greenmining 1.0.3__py3-none-any.whl → 1.0.4__py3-none-any.whl

greenmining/services/local_repo_analyzer.py

@@ -0,0 +1,387 @@
+# Local repository analyzer for direct GitHub URL analysis using PyDriller.
+
+from __future__ import annotations
+
+import os
+import re
+import shutil
+import tempfile
+from dataclasses import dataclass, field
+from datetime import datetime, timedelta
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Generator
+
+from pydriller import Repository
+from pydriller.metrics.process.change_set import ChangeSet
+from pydriller.metrics.process.code_churn import CodeChurn
+from pydriller.metrics.process.commits_count import CommitsCount
+from pydriller.metrics.process.contributors_count import ContributorsCount
+from pydriller.metrics.process.contributors_experience import ContributorsExperience
+from pydriller.metrics.process.history_complexity import HistoryComplexity
+from pydriller.metrics.process.hunks_count import HunksCount
+from pydriller.metrics.process.lines_count import LinesCount
+
+from greenmining.gsf_patterns import get_pattern_by_keywords, is_green_aware, GSF_PATTERNS
+from greenmining.utils import colored_print
+
+
+@dataclass
+class CommitAnalysis:
+    # Analysis result for a single commit.
+    
+    hash: str
+    message: str
+    author: str
+    author_email: str
+    date: datetime
+    green_aware: bool
+    gsf_patterns_matched: List[str]
+    pattern_count: int
+    pattern_details: List[Dict[str, Any]]
+    confidence: str
+    files_modified: List[str]
+    insertions: int
+    deletions: int
+    
+    # PyDriller DMM metrics
+    dmm_unit_size: Optional[float] = None
+    dmm_unit_complexity: Optional[float] = None
+    dmm_unit_interfacing: Optional[float] = None
+    
+    # Structural metrics (Lizard)
+    total_nloc: int = 0
+    total_complexity: int = 0
+    max_complexity: int = 0
+    methods_count: int = 0
+    
+    def to_dict(self) -> Dict[str, Any]:
+        # Convert to dictionary.
+        return {
+            "commit_hash": self.hash,
+            "message": self.message,
+            "author": self.author,
+            "author_email": self.author_email,
+            "date": self.date.isoformat() if self.date else None,
+            "green_aware": self.green_aware,
+            "gsf_patterns_matched": self.gsf_patterns_matched,
+            "pattern_count": self.pattern_count,
+            "pattern_details": self.pattern_details,
+            "confidence": self.confidence,
+            "files_modified": self.files_modified,
+            "insertions": self.insertions,
+            "deletions": self.deletions,
+            "dmm_unit_size": self.dmm_unit_size,
+            "dmm_unit_complexity": self.dmm_unit_complexity,
+            "dmm_unit_interfacing": self.dmm_unit_interfacing,
+            "total_nloc": self.total_nloc,
+            "total_complexity": self.total_complexity,
+            "max_complexity": self.max_complexity,
+            "methods_count": self.methods_count,
+        }
+
+
+@dataclass
+class RepositoryAnalysis:
+    # Complete analysis result for a repository.
+    
+    url: str
+    name: str
+    total_commits: int
+    green_commits: int
+    green_commit_rate: float
+    commits: List[CommitAnalysis] = field(default_factory=list)
+    process_metrics: Dict[str, Any] = field(default_factory=dict)
+    
+    def to_dict(self) -> Dict[str, Any]:
+        # Convert to dictionary.
+        return {
+            "url": self.url,
+            "name": self.name,
+            "total_commits": self.total_commits,
+            "green_commits": self.green_commits,
+            "green_commit_rate": self.green_commit_rate,
+            "commits": [c.to_dict() for c in self.commits],
+            "process_metrics": self.process_metrics,
+        }
+
+
+class LocalRepoAnalyzer:
+    # Analyze repositories directly from GitHub URLs using PyDriller.
+    
+    def __init__(
+        self,
+        clone_path: Optional[Path] = None,
+        max_commits: int = 500,
+        days_back: int = 730,
+        skip_merges: bool = True,
+        compute_process_metrics: bool = True,
+        cleanup_after: bool = True,
+    ):
+        # Initialize the local repository analyzer.
+        self.clone_path = clone_path or Path(tempfile.gettempdir()) / "greenmining_repos"
+        self.clone_path.mkdir(parents=True, exist_ok=True)
+        self.max_commits = max_commits
+        self.days_back = days_back
+        self.skip_merges = skip_merges
+        self.compute_process_metrics = compute_process_metrics
+        self.cleanup_after = cleanup_after
+        self.gsf_patterns = GSF_PATTERNS
+        
+    def _parse_repo_url(self, url: str) -> tuple[str, str]:
+        # Parse repository URL to extract owner and name.
+        # Handle HTTPS URLs
+        https_pattern = r"github\.com[/:]([^/]+)/([^/\.]+)"
+        match = re.search(https_pattern, url)
+        if match:
+            return match.group(1), match.group(2).replace(".git", "")
+        
+        # Handle SSH URLs
+        ssh_pattern = r"git@github\.com:([^/]+)/([^/\.]+)"
+        match = re.search(ssh_pattern, url)
+        if match:
+            return match.group(1), match.group(2).replace(".git", "")
+        
+        raise ValueError(f"Could not parse GitHub URL: {url}")
+    
+    def _get_pattern_details(self, matched_patterns: List[str]) -> List[Dict[str, Any]]:
+        # Get detailed pattern information.
