thailint 0.4.4__tar.gz → 0.4.5__tar.gz

{thailint-0.4.4 → thailint-0.4.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: thailint
-Version: 0.4.4
+Version: 0.4.5
 Summary: The AI Linter - Enterprise-grade linting and governance for AI-generated code across multiple languages
 License: MIT
 License-File: LICENSE

{thailint-0.4.4 → thailint-0.4.5}/pyproject.toml

@@ -17,7 +17,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "thailint"
-version = "0.4.4"
+version = "0.4.5"
 description = "The AI Linter - Enterprise-grade linting and governance for AI-generated code across multiple languages"
 authors = ["Steve Jackson"]
 license = "MIT"

{thailint-0.4.4 → thailint-0.4.5}/src/linters/dry/python_analyzer.py

@@ -65,8 +65,12 @@ class PythonDuplicateAnalyzer(BaseTokenAnalyzer):  # thailint: ignore[srp.violat
         # Performance optimization: Cache parsed AST to avoid re-parsing for each hash window
         self._cached_ast: ast.Module | None = None
         self._cached_content: str | None = None
+        # Performance optimization: Line-to-node index for O(1) lookups instead of O(n) ast.walk()
+        self._line_to_nodes: dict[int, list[ast.AST]] | None = None
 
-    def analyze(self, file_path: Path, content: str, config: DRYConfig) -> list[CodeBlock]:
+    def analyze(  # thailint: ignore[nesting.excessive-depth]
+        self, file_path: Path, content: str, config: DRYConfig
+    ) -> list[CodeBlock]:
         """Analyze Python file for duplicate code blocks, excluding docstrings.
 
         Args:
@@ -81,6 +85,9 @@ class PythonDuplicateAnalyzer(BaseTokenAnalyzer):  # thailint: ignore[srp.violat
         self._cached_ast = self._parse_content_safe(content)
         self._cached_content = content
 
+        # Performance optimization: Build line-to-node index for O(1) lookups
+        self._line_to_nodes = self._build_line_to_node_index(self._cached_ast)
+
         try:
             # Get docstring line ranges
             docstring_ranges = self._get_docstring_ranges_from_content(content)
@@ -89,34 +96,58 @@ class PythonDuplicateAnalyzer(BaseTokenAnalyzer):  # thailint: ignore[srp.violat
             lines_with_numbers = self._tokenize_with_line_numbers(content, docstring_ranges)
 
             # Generate rolling hash windows
-            windows = self._rolling_hash_with_tracking(lines_with_numbers, config.min_duplicate_lines)
-
-            blocks = []
-            for hash_val, start_line, end_line, snippet in windows:
-                # Skip blocks that are single logical statements
-                # Check the original source code, not the normalized snippet
-                if self._is_single_statement_in_source(content, start_line, end_line):
-                    continue
-
-                block = CodeBlock(
-                    file_path=file_path,
-                    start_line=start_line,
-                    end_line=end_line,
-                    snippet=snippet,
-                    hash_value=hash_val,
-                )
-
-                # Apply extensible filters (keyword arguments, imports, etc.)
-                if self._filter_registry.should_filter_block(block, content):
-                    continue
-
-                blocks.append(block)
+            windows = self._rolling_hash_with_tracking(
+                lines_with_numbers, config.min_duplicate_lines
+            )
 
-            return blocks
+            return self._filter_valid_blocks(windows, file_path, content)
         finally:
             # Clear cache after analysis to avoid memory leaks
             self._cached_ast = None
             self._cached_content = None
+            self._line_to_nodes = None
+
+    def _filter_valid_blocks(
+        self,
+        windows: list[tuple[int, int, int, str]],
+        file_path: Path,
+        content: str,
+    ) -> list[CodeBlock]:
+        """Filter hash windows and create valid CodeBlock instances."""
+        blocks = []
+        for hash_val, start_line, end_line, snippet in windows:
+            block = self._create_block_if_valid(
+                file_path, content, hash_val, start_line, end_line, snippet
+            )
+            if block:
+                blocks.append(block)
+        return blocks
+
+    def _create_block_if_valid(  # pylint: disable=too-many-arguments,too-many-positional-arguments
+        self,
+        file_path: Path,
+        content: str,
+        hash_val: int,
+        start_line: int,
+        end_line: int,
+        snippet: str,
+    ) -> CodeBlock | None:
+        """Create CodeBlock if it passes all validation checks."""
+        if self._is_single_statement_in_source(content, start_line, end_line):
+            return None
+
+        block = CodeBlock(
+            file_path=file_path,
+            start_line=start_line,
+            end_line=end_line,
+            snippet=snippet,
+            hash_value=hash_val,
+        )
+
+        if self._filter_registry.should_filter_block(block, content):
+            return None
+
+        return block
 
