thailint 0.9.0__py3-none-any.whl → 0.10.0__py3-none-any.whl

src/cli.py

@@ -2010,5 +2010,132 @@ def _execute_stateless_class_lint(  # pylint: disable=too-many-arguments,too-man
     sys.exit(1 if stateless_class_violations else 0)
 
 
+# Collection Pipeline command helper functions
+
+
+def _setup_pipeline_orchestrator(
+    path_objs: list[Path], config_file: str | None, verbose: bool, project_root: Path | None = None
+):
+    """Set up orchestrator for pipeline command."""
+    from src.orchestrator.core import Orchestrator
+    from src.utils.project_root import get_project_root
+
+    # Use provided project_root or fall back to auto-detection
+    if project_root is None:
+        first_path = path_objs[0] if path_objs else Path.cwd()
+        search_start = first_path if first_path.is_dir() else first_path.parent
+        project_root = get_project_root(search_start)
+
+    orchestrator = Orchestrator(project_root=project_root)
+
+    if config_file:
+        _load_config_file(orchestrator, config_file, verbose)
+
+    return orchestrator
+
+
+def _apply_pipeline_config_override(orchestrator, min_continues: int | None, verbose: bool):
+    """Apply min_continues override to orchestrator config."""
+    if min_continues is None:
+        return
+
+    if "collection_pipeline" not in orchestrator.config:
+        orchestrator.config["collection_pipeline"] = {}
+
+    orchestrator.config["collection_pipeline"]["min_continues"] = min_continues
+    if verbose:
+        logger.debug(f"Overriding min_continues to {min_continues}")
+
+
+def _run_pipeline_lint(orchestrator, path_objs: list[Path], recursive: bool):
+    """Execute collection-pipeline lint on files or directories."""
+    all_violations = _execute_linting_on_paths(orchestrator, path_objs, recursive)
+    return [v for v in all_violations if "collection-pipeline" in v.rule_id]
+
+
+@cli.command("pipeline")
+@click.argument("paths", nargs=-1, type=click.Path())
+@click.option("--config", "-c", "config_file", type=click.Path(), help="Path to config file")
+@format_option
+@click.option("--min-continues", type=int, help="Override min continue guards to flag (default: 1)")
+@click.option("--recursive/--no-recursive", default=True, help="Scan directories recursively")
+@click.pass_context
+def pipeline(  # pylint: disable=too-many-arguments,too-many-positional-arguments
+    ctx,
+    paths: tuple[str, ...],
+    config_file: str | None,
+    format: str,
+    min_continues: int | None,
+    recursive: bool,
+):
+    """Check for collection pipeline anti-patterns in code.
+
+    Detects for loops with embedded if/continue filtering patterns that could
+    be refactored to use collection pipelines (generator expressions, filter()).
+
+    PATHS: Files or directories to lint (defaults to current directory if none provided)
+
+    Examples:
+
+        \b
+        # Check current directory (all Python files recursively)
+        thai-lint pipeline
+
+        \b
+        # Check specific directory
+        thai-lint pipeline src/
+
+        \b
+        # Check single file
+        thai-lint pipeline src/app.py
+
+        \b
+        # Only flag loops with 2+ continue guards
+        thai-lint pipeline --min-continues 2 src/
+
+        \b
+        # Get JSON output
+        thai-lint pipeline --format json .
+
+        \b
+        # Get SARIF output for CI/CD integration
+        thai-lint pipeline --format sarif src/
+
+        \b
+        # Use custom config file
+        thai-lint pipeline --config .thailint.yaml src/
+    """
+    verbose = ctx.obj.get("verbose", False)
+    project_root = _get_project_root_from_context(ctx)
+
+    if not paths:
+        paths = (".",)
+
+    path_objs = [Path(p) for p in paths]
+
+    try:
+        _execute_pipeline_lint(
+            path_objs, config_file, format, min_continues, recursive, verbose, project_root
+        )
+    except Exception as e:
+        _handle_linting_error(e, verbose)
+
+
+def _execute_pipeline_lint(  # pylint: disable=too-many-arguments,too-many-positional-arguments
+    path_objs, config_file, format, min_continues, recursive, verbose, project_root=None
+):
+    """Execute collection-pipeline lint."""
+    _validate_paths_exist(path_objs)
+    orchestrator = _setup_pipeline_orchestrator(path_objs, config_file, verbose, project_root)
+    _apply_pipeline_config_override(orchestrator, min_continues, verbose)
+    pipeline_violations = _run_pipeline_lint(orchestrator, path_objs, recursive)
+
+    if verbose:
+        logger.info(f"Found {len(pipeline_violations)} collection-pipeline violation(s)")
+
+    format_violations(pipeline_violations, format)
+    sys.exit(1 if pipeline_violations else 0)
+
+
 if __name__ == "__main__":
     cli()

