-                if child.stat().st_size > _MAX_FILE_BYTES:
-                    continue
-            except OSError:
+        out.extend(_python_files(p))
+    return out
+
+
+def _python_files(p: Path) -> list[Path]:
+    out: list[Path] = []
+    for child in p.rglob("*.py"):
+        if not child.is_file():
+            continue
+        if any(part in SKIP_DIR_NAMES for part in child.parts):
+            continue
+        try:
+            if child.stat().st_size > _MAX_FILE_BYTES:
                 continue
-            out.append(child)
+        except OSError:
+            continue
+        out.append(child)
     return out
 
 
@@ -98,22 +104,23 @@ def _check(rule_ids: list[str], paths: list[Path]) -> list[Diagnostic]:
             rule.prepare(indexes)
     diags: list[Diagnostic] = []
     for p, source in loaded.items():
-        source_lines = source.splitlines()
-        raw = [diagnostic for rule in rules for diagnostic in rule.check(p, source)]
-        diags.extend(
-            diagnostic
-            for diagnostic in deduplicate_diagnostics(
-                [
-                    diagnostic
-                    for diagnostic in raw
-                    if diagnostic.code == "SARJ419" or not is_suppressed(source_lines, diagnostic.line, diagnostic.code)
-                ],
-                source=source,
-            )
-        )
+        diags.extend(_check_source(rules, p, source))
     return diags
 
 
+def _check_source(rules: list[Rule], p: Path, source: str) -> list[Diagnostic]:
+    source_lines = source.splitlines()
+    raw = [diagnostic for rule in rules for diagnostic in rule.check(p, source)]
+    return deduplicate_diagnostics(
+        [
+            diagnostic
+            for diagnostic in raw
+            if diagnostic.code == "SARJ419" or not is_suppressed(source_lines, diagnostic.line, diagnostic.code)
+        ],
+        source=source,
+    )
+
+
 def analyze(
     rule_ids: list[str],
     paths: list[Path],
@@ -172,15 +179,7 @@ def deduplicate_diagnostics(diags: list[Diagnostic], *, source: str | None = Non
         location = owner_location(diagnostic)
         by_code = present.setdefault((diagnostic.path, location.line, location.column), {})
         by_code.setdefault(diagnostic.code, set()).add(diagnostic.severity)
-    suppressed = {
-        (location, generic, generic_severity)
-        for location, codes in present.items()
-        for specific, generics in _DIAGNOSTIC_PRECEDENCE.items()
-        if specific in codes
-        for generic in generics
-        for generic_severity in codes.get(generic, set())
-        if generic_severity is Severity.WARNING or Severity.ERROR in codes[specific]
-    }
+    suppressed = _suppressed_diagnostics(present)
     return [
         diagnostic
         for diagnostic in diags
@@ -199,6 +198,27 @@ def deduplicate_diagnostics(diags: list[Diagnostic], *, source: str | None = Non
     ]
 
 
+def _suppressed_diagnostics(
+    present: dict[tuple[Path, int, int], dict[str, set[Severity]]],
+) -> set[tuple[tuple[Path, int, int], str, Severity]]:
+    return {
+        (location, generic, severity)
+        for location, codes in present.items()
+        for generic, severity in _suppressed_codes(codes)
+    }
+
+
+def _suppressed_codes(codes: dict[str, set[Severity]]) -> set[tuple[str, Severity]]:
+    return {
+        (generic, generic_severity)
+        for specific, generics in _DIAGNOSTIC_PRECEDENCE.items()
+        if specific in codes
+        for generic in generics
+        for generic_severity in codes.get(generic, set())
+        if generic_severity is Severity.WARNING or Severity.ERROR in codes[specific]
+    }
+
+
 def _function_signature_owner_locations(source: str) -> dict[int, _OwnerLocation]:
     try:
         tree = ast.parse(source)

{sarj_python_lint-0.82.1 → sarj_python_lint-0.82.3}/src/sarj_python_lint/_python_target.py

@@ -53,6 +53,9 @@ class PythonTargetFacts:
                 if root not in self._installed:
                     self._installed[root] = _installed_declarations(root)
                 return self._installed[root].get(resolved)
+        return self._project_target(resolved)
+
+    def _project_target(self, resolved: Path) -> str | None:
         for ancestor in list(resolved.parents)[:_MAX_ANCESTORS]:
             try:
                 if any((ancestor / marker).exists() for marker in _PROJECT_BOUNDARIES):

{sarj_python_lint-0.82.1 → sarj_python_lint-0.82.3}/src/sarj_python_lint/_ratchet_cli.py