     def _get_docstring_ranges_from_content(self, content: str) -> set[int]:
         """Extract line numbers that are part of docstrings.
@@ -184,20 +215,21 @@ class PythonDuplicateAnalyzer(BaseTokenAnalyzer):  # thailint: ignore[srp.violat
             List of (original_line_number, normalized_code) tuples
         """
         lines_with_numbers = []
+        in_multiline_import = False
 
         for line_num, line in enumerate(content.split("\n"), start=1):
-            # Skip docstring lines
             if line_num in docstring_lines:
                 continue
 
-            # Use hasher's existing tokenization logic
-            line = self._hasher._strip_comments(line)  # pylint: disable=protected-access
-            line = " ".join(line.split())
-
+            line = self._hasher._normalize_line(line)  # pylint: disable=protected-access
             if not line:
                 continue
 
-            if self._hasher._is_import_statement(line):  # pylint: disable=protected-access
+            # Update multi-line import state and check if line should be skipped
+            in_multiline_import, should_skip = self._hasher._should_skip_import_line(  # pylint: disable=protected-access
+                line, in_multiline_import
+            )
+            if should_skip:
                 continue
 
             lines_with_numbers.append((line_num, line))
@@ -243,6 +275,7 @@ class PythonDuplicateAnalyzer(BaseTokenAnalyzer):  # thailint: ignore[srp.violat
         to avoid re-parsing the entire file for each hash window check.
         """
         # Use cached AST if available and content matches
+        tree: ast.Module | None
         if self._cached_ast is not None and content == self._cached_content:
             tree = self._cached_ast
         else:
@@ -261,23 +294,99 @@ class PythonDuplicateAnalyzer(BaseTokenAnalyzer):  # thailint: ignore[srp.violat
         except SyntaxError:
             return None
 
-    def _check_overlapping_nodes(self, tree: ast.Module, start_line: int, end_line: int) -> bool:
-        """Check if any AST node overlaps and matches single-statement pattern.
+    @staticmethod
+    def _build_line_to_node_index(tree: ast.Module | None) -> dict[int, list[ast.AST]] | None:
+        """Build an index mapping each line number to overlapping AST nodes.
+
+        Performance optimization: This allows O(1) lookups instead of O(n) ast.walk() calls.
+        For a file with 5,144 nodes and 673 hash windows, this reduces 3.46M node operations
+        to just ~3,365 relevant node checks (99.9% reduction).
 
-        Performance optimization: Pre-filter nodes by line range before expensive pattern checks.
+        Args:
+            tree: Parsed AST tree (None if parsing failed)
+
+        Returns:
+            Dictionary mapping line numbers to list of AST nodes overlapping that line,
+            or None if tree is None
         """
+        if tree is None:
+            return None
+
+        line_to_nodes: dict[int, list[ast.AST]] = {}
         for node in ast.walk(tree):
-            # Quick line range check to skip nodes that don't overlap
-            if not hasattr(node, "lineno") or not hasattr(node, "end_lineno"):
-                continue
-            if node.end_lineno < start_line or node.lineno > end_line:
-                continue  # No overlap, skip expensive pattern matching
+            if PythonDuplicateAnalyzer._node_has_line_info(node):
+                PythonDuplicateAnalyzer._add_node_to_index(node, line_to_nodes)
+
+        return line_to_nodes
 