src/config.py

@@ -103,9 +103,8 @@ def _load_from_explicit_path(config_path: Path) -> dict[str, Any]:
 
 def _load_from_default_locations() -> dict[str, Any]:
     """Load config from default locations."""
-    for location in CONFIG_LOCATIONS:
-        if not location.exists():
-            continue
+    existing_locations = (loc for loc in CONFIG_LOCATIONS if loc.exists())
+    for location in existing_locations:
         loaded_config = _try_load_from_location(location)
         if loaded_config:
             return loaded_config

src/core/rule_discovery.py

@@ -20,6 +20,7 @@ Implementation: Package traversal with pkgutil, class introspection with inspect
 import importlib
 import inspect
 import pkgutil
+from types import ModuleType
 from typing import Any
 
 from .base import BaseLintRule
@@ -87,19 +88,51 @@ def _discover_from_module(module_path: str) -> list[BaseLintRule]:
     Returns:
         List of discovered rule instances
     """
+    module = _try_import_module(module_path)
+    if module is None:
+        return []
+    return _extract_rules_from_module(module)
+
+
+def _try_import_module(module_path: str) -> ModuleType | None:
+    """Try to import a module, returning None on failure.
+
+    Args:
+        module_path: Full module path to import
+
+    Returns:
+        Module object or None if import fails
+    """
     try:
-        module = importlib.import_module(module_path)
+        return importlib.import_module(module_path)
     except (ImportError, AttributeError):
-        return []
+        return None
 
-    rules = []
-    for _name, obj in inspect.getmembers(module):
-        if not _is_rule_class(obj):
-            continue
-        rule_instance = _try_instantiate_rule(obj)
-        if rule_instance:
-            rules.append(rule_instance)
-    return rules
+
+def _extract_rules_from_module(module: ModuleType) -> list[BaseLintRule]:
+    """Extract rule instances from a module.
+
+    Args:
+        module: Imported module to scan
+
+    Returns:
+        List of discovered rule instances
+    """
+    rule_classes = [obj for _name, obj in inspect.getmembers(module) if _is_rule_class(obj)]
+    return _instantiate_rules(rule_classes)
+
+
+def _instantiate_rules(rule_classes: list[type[BaseLintRule]]) -> list[BaseLintRule]:
+    """Instantiate a list of rule classes.
+
+    Args:
+        rule_classes: List of rule classes to instantiate
+
+    Returns:
+        List of successfully instantiated rules
+    """
+    instances = (_try_instantiate_rule(cls) for cls in rule_classes)
+    return [inst for inst in instances if inst is not None]
 
 
 def _try_instantiate_rule(rule_class: type[BaseLintRule]) -> BaseLintRule | None:

