elspais 0.11.1__py3-none-any.whl → 0.43.5__py3-none-any.whl

elspais/commands/reformat_cmd.py

@@ -2,457 +2,54 @@
 """
 elspais.commands.reformat_cmd - Reformat requirements using AI.
 
-Transforms requirements from old format (Acceptance Criteria) to new format
-(labeled Assertions). Also provides line break normalization.
-
-REQ-int-d00008-A: Format transformation SHALL be available via
-                  `elspais reformat-with-claude`.
-REQ-int-d00008-B: The command SHALL support --dry-run, --backup, --start-req flags.
-REQ-int-d00008-C: Line break normalization SHALL be included.
+FUNCTIONALITY (to be reimplemented using Graph):
+- `elspais reformat-with-claude` - Transform requirements format using Claude AI
+  - Converts old "Acceptance Criteria" format to new "Assertions" format
+  - Uses elspais.reformat module for AI calls and content assembly
+
+OPTIONS:
+- --start-req REQ_ID: Start from specific requirement (traverse descendants)
+- --depth N: Maximum traversal depth
+- --dry-run: Preview changes without writing
+- --backup: Create .bak files before modifying
+- --force: Reformat even if already in new format
+- --fix-line-breaks: Normalize line breaks in output
+- --line-breaks-only: Only fix line breaks, no format conversion
+- --mode core|combined|local-only: Which repos to include
+
+WORKFLOW:
+1. Build requirement graph from spec directories
+2. Identify requirements needing reformat (via validation)
+3. Traverse from start_req or all PRD requirements
+4. For each requirement:
+   - Call Claude to generate new assertions format
+   - Validate the result
+   - Replace content in source file (with backup if requested)
+
+GRAPH INTEGRATION:
+- Receives TraceGraph from CLI dispatcher
+- Uses graph.find_by_id() to locate start requirement
+- Uses node.children for BFS traversal
+- Uses node.requirement.file_path for file modifications
+- Filters by NodeKind.REQUIREMENT
 """
 
 import argparse
-import shutil
-import sys
-from pathlib import Path
-from typing import List, Optional
-
-from elspais.config.loader import load_config, find_config_file, get_spec_directories
-from elspais.core.parser import RequirementParser
-from elspais.core.patterns import PatternValidator, PatternConfig
-from elspais.core.rules import RuleEngine, RulesConfig
 
 
 def run(args: argparse.Namespace) -> int:
     """Run the reformat-with-claude command.
 