@@ -111,6 +111,10 @@ def _run(args: _Options) -> int:
             allow_increase=args.allow_increase or not baseline_path.exists(),
         )
 
+    return _report_gate(measurement, baseline, packages)
+
+
+def _report_gate(measurement: Measurement, baseline: Baseline, packages: list[str]) -> int:
     sys.stdout.write(
         f"{measurement.total} suppressions across {len(measurement.codes)} codes "
         f"in {len(packages)} package(s) (baseline total {sum(baseline.codes.values())})\n"

{sarj_python_lint-0.82.1 → sarj_python_lint-0.82.3}/src/sarj_python_lint/ratchet.py

@@ -323,29 +323,7 @@ def _count_comment(
     standalone: bool,
     ruff_aliases: Mapping[str, str] | None,
 ) -> None:
-    if standalone and (file_ignore := _RUFF_FILE_IGNORE_RE.match(comment)):
-        counts.update(
-            f"file-noqa:{_normalized_ruff_selector(code, ruff_aliases)}" for code in file_ignore.group(1).split(",")
-        )
-        return
-    if standalone and (file_noqa := _FILE_NOQA_RE.match(comment)):
-        listed = file_noqa.group(1)
-        if listed:
-            counts.update(f"file-noqa:{_normalized_ruff_selector(code, ruff_aliases)}" for code in listed.split(","))
-        else:
-            counts[_BLANKET_KEY] += 1
-        return
-    if standalone and _FLAKE8_FILE_NOQA_RE.match(comment):
-        counts[_FLAKE8_BLANKET_KEY] += 1
-        return
-    if standalone and (disabled := _RUFF_DISABLE_RE.match(comment)):
-        counts.update(
-            f"ruff-range:{_normalized_ruff_selector(code, ruff_aliases)}" for code in disabled.group(1).split(",")
-        )
-        return
-    if standalone and _FILE_PYRIGHT_RE.match(comment):
-        rules: list[str] = _FILE_PYRIGHT_RULE_RE.findall(comment)
-        counts.update(f"file-pyright:{rule}" for rule in rules)
+    if standalone and _count_file_comment(comment, counts, ruff_aliases=ruff_aliases):
         return
     for match in _NOQA_RE.finditer(comment):
         listed = match.group(1)
@@ -358,6 +336,10 @@ def _count_comment(
         counts.update(f"{prefix}{_normalized_ruff_selector(code, ruff_aliases)}" for code in match.group(1).split(","))
     for match in _SARJ_NOQA_RE.finditer(comment):
         counts.update(f"sarj-noqa:{code.strip().upper()}" for code in match.group(1).split(","))
+    _count_type_checker_comments(comment, counts)
+
+
+def _count_type_checker_comments(comment: str, counts: Counter[str]) -> None:
     for match in _PYRIGHT_IGNORE_RE.finditer(comment):
         counts.update(f"pyright:{code.strip()}" for code in match.group(1).split(","))
     if re.search(r"#\s*pyright:\s*ignore\b(?!\s*\[)", comment):
@@ -370,6 +352,34 @@ def _count_comment(
             counts[_BARE_TYPE_IGNORE_KEY] += 1
 
 
+def _count_file_comment(comment: str, counts: Counter[str], *, ruff_aliases: Mapping[str, str] | None) -> bool:
+    if file_ignore := _RUFF_FILE_IGNORE_RE.match(comment):
+        counts.update(
+            f"file-noqa:{_normalized_ruff_selector(code, ruff_aliases)}" for code in file_ignore.group(1).split(",")
+        )
+        return True
+    if file_noqa := _FILE_NOQA_RE.match(comment):
+        listed = file_noqa.group(1)
+        if listed:
+            counts.update(f"file-noqa:{_normalized_ruff_selector(code, ruff_aliases)}" for code in listed.split(","))
+        else:
+            counts[_BLANKET_KEY] += 1
+        return True
+    if _FLAKE8_FILE_NOQA_RE.match(comment):
+        counts[_FLAKE8_BLANKET_KEY] += 1
+        return True
+    if disabled := _RUFF_DISABLE_RE.match(comment):
+        counts.update(
+            f"ruff-range:{_normalized_ruff_selector(code, ruff_aliases)}" for code in disabled.group(1).split(",")
+        )
+        return True
+    if _FILE_PYRIGHT_RE.match(comment):
+        rules: list[str] = _FILE_PYRIGHT_RULE_RE.findall(comment)
+        counts.update(f"file-pyright:{rule}" for rule in rules)
+        return True
+    return False
+
+
 def _normalized_ruff_selector(selector: str, aliases: Mapping[str, str] | None) -> str:
     stripped = selector.strip()
     normalized = stripped.upper() if stripped.isupper() else stripped.lower()