src/linters/collection_pipeline/__init__.py

@@ -0,0 +1,90 @@
+"""
+Purpose: Collection pipeline linter package initialization
+
+Scope: Exports for collection-pipeline linter module
+
+Overview: Initializes the collection-pipeline linter package and exposes the main rule class
+    for external use. Exports CollectionPipelineRule as the primary interface for the linter,
+    allowing the orchestrator to discover and instantiate the rule. Also exports configuration
+    and detector classes for advanced use cases. Provides a convenience lint() function for
+    direct usage without orchestrator setup. This module serves as the entry point for
+    the collection-pipeline linter functionality within the thai-lint framework.
+
+Dependencies: CollectionPipelineRule, CollectionPipelineConfig, PipelinePatternDetector
+
+Exports: CollectionPipelineRule (primary), CollectionPipelineConfig, PipelinePatternDetector, lint
+
+Interfaces: Standard Python package initialization with __all__ for explicit exports
+
+Implementation: Simple re-export pattern for package interface, convenience lint function
+"""
+
+from pathlib import Path
+from typing import Any
+
+from .config import DEFAULT_MIN_CONTINUES, CollectionPipelineConfig
+from .detector import PatternMatch, PipelinePatternDetector
+from .linter import CollectionPipelineRule
+
+__all__ = [
+    "CollectionPipelineRule",
+    "CollectionPipelineConfig",
+    "PipelinePatternDetector",
+    "PatternMatch",
+    "lint",
+]
+
+
+def lint(
+    path: Path | str,
+    config: dict[str, Any] | None = None,
+    min_continues: int = DEFAULT_MIN_CONTINUES,
+) -> list:
+    """Lint a file or directory for collection pipeline violations.
+
+    Args:
+        path: Path to file or directory to lint
+        config: Configuration dict (optional, uses defaults if not provided)
+        min_continues: Minimum if/continue patterns to flag (default: 1)
+
+    Returns:
+        List of violations found
+
+    Example:
+        >>> from src.linters.collection_pipeline import lint
+        >>> violations = lint('src/my_module.py', min_continues=2)
+        >>> for v in violations:
+        ...     print(f"{v.file_path}:{v.line} - {v.message}")
+    """
+    path_obj = Path(path) if isinstance(path, str) else path
+    project_root = path_obj if path_obj.is_dir() else path_obj.parent
+
+    orchestrator = _setup_pipeline_orchestrator(project_root, config, min_continues)
+    violations = _execute_pipeline_lint(orchestrator, path_obj)
+
+    return [v for v in violations if "collection-pipeline" in v.rule_id]
+
+
+def _setup_pipeline_orchestrator(
+    project_root: Path, config: dict[str, Any] | None, min_continues: int
+) -> Any:
+    """Set up orchestrator with collection-pipeline config."""
+    from src.orchestrator.core import Orchestrator
+
+    orchestrator = Orchestrator(project_root=project_root)
+
+    if config:
+        orchestrator.config["collection-pipeline"] = config
+    else:
+        orchestrator.config["collection-pipeline"] = {"min_continues": min_continues}
+
+    return orchestrator
+
+
+def _execute_pipeline_lint(orchestrator: Any, path_obj: Path) -> list:
+    """Execute linting on file or directory."""
+    if path_obj.is_file():
+        return orchestrator.lint_file(path_obj)
+    if path_obj.is_dir():
+        return orchestrator.lint_directory(path_obj)
+    return []

src/linters/collection_pipeline/config.py

@@ -0,0 +1,63 @@
+"""
+Purpose: Configuration dataclass for collection-pipeline linter
+
+Scope: Define configurable options for embedded filtering pattern detection
+
+Overview: Provides CollectionPipelineConfig for customizing linter behavior including
+    minimum number of continue patterns to flag, enable/disable toggle, and ignore
+    patterns. Integrates with the orchestrator's configuration system to allow users
+    to customize collection-pipeline detection via .thailint.yaml configuration files.
+    Follows the same configuration pattern as other thai-lint linters.
+
+Dependencies: dataclasses, typing
+
+Exports: CollectionPipelineConfig dataclass, DEFAULT_MIN_CONTINUES constant
+
+Interfaces: CollectionPipelineConfig.from_dict() class method for configuration loading
+
+Implementation: Dataclass with sensible defaults and config loading from dictionary
+"""
+
+from dataclasses import dataclass, field
+from typing import Any
+
+# Default threshold for minimum continue guards to flag
+DEFAULT_MIN_CONTINUES = 1
+
+
+@dataclass
+class CollectionPipelineConfig:
+    """Configuration for collection-pipeline linter."""
+
+    enabled: bool = True
+    """Whether the linter is enabled."""
+
+    min_continues: int = DEFAULT_MIN_CONTINUES
+    """Minimum number of if/continue patterns required to flag a violation."""
+
+    ignore: list[str] = field(default_factory=list)
+    """File patterns to ignore."""
+
+    def __post_init__(self) -> None:
+        """Validate configuration values."""
+        if self.min_continues < 1:
+            raise ValueError(f"min_continues must be at least 1, got {self.min_continues}")
+
+    @classmethod
+    def from_dict(
+        cls, config: dict[str, Any], language: str | None = None
+    ) -> "CollectionPipelineConfig":
+        """Load configuration from dictionary.
+
+        Args:
+            config: Dictionary containing configuration values
+            language: Programming language (unused, for interface compatibility)
+
+        Returns:
+            CollectionPipelineConfig instance with values from dictionary
+        """
+        return cls(
+            enabled=config.get("enabled", True),
+            min_continues=config.get("min_continues", DEFAULT_MIN_CONTINUES),
+            ignore=config.get("ignore", []),
+        )

