gitflow-analytics 1.0.3__py3-none-any.whl → 1.3.6__py3-none-any.whl

gitflow_analytics/identity_llm/__init__.py

@@ -0,0 +1,6 @@
+"""LLM-based developer identity resolution and auto-aliasing."""
+
+from .analyzer import LLMIdentityAnalyzer
+from .models import DeveloperCluster, IdentityAnalysisResult
+
+__all__ = ["LLMIdentityAnalyzer", "IdentityAnalysisResult", "DeveloperCluster"]

gitflow_analytics/identity_llm/analysis_pass.py

@@ -0,0 +1,231 @@
+"""Analysis pass for auto-aliasing developer identities."""
+
+import logging
+import os
+from pathlib import Path
+from typing import Any, Optional
+
+import yaml
+
+from .analyzer import LLMIdentityAnalyzer
+from .models import IdentityAnalysisResult
+
+logger = logging.getLogger(__name__)
+
+
+class IdentityAnalysisPass:
+    """Performs an analysis pass to auto-alias developer identities."""
+
+    def __init__(self, config_path: Path):
+        """Initialize with configuration."""
+        self.config_path = config_path
+        self.config = self._load_config()
+
+    def _load_config(self) -> dict[str, Any]:
+        """Load configuration from file."""
+        with open(self.config_path) as f:
+            config = yaml.safe_load(f)
+
+        # Handle environment variables
+        config_dir = self.config_path.parent
+        env_file = config_dir / ".env"
+
+        if env_file.exists():
+            from dotenv import load_dotenv
+
+            load_dotenv(env_file)
+
+        return config
+
+    def run_analysis(
+        self,
+        commits: list[dict[str, Any]],
+        output_path: Optional[Path] = None,
+        apply_to_config: bool = False,
+    ) -> IdentityAnalysisResult:
+        """Run identity analysis pass on commits."""
+        logger.info("Starting identity analysis pass...")
+
+        # Get OpenRouter API key from config or env
+        api_key = None
+        if "qualitative" in self.config and self.config["qualitative"].get("enabled"):
+            api_key = self._resolve_env_var(
+                self.config["qualitative"].get("openrouter_api_key", "")
+            )
+
+        # Initialize analyzer
+        analyzer = LLMIdentityAnalyzer(
+            api_key=api_key,
+            model=self.config.get("qualitative", {}).get("model", "openai/gpt-4o-mini"),
+            confidence_threshold=self.config.get("analysis", {}).get("similarity_threshold", 0.85),
+        )
+
+        # Run analysis
+        result = analyzer.analyze_identities(commits)
+
+        logger.info(f"Analysis complete: {len(result.clusters)} identity clusters found")
+
+        # Save analysis report
+        if output_path:
+            self._save_analysis_report(result, output_path)
+
+        # Apply to configuration if requested
+        if apply_to_config:
+            self._apply_to_config(result)
+
+        return result
+
+    def _resolve_env_var(self, value: str) -> str:
+        """Resolve environment variable references."""
+        if value.startswith("${") and value.endswith("}"):
+            var_name = value[2:-1]
+            return os.getenv(var_name, value)
+        return value
+
+    def _save_analysis_report(self, result: IdentityAnalysisResult, output_path: Path):
+        """Save analysis report to file."""
+        report = {"analysis_metadata": result.analysis_metadata, "identity_clusters": []}
+
+        for cluster in result.clusters:
+            cluster_data = {
+                "canonical_name": cluster.canonical_name,
+                "canonical_email": cluster.canonical_email,
+                "confidence": cluster.confidence,
+                "reasoning": cluster.reasoning,
+                "total_commits": cluster.total_commits,
+                "aliases": [],
+            }
+
+            for alias in cluster.aliases:
+                cluster_data["aliases"].append(
+                    {
+                        "name": alias.name,
+                        "email": alias.email,
+                        "commit_count": alias.commit_count,
+                        "repositories": list(alias.repositories),
+                    }
+                )
+
+            report["identity_clusters"].append(cluster_data)
+
+        # Add unresolved identities
+        if result.unresolved_identities:
+            report["unresolved_identities"] = []
+            for identity in result.unresolved_identities:
+                report["unresolved_identities"].append(
+                    {
+                        "name": identity.name,
+                        "email": identity.email,
+                        "commit_count": identity.commit_count,
+                        "repositories": list(identity.repositories),
+                    }
+                )
+
+        # Write YAML report