-            # Node overlaps - check if it matches single-statement pattern
+    @staticmethod
+    def _node_has_line_info(node: ast.AST) -> bool:
+        """Check if node has valid line number information."""
+        if not hasattr(node, "lineno") or not hasattr(node, "end_lineno"):
+            return False
+        return node.lineno is not None and node.end_lineno is not None
+
+    @staticmethod
+    def _add_node_to_index(node: ast.AST, line_to_nodes: dict[int, list[ast.AST]]) -> None:
+        """Add node to all lines it overlaps in the index."""
+        for line_num in range(node.lineno, node.end_lineno + 1):  # type: ignore[attr-defined]
+            if line_num not in line_to_nodes:
+                line_to_nodes[line_num] = []
+            line_to_nodes[line_num].append(node)
+
+    def _check_overlapping_nodes(self, tree: ast.Module, start_line: int, end_line: int) -> bool:
+        """Check if any AST node overlaps and matches single-statement pattern.
+
+        Performance optimization: Use line-to-node index for O(1) lookups instead of O(n) ast.walk().
+        """
+        if self._line_to_nodes is not None:
+            return self._check_nodes_via_index(start_line, end_line)
+        return self._check_nodes_via_walk(tree, start_line, end_line)
+
+    def _check_nodes_via_index(self, start_line: int, end_line: int) -> bool:
+        """Check nodes using line-to-node index for O(1) lookups."""
+        candidates = self._collect_candidate_nodes_from_index(start_line, end_line)
+        return self._any_node_matches_pattern(candidates, start_line, end_line)
+
+    def _collect_candidate_nodes_from_index(self, start_line: int, end_line: int) -> set[ast.AST]:
+        """Collect unique nodes that overlap with the line range from index."""
+        candidate_nodes: set[ast.AST] = set()
+        for line_num in range(start_line, end_line + 1):
+            if self._line_to_nodes and line_num in self._line_to_nodes:
+                candidate_nodes.update(self._line_to_nodes[line_num])
+        return candidate_nodes
+
+    def _any_node_matches_pattern(
+        self, nodes: set[ast.AST], start_line: int, end_line: int
+    ) -> bool:
+        """Check if any node matches single-statement pattern."""
+        for node in nodes:
             if self._is_single_statement_pattern(node, start_line, end_line):
                 return True
         return False
 
+    def _check_nodes_via_walk(self, tree: ast.Module, start_line: int, end_line: int) -> bool:
+        """Check nodes using ast.walk() fallback for tests or standalone calls."""
+        for node in ast.walk(tree):
+            if self._node_matches_via_walk(node, start_line, end_line):
+                return True
+        return False
+
+    def _node_matches_via_walk(self, node: ast.AST, start_line: int, end_line: int) -> bool:
+        """Check if a single node overlaps and matches pattern."""
+        if not self._node_overlaps_range(node, start_line, end_line):
+            return False
+        return self._is_single_statement_pattern(node, start_line, end_line)
+
+    @staticmethod
+    def _node_overlaps_range(node: ast.AST, start_line: int, end_line: int) -> bool:
+        """Check if node overlaps with the given line range."""
+        if not hasattr(node, "lineno") or not hasattr(node, "end_lineno"):
+            return False
+        node_end = node.end_lineno
+        node_start = node.lineno
+        return not (node_end < start_line or node_start > end_line)
+
     def _node_overlaps_and_matches(self, node: ast.AST, start_line: int, end_line: int) -> bool:
         """Check if node overlaps with range and matches single-statement pattern."""
         if not hasattr(node, "lineno") or not hasattr(node, "end_lineno"):

{thailint-0.4.4 → thailint-0.4.5}/src/linters/dry/token_hasher.py

@@ -33,26 +33,80 @@ class TokenHasher:
             List of normalized code lines (non-empty, comments removed, imports filtered)
         """
         lines = []
+        in_multiline_import = False
 
         for line in code.split("\n"):
-            # Remove comments (language-specific logic can be added)
-            line = self._strip_comments(line)
-
-            # Normalize whitespace (collapse to single space)
-            line = " ".join(line.split())
-
-            # Skip empty lines
+            line = self._normalize_line(line)
             if not line:
                 continue
 
-            # Skip import statements (common false positive)
-            if self._is_import_statement(line):
+            # Update multi-line import state and check if line should be skipped
+            in_multiline_import, should_skip = self._should_skip_import_line(
+                line, in_multiline_import
+            )
+            if should_skip:
                 continue
 
             lines.append(line)
 
         return lines
 
+    def _normalize_line(self, line: str) -> str:
+        """Normalize a line by removing comments and excess whitespace.
+
+        Args:
+            line: Raw source code line
+
+        Returns:
+            Normalized line (empty string if line has no content)
+        """
+        line = self._strip_comments(line)
+        return " ".join(line.split())
+
+    def _should_skip_import_line(self, line: str, in_multiline_import: bool) -> tuple[bool, bool]:
+        """Determine if an import line should be skipped.
+
+        Args:
+            line: Normalized code line
+            in_multiline_import: Whether we're currently inside a multi-line import
+
+        Returns:
+            Tuple of (new_in_multiline_import_state, should_skip_line)
+        """
+        if self._is_multiline_import_start(line):
+            return True, True
+
+        if in_multiline_import:
+            return self._handle_multiline_import_continuation(line)
+
+        if self._is_import_statement(line):
+            return False, True
+
+        return False, False
+
+    def _is_multiline_import_start(self, line: str) -> bool:
+        """Check if line starts a multi-line import statement.
+
+        Args:
+            line: Normalized code line
+
+        Returns:
+            True if line starts a multi-line import (has opening paren but no closing)
+        """
+        return self._is_import_statement(line) and "(" in line and ")" not in line
+
+    def _handle_multiline_import_continuation(self, line: str) -> tuple[bool, bool]:
+        """Handle a line that's part of a multi-line import.
+
+        Args:
+            line: Normalized code line inside a multi-line import
+
+        Returns:
+            Tuple of (still_in_import, should_skip)
+        """
+        closes_import = ")" in line
+        return not closes_import, True
+
     def _strip_comments(self, line: str) -> str:
         """Remove comments from line (Python # and // style).
 