src/linters/collection_pipeline/continue_analyzer.py

@@ -0,0 +1,100 @@
+"""
+Purpose: Analyze continue guard patterns in for loops
+
+Scope: Extract and validate if/continue patterns from loop bodies
+
+Overview: Provides helper functions for analyzing continue guard patterns in for loop
+    bodies. Handles extraction of sequential if/continue statements, validation of
+    simple continue-only patterns, and detection of side effects in conditions.
+    Separates pattern analysis logic from main detection for better maintainability.
+
+Dependencies: ast module for Python AST processing
+
+Exports: extract_continue_patterns, is_continue_only, has_side_effects, has_body_after_continues
+
+Interfaces: Functions for analyzing continue patterns in AST structures
+
+Implementation: AST-based pattern matching for continue guard identification
+"""
+
+import ast
+
+
+def extract_continue_patterns(body: list[ast.stmt]) -> list[ast.If]:
+    """Extract leading if statements that only contain continue.
+
+    Args:
+        body: List of statements in for loop body
+
+    Returns:
+        List of ast.If nodes that are continue guards
+    """
+    continues: list[ast.If] = []
+    for stmt in body:
+        if not isinstance(stmt, ast.If):
+            break
+        if not is_continue_only(stmt):
+            break
+        continues.append(stmt)
+    return continues
+
+
+def is_continue_only(if_node: ast.If) -> bool:
+    """Check if an if statement only contains continue.
+
+    Args:
+        if_node: AST If node to check
+
+    Returns:
+        True if the if statement is a simple continue guard
+    """
+    if len(if_node.body) != 1:
+        return False
+    if not isinstance(if_node.body[0], ast.Continue):
+        return False
+    if if_node.orelse:
+        return False
+    return True
+
+
+def has_side_effects(continues: list[ast.If]) -> bool:
+    """Check if any condition has side effects.
+
+    Args:
+        continues: List of continue guard if statements
+
+    Returns:
+        True if any condition has side effects (e.g., walrus operator)
+    """
+    for if_node in continues:
+        if _condition_has_side_effects(if_node.test):
+            return True
+    return False
+
+
+def _condition_has_side_effects(node: ast.expr) -> bool:
+    """Check if expression has side effects.
+
+    Args:
+        node: AST expression node to check
+
+    Returns:
+        True if expression has side effects
+    """
+    for child in ast.walk(node):
+        if isinstance(child, ast.NamedExpr):
+            return True
+    return False
+
+
+def has_body_after_continues(body: list[ast.stmt], num_continues: int) -> bool:
+    """Check if there are statements after continue guards.
+
+    Args:
+        body: List of statements in for loop body
+        num_continues: Number of continue guards detected
+
+    Returns:
+        True if there are statements after the continue guards
+    """
+    return len(body) > num_continues