+        with open(output_path, "w") as f:
+            yaml.dump(report, f, default_flow_style=False, sort_keys=False)
+
+        logger.info(f"Analysis report saved to: {output_path}")
+
+    def _apply_to_config(self, result: IdentityAnalysisResult):
+        """Apply analysis results to configuration file."""
+        # Get manual mappings from analysis
+        new_mappings = result.get_manual_mappings()
+
+        if not new_mappings:
+            logger.info("No new identity mappings to apply")
+            return
+
+        # Load current config
+        with open(self.config_path) as f:
+            config = yaml.safe_load(f)
+
+        # Ensure analysis section exists
+        if "analysis" not in config:
+            config["analysis"] = {}
+
+        # Get existing manual mappings
+        existing_mappings = config["analysis"].get("manual_identity_mappings", [])
+
+        # Merge new mappings
+        existing_emails = set()
+        for mapping in existing_mappings:
+            existing_emails.add(mapping.get("canonical_email", "").lower())
+
+        for new_mapping in new_mappings:
+            canonical_email = new_mapping["canonical_email"].lower()
+            if canonical_email not in existing_emails:
+                existing_mappings.append(new_mapping)
+                logger.info(f"Added identity mapping for: {canonical_email}")
+            else:
+                # Update existing mapping with new aliases
+                for existing in existing_mappings:
+                    if existing.get("canonical_email", "").lower() == canonical_email:
+                        existing_aliases = set(
+                            alias.lower() for alias in existing.get("aliases", [])
+                        )
+                        new_aliases = set(alias.lower() for alias in new_mapping["aliases"])
+                        combined_aliases = existing_aliases | new_aliases
+                        existing["aliases"] = list(combined_aliases)
+                        if new_aliases - existing_aliases:
+                            logger.info(f"Updated aliases for: {canonical_email}")
+                        break
+
+        # Update config
+        config["analysis"]["manual_identity_mappings"] = existing_mappings
+
+        # Write updated config
+        with open(self.config_path, "w") as f:
+            yaml.dump(config, f, default_flow_style=False, sort_keys=False)
+
+        logger.info(f"Updated configuration with {len(new_mappings)} identity mappings")
+
+    def generate_suggested_config(self, result: IdentityAnalysisResult) -> dict[str, Any]:
+        """Generate suggested configuration snippet for manual review."""
+        manual_mappings = result.get_manual_mappings()
+
+        # Also generate exclusions for bots
+        bot_patterns = [
+            r".*\[bot\]$",  # Matches "ewtn-version-bumper[bot]", "dependabot[bot]"
+            r".*-bot$",  # Matches names ending in "-bot"
+            r"^bot-.*",  # Matches names starting with "bot-"
+            r".*\sBot$",  # Matches "CNA Studio Bot", "GitHub Bot", etc.
+            r".*\sbot$",  # Matches "studio bot", "merge bot", etc.
+            r"^Bot\s.*",  # Matches "Bot User", "Bot Account", etc.
+            r"^bot\s.*",  # Matches "bot user", "bot account", etc.
+        ]
+
+        suggested_exclusions = []
+        logger.debug(
+            f"Bot detection: checking {len(result.unresolved_identities)} unresolved identities"
+        )
+
+        for identity in result.unresolved_identities:
+            logger.debug(
+                f"Bot detection: checking identity '{identity.name}' against {len(bot_patterns)} patterns"
+            )
+            for pattern in bot_patterns:
+                import re
+
+                if re.match(pattern, identity.name, re.IGNORECASE):
+                    logger.debug(
+                        f"Bot detection: MATCH - '{identity.name}' matches pattern '{pattern}'"
+                    )
+                    suggested_exclusions.append(identity.name)
+                    break
+            else:
+                logger.debug(
+                    f"Bot detection: NO MATCH - '{identity.name}' doesn't match any bot patterns"
+                )
+
+        if suggested_exclusions:
+            logger.debug(
+                f"Bot detection: found {len(suggested_exclusions)} bots to exclude: {suggested_exclusions}"
+            )
+        else:
+            logger.debug("Bot detection: no bots detected for exclusion")
+
+        return {
+            "analysis": {"manual_identity_mappings": manual_mappings},
+            "exclude": {"authors": suggested_exclusions} if suggested_exclusions else {},
+        }