{sarj_python_lint-0.82.1 → sarj_python_lint-0.82.3}/src/sarj_python_lint/rule_base.py

@@ -184,6 +184,9 @@ class RuleDocumentation:
         if len(example_ids) != len(set(example_ids)):
             msg = "rule example IDs must be unique"
             raise ValueError(msg)
+        self._validate_public_scenarios()
+
+    def _validate_public_scenarios(self) -> None:
         public_scenarios = {example.scenario for example in self.examples if example.public}
         for scenario in public_scenarios:
             pair = tuple(example for example in self.examples if example.public and example.scenario == scenario)

{sarj_python_lint-0.82.1 → sarj_python_lint-0.82.3}/src/sarj_python_lint/rules/_comments.py

@@ -457,32 +457,7 @@ def statement_comment_walls(
             statements = [item for item in value if _is_wall_statement(item)]
             if len(statements) < _WALL_MIN_STATEMENTS:
                 continue
-            attached: list[tuple[int, int, bool]] = []
-            for index, statement in enumerate(statements):
-                entry = comments.get(statement.lineno - 1)
-                if entry is None or entry[0] != statement.col_offset:
-                    continue
-                line = statement.lineno - 1
-                segment = ast.get_source_segment(source, statement) or ""
-                attached.append((index, line, _weak_walkthrough_comment(entry[1], segment)))
-            clusters: list[list[tuple[int, int, bool]]] = []
-            for item in attached:
-                if clusters and item[0] <= clusters[-1][-1][0] + 2:
-                    clusters[-1].append(item)
-                else:
-                    clusters.append([item])
-            for cluster in clusters:
-                span = cluster[-1][0] - cluster[0][0] + 1
-                weak = [line for _index, line, is_weak in cluster if is_weak]
-                if (
-                    span < _WALL_MIN_STATEMENTS
-                    or len(weak) < _WALL_MIN_COMMENTS
-                    or len(cluster) / span < _WALL_MIN_COMMENTED_RATIO
-                    or len(weak) / len(cluster) < _WALL_MIN_WEAK_RATIO
-                ):
-                    continue
-                leader = min(weak)
-                walls[leader] = frozenset(weak)
+            _collect_statement_walls(statements, comments, source, walls)
     return walls
 
 
@@ -513,6 +488,32 @@ def _is_wall_statement(node: ast.stmt) -> bool:
     )
 
 
+def _collect_statement_walls(
+    statements: list[ast.stmt], comments: dict[int, tuple[int, str]], source: str, walls: dict[int, frozenset[int]]
+) -> None:
+    attached: list[tuple[int, int, bool]] = []
+    for index, statement in enumerate(statements):
+        entry = comments.get(statement.lineno - 1)
+        if entry is None or entry[0] != statement.col_offset:
+            continue
+        line = statement.lineno - 1
+        segment = ast.get_source_segment(source, statement) or ""
+        attached.append((index, line, _weak_walkthrough_comment(entry[1], segment)))
+    clusters = _comment_clusters(attached)
+    for cluster in clusters:
+        span = cluster[-1][0] - cluster[0][0] + 1
+        weak = [line for _index, line, is_weak in cluster if is_weak]
+        if (
+            span < _WALL_MIN_STATEMENTS
+            or len(weak) < _WALL_MIN_COMMENTS
+            or len(cluster) / span < _WALL_MIN_COMMENTED_RATIO
+            or len(weak) / len(cluster) < _WALL_MIN_WEAK_RATIO
+        ):
+            continue
+        leader = min(weak)
+        walls[leader] = frozenset(weak)
+
+
 def _weak_walkthrough_comment(body: str, statement: str) -> bool:
     if (
         not body
@@ -533,3 +534,13 @@ def _weak_walkthrough_comment(body: str, statement: str) -> bool:
     matched = sum(1 for word in described if restates([word], known))
     novel = len(described) - matched
     return matched / len(described) >= _WALL_MIN_MATCHED_RATIO and novel <= _WALL_MAX_NOVEL_WORDS
+
+
+def _comment_clusters(attached: list[tuple[int, int, bool]]) -> list[list[tuple[int, int, bool]]]:
+    clusters: list[list[tuple[int, int, bool]]] = []
+    for item in attached:
+        if clusters and item[0] <= clusters[-1][-1][0] + 2:
+            clusters[-1].append(item)
+        else:
+            clusters.append([item])
+    return clusters