src/linters/collection_pipeline/detector.py

@@ -0,0 +1,130 @@
+"""
+Purpose: AST-based detection of collection pipeline anti-patterns
+
+Scope: Pattern matching for for loops with embedded filtering via if/continue
+
+Overview: Implements the core detection logic for identifying imperative loop patterns
+    that use if/continue for filtering instead of collection pipelines. Uses Python's
+    AST module to analyze code structure and identify refactoring opportunities. Detects
+    patterns like 'for x in iter: if not cond: continue; action(x)' and suggests
+    refactoring to generator expressions or filter(). Handles edge cases like walrus
+    operators (side effects), else branches, and empty loop bodies.
+
+Dependencies: ast module, continue_analyzer, suggestion_builder
+
+Exports: PipelinePatternDetector class, PatternMatch dataclass
+
+Interfaces: PipelinePatternDetector.detect_patterns() -> list[PatternMatch]
+
+Implementation: AST visitor pattern with delegated pattern matching and suggestion generation
+"""
+
+import ast
+from dataclasses import dataclass
+
+from . import continue_analyzer, suggestion_builder
+
+
+@dataclass
+class PatternMatch:
+    """Represents a detected anti-pattern."""
+
+    line_number: int
+    """Line number where the for loop starts (1-indexed)."""
+
+    loop_var: str
+    """Name of the loop variable."""
+
+    iterable: str
+    """Source representation of the iterable."""
+
+    conditions: list[str]
+    """List of filter conditions (inverted from continue guards)."""
+
+    has_side_effects: bool
+    """Whether any condition has side effects."""
+
+    suggestion: str
+    """Refactoring suggestion as a code snippet."""
+
+
+class PipelinePatternDetector(ast.NodeVisitor):
+    """Detects for loops with embedded filtering via if/continue patterns."""
+
+    def __init__(self, source_code: str) -> None:
+        """Initialize detector with source code.
+
+        Args:
+            source_code: Python source code to analyze
+        """
+        self.source_code = source_code
+        self.matches: list[PatternMatch] = []
+
+    def detect_patterns(self) -> list[PatternMatch]:
+        """Analyze source code and return detected patterns.
+
+        Returns:
+            List of PatternMatch objects for each detected anti-pattern
+        """
+        try:
+            tree = ast.parse(self.source_code)
+            self.visit(tree)
+        except SyntaxError:
+            pass  # Invalid Python, return empty list
+        return self.matches
+
+    def visit_For(self, node: ast.For) -> None:  # pylint: disable=invalid-name
+        """Visit for loop and check for filtering patterns.
+
+        Args:
+            node: AST For node to analyze
+        """
+        match = self._analyze_for_loop(node)
+        if match is not None:
+            self.matches.append(match)
+        self.generic_visit(node)
+
+    def _analyze_for_loop(self, node: ast.For) -> PatternMatch | None:
+        """Analyze a for loop for embedded filtering patterns.
+
+        Args:
+            node: AST For node to analyze
+
+        Returns:
+            PatternMatch if pattern detected, None otherwise
+        """
+        continues = continue_analyzer.extract_continue_patterns(node.body)
+        if not continues:
+            return None
+
+        if continue_analyzer.has_side_effects(continues):
+            return None
+
+        if not continue_analyzer.has_body_after_continues(node.body, len(continues)):
+            return None
+
+        return self._create_match(node, continues)
+
+    def _create_match(self, for_node: ast.For, continues: list[ast.If]) -> PatternMatch:
+        """Create a PatternMatch from detected pattern.
+
+        Args:
+            for_node: AST For node
+            continues: List of continue guard if statements
+
+        Returns:
+            PatternMatch object with detection information
+        """
+        loop_var = suggestion_builder.get_target_name(for_node.target)
+        iterable = ast.unparse(for_node.iter)
+        conditions = [suggestion_builder.invert_condition(c.test) for c in continues]
+        suggestion = suggestion_builder.build_suggestion(loop_var, iterable, conditions)
+
+        return PatternMatch(
+            line_number=for_node.lineno,
+            loop_var=loop_var,
+            iterable=iterable,
+            conditions=conditions,
+            has_side_effects=False,
+            suggestion=suggestion,
+        )