-    This command reformats requirements from the old Acceptance Criteria format
-    to the new Assertions format using Claude AI.
+    This command requires reimplementation using the graph-based system.
     """
-    from elspais.reformat import (
-        get_all_requirements,
-        build_hierarchy,
-        traverse_top_down,
-        normalize_req_id,
-        reformat_requirement,
-        assemble_new_format,
-        validate_reformatted_content,
-        normalize_line_breaks,
-        fix_requirement_line_breaks,
-    )
-
-    print("elspais reformat-with-claude")
-    print()
-
-    # Handle line-breaks-only mode
-    if args.line_breaks_only:
-        return run_line_breaks_only(args)
-
-    # Configuration
-    start_req = args.start_req
-    max_depth = args.depth
-    dry_run = args.dry_run
-    backup = args.backup
-    force = args.force
-    fix_line_breaks = args.fix_line_breaks
-    verbose = getattr(args, 'verbose', False)
-    mode = getattr(args, 'mode', 'combined')
-
-    print(f"Options:")
-    print(f"  Start REQ:       {start_req or 'All PRD requirements'}")
-    print(f"  Max depth:       {max_depth or 'Unlimited'}")
-    print(f"  Mode:            {mode}")
-    print(f"  Dry run:         {dry_run}")
-    print(f"  Backup:          {backup}")
-    print(f"  Force reformat:  {force}")
-    print(f"  Fix line breaks: {fix_line_breaks}")
+    print("Error: 'reformat-with-claude' command not yet implemented with graph-based system")
     print()
-
-    if dry_run:
-        print("DRY RUN MODE - no changes will be made")
-        print()
-
-    # Create cached validator for ID normalization
-    config_path = find_config_file(Path.cwd())
-    config = load_config(config_path) if config_path else {}
-    pattern_config = PatternConfig.from_dict(config.get("patterns", {}))
-    validator = PatternValidator(pattern_config)
-
-    # Determine local base path for filtering (only modify local files)
-    local_base_path = config_path.parent if config_path else Path.cwd()
-
-    # Get all requirements (including cross-repo if mode allows)
-    print("Loading requirements...", end=" ", flush=True)
-    requirements = get_all_requirements(mode=mode)
-    if not requirements:
-        print("FAILED")
-        print("Error: Could not load requirements. Run 'elspais validate' first.",
-              file=sys.stderr)
-        return 1
-    print(f"found {len(requirements)} requirements")
-
-    # Build hierarchy
-    print("Building hierarchy...", end=" ", flush=True)
-    build_hierarchy(requirements)
-    print("done", flush=True)
-
-    # Determine which requirements to process
-    if start_req:
-        # Normalize and validate start requirement
-        print(f"Normalizing {start_req}...", end=" ", flush=True)
-        start_req = normalize_req_id(start_req, validator)
-        print(f"-> {start_req}", flush=True)
-        if start_req not in requirements:
-            print(f"Error: Requirement {start_req} not found", file=sys.stderr)
-            return 1
-
-        print(f"Traversing from {start_req}...", flush=True)
-        req_ids = traverse_top_down(requirements, start_req, max_depth)
-        print(f"Traversal complete", flush=True)
-    else:
-        # Process all PRD requirements first, then their descendants
-        prd_reqs = [
-            req_id for req_id, node in requirements.items()
-            if node.level.upper() == 'PRD'
-        ]
-        prd_reqs.sort()
-
-        print(f"Processing {len(prd_reqs)} PRD requirements and their descendants...")
-        req_ids = []
-        seen = set()
-        for prd_id in prd_reqs:
-            for req_id in traverse_top_down(requirements, prd_id, max_depth):
-                if req_id not in seen:
-                    req_ids.append(req_id)
-                    seen.add(req_id)
-
-    print(f"Found {len(req_ids)} requirements to process", flush=True)
-
-    # Run validation to identify requirements with acceptance_criteria issues
-    print("Running validation to identify old format...", end=" ", flush=True)
-    needs_reformat_ids = get_requirements_needing_reformat(config, local_base_path)
-    print(f"found {len(needs_reformat_ids)} with old format", flush=True)
-    print(flush=True)
-
-    # Filter to only requirements that need reformatting (unless --force)
-    if not force:
-        req_ids = [r for r in req_ids if r in needs_reformat_ids]
-        print(f"Filtered to {len(req_ids)} requirements needing reformat")
-        print(flush=True)
-
-    # Process each requirement
-    reformatted = 0
-    skipped = 0
-    errors = 0
-    line_break_fixes = 0
-
-    for i, req_id in enumerate(req_ids):
-        if i % 10 == 0 and i > 0:
-            print(f"Processing {i}/{len(req_ids)}...", flush=True)
-        node = requirements[req_id]
-
-        # Skip non-local files (from core/associated repos)
-        if not is_local_file(node.file_path, local_base_path):
-            skipped += 1
-            continue
-
-        print(f"[PROC] {req_id}: {node.title[:50]}...")
-
-        # Call Claude to reformat
-        result, success, error_msg = reformat_requirement(node, verbose=verbose)
-
-        if not success:
-            print(f"  ERROR: {error_msg}")