thailint-0.4.5/src/linters/file_header/__init__.py

@@ -0,0 +1,24 @@
+"""
+File: src/linters/file_header/__init__.py
+Purpose: File header linter module initialization
+Exports: FileHeaderRule
+Depends: linter.FileHeaderRule
+Implements: Module-level exports for clean API
+Related: linter.py for main rule implementation
+
+Overview:
+    Initializes the file header linter module providing multi-language file header
+    validation with mandatory field checking, atemporal language detection, and configuration
+    support. Main entry point for file header linting functionality.
+
+Usage:
+    from src.linters.file_header import FileHeaderRule
+    rule = FileHeaderRule()
+    violations = rule.check(context)
+
+Notes: Follows standard Python module initialization pattern with __all__ export control
+"""
+
+from .linter import FileHeaderRule
+
+__all__ = ["FileHeaderRule"]

thailint-0.4.5/src/linters/file_header/atemporal_detector.py

@@ -0,0 +1,87 @@
+"""
+File: src/linters/file_header/atemporal_detector.py
+Purpose: Detects temporal language patterns in file headers
+Exports: AtemporalDetector class
+Depends: re module for regex matching
+Implements: Regex-based pattern matching with configurable patterns
+Related: linter.py for detector usage, violation_builder.py for violation creation
+
+Overview:
+    Implements pattern-based detection of temporal language that violates atemporal
+    documentation requirements. Detects dates, temporal qualifiers, state change language,
+    and future references using regex patterns. Provides violation details for each pattern match.
+
+Usage:
+    detector = AtemporalDetector()
+    violations = detector.detect_violations(header_text)
+
+Notes: Four pattern categories - dates, temporal qualifiers, state changes, future references
+"""
+
+import re
+
+
+class AtemporalDetector:
+    """Detects temporal language patterns in text."""
+
+    # Date patterns
+    DATE_PATTERNS = [
+        (r"\d{4}-\d{2}-\d{2}", "ISO date format (YYYY-MM-DD)"),
+        (
+            r"(?:January|February|March|April|May|June|July|August|September|October|November|December)\s+\d{4}",
+            "Month Year format",
+        ),
+        (r"(?:Created|Updated|Modified):\s*\d{4}", "Date metadata"),
+    ]
+
+    # Temporal qualifiers
+    TEMPORAL_QUALIFIERS = [
+        (r"\bcurrently\b", 'temporal qualifier "currently"'),
+        (r"\bnow\b", 'temporal qualifier "now"'),
+        (r"\brecently\b", 'temporal qualifier "recently"'),
+        (r"\bsoon\b", 'temporal qualifier "soon"'),
+        (r"\bfor now\b", 'temporal qualifier "for now"'),
+    ]
+
+    # State change language
+    STATE_CHANGE = [
+        (r"\breplaces?\b", 'state change "replaces"'),
+        (r"\bmigrated from\b", 'state change "migrated from"'),
+        (r"\bformerly\b", 'state change "formerly"'),
+        (r"\bold implementation\b", 'state change "old"'),
+        (r"\bnew implementation\b", 'state change "new"'),
+    ]
+
+    # Future references
+    FUTURE_REFS = [
+        (r"\bwill be\b", 'future reference "will be"'),
+        (r"\bplanned\b", 'future reference "planned"'),
+        (r"\bto be added\b", 'future reference "to be added"'),
+        (r"\bcoming soon\b", 'future reference "coming soon"'),
+    ]
+
+    def detect_violations(  # thailint: ignore[nesting]
+        self, text: str
+    ) -> list[tuple[str, str, int]]:
+        """Detect all temporal language violations in text.
+
+        Args:
+            text: Text to check for temporal language
+
+        Returns:
+            List of (pattern, description, line_number) tuples for each violation
+        """
+        violations = []
+
+        # Check all pattern categories
+        all_patterns = (
+            self.DATE_PATTERNS + self.TEMPORAL_QUALIFIERS + self.STATE_CHANGE + self.FUTURE_REFS
+        )
+
+        lines = text.split("\n")
+        for line_num, line in enumerate(lines, start=1):
+            for pattern, description in all_patterns:
+                if re.search(pattern, line, re.IGNORECASE):
+                    violations.append((pattern, description, line_num))
+
+        return violations