-            errors += 1
-            continue
-
-        # Validate the result
-        rationale = result.get('rationale', '')
-        assertions = result.get('assertions', [])
-
-        is_valid, warnings = validate_reformatted_content(node, rationale, assertions)
-
-        if warnings:
-            for warning in warnings:
-                print(f"  WARNING: {warning}")
-
-        if not is_valid:
-            print(f"  INVALID: Skipping due to validation errors")
-            errors += 1
-            continue
-
-        # Assemble the new format
-        new_content = assemble_new_format(
-            req_id=node.req_id,
-            title=node.title,
-            level=node.level,
-            status=node.status,
-            implements=node.implements,
-            rationale=rationale,
-            assertions=assertions
-        )
-
-        # Optionally normalize line breaks
-        if fix_line_breaks:
-            new_content = normalize_line_breaks(new_content)
-            line_break_fixes += 1
-
-        if dry_run:
-            print(f"  Would write to: {node.file_path}")
-            print(f"  Assertions: {len(assertions)}")
-            reformatted += 1
-        else:
-            # Write the reformatted content
-            try:
-                file_path = Path(node.file_path)
-
-                if backup:
-                    backup_path = file_path.with_suffix(file_path.suffix + '.bak')
-                    shutil.copy2(file_path, backup_path)
-                    print(f"  Backup: {backup_path}")
-
-                # Read the entire file
-                content = file_path.read_text()
-
-                # Find and replace this requirement's content
-                # The requirement starts with its header and ends before the next
-                # requirement or end of file
-                updated_content = replace_requirement_content(
-                    content, node.req_id, node.title, new_content
-                )
-
-                if updated_content:
-                    file_path.write_text(updated_content)
-                    print(f"  Written: {file_path}")
-                    reformatted += 1
-                else:
-                    print(f"  ERROR: Could not locate requirement in file")
-                    errors += 1
-
-            except Exception as e:
-                print(f"  ERROR: {e}")
-                errors += 1
-
-    # Summary
-    print()
-    print("=" * 60)
-    print(f"Summary:")
-    print(f"  Reformatted: {reformatted}")
-    print(f"  Skipped:     {skipped}")
-    print(f"  Errors:      {errors}")
-    if fix_line_breaks:
-        print(f"  Line breaks: {line_break_fixes} files normalized")
-
-    return 0 if errors == 0 else 1
-
-
-def replace_requirement_content(
-    file_content: str,
-    req_id: str,
-    title: str,
-    new_content: str
-) -> Optional[str]:
-    """
-    Replace a requirement's content in a file.
-
-    Finds the requirement by its header pattern and replaces everything
-    up to the footer line.
-
-    Args:
-        file_content: Full file content
-        req_id: Requirement ID (e.g., 'REQ-d00027')
-        title: Requirement title
-        new_content: New requirement content
-
-    Returns:
-        Updated file content, or None if requirement not found
-    """
-    import re
-
-    # Pattern to match the requirement header
-    # # REQ-d00027: Title
-    header_pattern = rf'^# {re.escape(req_id)}:\s*'
-
-    # Pattern to match the footer
-    # *End* *Title* | **Hash**: xxxxxxxx
-    footer_pattern = rf'^\*End\*\s+\*{re.escape(title)}\*\s+\|\s+\*\*Hash\*\*:\s*[a-fA-F0-9]+'
-
-    lines = file_content.split('\n')
-    result_lines = []
-    in_requirement = False
-    found = False
-
-    i = 0
-    while i < len(lines):
-        line = lines[i]
-
-        if not in_requirement:
-            # Check if this line starts the requirement
-            if re.match(header_pattern, line, re.IGNORECASE):
-                in_requirement = True
-                found = True
-                # Insert new content (without trailing newline, we'll add it)
-                new_lines = new_content.rstrip('\n').split('\n')
-                result_lines.extend(new_lines)
-                i += 1
-                continue
-            else:
-                result_lines.append(line)
-                i += 1
-        else:
-            # We're inside the requirement, skip until we find the footer
-            if re.match(footer_pattern, line, re.IGNORECASE):
-                # Found the footer, we've already added the new content
-                # with its own footer, so skip this old footer
-                in_requirement = False
-                i += 1
-                # Skip any trailing blank lines after the footer
-                while i < len(lines) and lines[i].strip() == '':
-                    i += 1
-            else:
-                # Skip this line (part of old requirement)
-                i += 1
-
-    if not found:
-        return None
-
-    return '\n'.join(result_lines)
-
-
-def run_line_breaks_only(args: argparse.Namespace) -> int:
-    """Run line break normalization only."""
-    from elspais.reformat import (
-        get_all_requirements,
-        normalize_line_breaks,
-        detect_line_break_issues,
-    )
-
-    dry_run = args.dry_run
-    backup = args.backup
-
-    print("Line break normalization mode")
-    print(f"  Dry run: {dry_run}")
-    print(f"  Backup:  {backup}")
-    print()
-
-    # Get all requirements
-    print("Loading requirements...", end=" ", flush=True)
-    requirements = get_all_requirements()
-    if not requirements:
-        print("FAILED")
-        print("Error: Could not load requirements.", file=sys.stderr)
-        return 1
-    print(f"found {len(requirements)} requirements")
-
-    # Group by file
-    files_to_process = {}
-    for req_id, node in requirements.items():
-        if node.file_path not in files_to_process:
-            files_to_process[node.file_path] = []
-        files_to_process[node.file_path].append(req_id)
-
-    print(f"Processing {len(files_to_process)} files...")
-    print()
-
-    fixed = 0
-    unchanged = 0
-    errors = 0
-
-    for file_path_str, req_ids in sorted(files_to_process.items()):
-        file_path = Path(file_path_str)
-
-        try:
-            content = file_path.read_text()
-            issues = detect_line_break_issues(content)
-
-            if not issues:
-                unchanged += 1
-                continue
-
-            print(f"[FIX] {file_path}")
-            for issue in issues:
-                print(f"  - {issue}")
-
-            if dry_run:
-                fixed += 1
-                continue
-
-            # Apply fixes
-            fixed_content = normalize_line_breaks(content)
-
-            if backup:
-                backup_path = file_path.with_suffix(file_path.suffix + '.bak')
-                shutil.copy2(file_path, backup_path)
-
-            file_path.write_text(fixed_content)
-            fixed += 1
-
-        except Exception as e:
-            print(f"[ERR] {file_path}: {e}")
-            errors += 1
-
-    print()
-    print("=" * 60)
-    print(f"Summary:")
-    print(f"  Fixed:     {fixed}")
-    print(f"  Unchanged: {unchanged}")
-    print(f"  Errors:    {errors}")
-
-    return 0 if errors == 0 else 1
-
-
-def get_requirements_needing_reformat(config: dict, base_path: Path) -> set:
-    """Run validation to identify requirements with old format.
-
-    Args:
-        config: Configuration dictionary
-        base_path: Base path of the local repository
-
-    Returns:
-        Set of requirement IDs that have format.acceptance_criteria violations
-    """
-    # Get local spec directories only
-    spec_dirs = get_spec_directories(None, config, base_path)
-    if not spec_dirs:
-        return set()
-
-    # Parse local requirements
-    pattern_config = PatternConfig.from_dict(config.get("patterns", {}))
-    spec_config = config.get("spec", {})
-    no_reference_values = spec_config.get("no_reference_values")
-    parser = RequirementParser(pattern_config, no_reference_values=no_reference_values)
-    skip_files = spec_config.get("skip_files", [])
-
-    try:
-        parse_result = parser.parse_directories(spec_dirs, skip_files=skip_files)
-        requirements = dict(parse_result)
-    except Exception:
-        return set()
-
-    # Run validation
-    rules_config = RulesConfig.from_dict(config.get("rules", {}))
-    engine = RuleEngine(rules_config)
-    violations = engine.validate(requirements)
-
-    # Filter to acceptance_criteria violations
-    return {
-        v.requirement_id for v in violations
-        if v.rule_name == "format.acceptance_criteria"
-    }
-
-
-def is_local_file(file_path: str, base_path: Path) -> bool:
-    """Check if file is in the local repo (not core/associated).
-
-    Args:
-        file_path: Path to the file (string)
-        base_path: Base path of the local repository
-
-    Returns:
-        True if file is within the local repo, False otherwise
-    """
-    try:
-        Path(file_path).resolve().relative_to(base_path.resolve())
-        return True
-    except ValueError:
-        return False
+    print("Planned features:")
+    print("  --start-req REQ_ID")
+    print("  --depth N")
+    print("  --dry-run")
+    print("  --backup")
+    print("  --force")
+    print("  --fix-line-breaks")
+    print("  --line-breaks-only")
+    return 1

elspais/commands/rules_cmd.py

@@ -9,8 +9,8 @@ import sys
 from pathlib import Path
 from typing import Optional
 
-from elspais.config.loader import find_config_file, load_config
-from elspais.core.content_rules import load_content_rules, load_content_rule
+from elspais.config import find_config_file, load_config
+from elspais.content_rules import load_content_rule, load_content_rules
 
 
 def run(args: argparse.Namespace) -> int:
@@ -56,7 +56,11 @@ def cmd_list(args: argparse.Namespace) -> int:
     print("Content Rules:")
     print("-" * 60)
     for rule in rules:
-        rel_path = rule.file_path.relative_to(base_path) if base_path in rule.file_path.parents else rule.file_path
+        rel_path = (
+            rule.file_path.relative_to(base_path)
+            if base_path in rule.file_path.parents
+            else rule.file_path
+        )
         print(f"  {rel_path}")
         print(f"    Title: {rule.title}")
         print(f"    Type: {rule.type}")