thailint-0.4.5/src/linters/file_header/config.py

@@ -0,0 +1,66 @@
+"""
+File: src/linters/file_header/config.py
+Purpose: Configuration model for file header linter
+Exports: FileHeaderConfig dataclass
+Depends: dataclasses, pathlib
+Implements: Configuration with validation and defaults
+Related: linter.py for configuration usage
+
+Overview:
+    Defines configuration structure for file header linter including required fields
+    per language, ignore patterns, and validation options. Provides defaults matching
+    ai-doc-standard.md requirements and supports loading from .thailint.yaml configuration.
+
+Usage:
+    config = FileHeaderConfig()
+    config = FileHeaderConfig.from_dict(config_dict, "python")
+
+Notes: Dataclass with validation and language-specific defaults
+"""
+
+from dataclasses import dataclass, field
+
+
+@dataclass
+class FileHeaderConfig:
+    """Configuration for file header linting."""
+
+    # Required fields by language
+    required_fields_python: list[str] = field(
+        default_factory=lambda: [
+            "Purpose",
+            "Scope",
+            "Overview",
+            "Dependencies",
+            "Exports",
+            "Interfaces",
+            "Implementation",
+        ]
+    )
+
+    # Enforce atemporal language checking
+    enforce_atemporal: bool = True
+
+    # Patterns to ignore (file paths)
+    ignore: list[str] = field(
+        default_factory=lambda: ["test/**", "**/migrations/**", "**/__init__.py"]
+    )
+
+    @classmethod
+    def from_dict(cls, config_dict: dict, language: str) -> "FileHeaderConfig":
+        """Create config from dictionary.
+
+        Args:
+            config_dict: Dictionary of configuration values
+            language: Programming language for language-specific config
+
+        Returns:
+            FileHeaderConfig instance with values from dictionary
+        """
+        return cls(
+            required_fields_python=config_dict.get("required_fields", {}).get(
+                "python", cls().required_fields_python
+            ),
+            enforce_atemporal=config_dict.get("enforce_atemporal", True),
+            ignore=config_dict.get("ignore", cls().ignore),
+        )

thailint-0.4.5/src/linters/file_header/field_validator.py

@@ -0,0 +1,69 @@
+"""
+File: src/linters/file_header/field_validator.py
+Purpose: Validates mandatory fields in file headers
+Exports: FieldValidator class
+Depends: FileHeaderConfig for field requirements
+Implements: Configuration-driven validation with field presence checking
+Related: linter.py for validator usage, config.py for configuration
+
+Overview:
+    Validates presence and quality of mandatory header fields. Checks that all
+    required fields are present, non-empty, and meet minimum content requirements.
+    Supports language-specific required fields and provides detailed violation messages.
+
+Usage:
+    validator = FieldValidator(config)
+    violations = validator.validate_fields(fields, "python")
+
+Notes: Language-specific field requirements defined in config
+"""
+
+from .config import FileHeaderConfig
+
+
+class FieldValidator:
+    """Validates mandatory fields in headers."""
+
+    def __init__(self, config: FileHeaderConfig):
+        """Initialize validator with configuration.
+
+        Args:
+            config: File header configuration with required fields
+        """
+        self.config = config
+
+    def validate_fields(  # thailint: ignore[nesting]
+        self, fields: dict[str, str], language: str
+    ) -> list[tuple[str, str]]:
+        """Validate all required fields are present.
+
+        Args:
+            fields: Dictionary of parsed header fields
+            language: File language (python, typescript, etc.)
+
+        Returns:
+            List of (field_name, error_message) tuples for missing/invalid fields
+        """
+        violations = []
+        required_fields = self._get_required_fields(language)
+
+        for field_name in required_fields:
+            if field_name not in fields:
+                violations.append((field_name, f"Missing mandatory field: {field_name}"))
+            elif not fields[field_name] or len(fields[field_name].strip()) == 0:
+                violations.append((field_name, f"Empty mandatory field: {field_name}"))
+
+        return violations
+
+    def _get_required_fields(self, language: str) -> list[str]:
+        """Get required fields for language.
+
+        Args:
+            language: Programming language
+
+        Returns:
+            List of required field names for the language
+        """
+        if language == "python":
+            return self.config.required_fields_python
+        return []  # Other languages in PR5