+        details = []
+        for pattern_id, pattern in self.gsf_patterns.items():
+            if pattern["name"] in matched_patterns:
+                details.append({
+                    "name": pattern["name"],
+                    "category": pattern["category"],
+                    "description": pattern["description"],
+                    "sci_impact": pattern["sci_impact"],
+                })
+        return details
+    
+    def analyze_commit(self, commit) -> CommitAnalysis:
+        # Analyze a single PyDriller commit object.
+        message = commit.msg or ""
+        
+        # Green awareness check
+        green_aware = is_green_aware(message)
+        
+        # GSF pattern matching
+        matched_patterns = get_pattern_by_keywords(message)
+        pattern_details = self._get_pattern_details(matched_patterns)
+        
+        # Confidence calculation
+        pattern_count = len(matched_patterns)
+        confidence = "high" if pattern_count >= 2 else "medium" if pattern_count == 1 else "low"
+        
+        # File modifications
+        files_modified = [mod.filename for mod in commit.modified_files]
+        insertions = sum(mod.added_lines for mod in commit.modified_files)
+        deletions = sum(mod.deleted_lines for mod in commit.modified_files)
+        
+        # Delta Maintainability Model (if available)
+        dmm_unit_size = None
+        dmm_unit_complexity = None
+        dmm_unit_interfacing = None
+        
+        try:
+            dmm_unit_size = commit.dmm_unit_size
+            dmm_unit_complexity = commit.dmm_unit_complexity
+            dmm_unit_interfacing = commit.dmm_unit_interfacing
+        except Exception:
+            pass  # DMM may not be available for all commits
+        
+        # Structural metrics from Lizard (via PyDriller)
+        total_nloc = 0
+        total_complexity = 0
+        max_complexity = 0
+        methods_count = 0
+        
+        try:
+            for mod in commit.modified_files:
+                if mod.nloc:
+                    total_nloc += mod.nloc
+                if mod.complexity:
+                    total_complexity += mod.complexity
+                    if mod.complexity > max_complexity:
+                        max_complexity = mod.complexity
+                if mod.methods:
+                    methods_count += len(mod.methods)
+        except Exception:
+            pass  # Structural metrics may fail for some files
+        
+        return CommitAnalysis(
+            hash=commit.hash,
+            message=message,
+            author=commit.author.name,
+            author_email=commit.author.email,
+            date=commit.author_date,
+            green_aware=green_aware,
+            gsf_patterns_matched=matched_patterns,
+            pattern_count=pattern_count,
+            pattern_details=pattern_details,
+            confidence=confidence,
+            files_modified=files_modified,
+            insertions=insertions,
+            deletions=deletions,
+            dmm_unit_size=dmm_unit_size,
+            dmm_unit_complexity=dmm_unit_complexity,
+            dmm_unit_interfacing=dmm_unit_interfacing,
+            total_nloc=total_nloc,
+            total_complexity=total_complexity,
+            max_complexity=max_complexity,
+            methods_count=methods_count,
+        )
+    
+    def analyze_repository(self, url: str) -> RepositoryAnalysis:
+        # Analyze a repository from its URL.
+        owner, repo_name = self._parse_repo_url(url)
+        full_name = f"{owner}/{repo_name}"
+        
+        colored_print(f"\n Analyzing repository: {full_name}", "cyan")
+        
+        # Calculate date range
+        since_date = datetime.now() - timedelta(days=self.days_back)
+        
+        # Configure PyDriller Repository
+        repo_config = {
+            "path_to_repo": url,
+            "since": since_date,
+            "only_no_merge": self.skip_merges,
+        }
+        
+        # Clone to specific path if needed
+        local_path = self.clone_path / repo_name
+        if local_path.exists():
+            shutil.rmtree(local_path)
+        
+        repo_config["clone_repo_to"] = str(self.clone_path)
+        
+        colored_print(f"   Cloning to: {local_path}", "cyan")
+        
+        commits_analyzed = []
+        commit_count = 0
+        
+        try:
+            for commit in Repository(**repo_config).traverse_commits():
+                if commit_count >= self.max_commits:
+                    break
+                
+                try:
+                    analysis = self.analyze_commit(commit)
+                    commits_analyzed.append(analysis)
+                    commit_count += 1
+                    
+                    if commit_count % 50 == 0:
+                        colored_print(f"   Processed {commit_count} commits...", "cyan")
+                        
+                except Exception as e:
+                    colored_print(f"   Warning: Error analyzing commit {commit.hash[:8]}: {e}", "yellow")
+                    continue
+            
+            colored_print(f"    Analyzed {len(commits_analyzed)} commits", "green")
+            
+            # Compute process metrics if enabled
+            process_metrics = {}
+            if self.compute_process_metrics and local_path.exists():
+                colored_print("   Computing process metrics...", "cyan")
+                process_metrics = self._compute_process_metrics(str(local_path))
+            
+            # Calculate summary
+            green_commits = sum(1 for c in commits_analyzed if c.green_aware)
+            green_rate = green_commits / len(commits_analyzed) if commits_analyzed else 0
+            
+            result = RepositoryAnalysis(
+                url=url,
+                name=full_name,
+                total_commits=len(commits_analyzed),
+                green_commits=green_commits,
+                green_commit_rate=green_rate,
+                commits=commits_analyzed,
+                process_metrics=process_metrics,
+            )
+            
+            return result
+            
+        finally:
+            # Cleanup if requested
+            if self.cleanup_after and local_path.exists():
+                colored_print(f"   Cleaning up: {local_path}", "cyan")
+                shutil.rmtree(local_path, ignore_errors=True)
+    
+    def _compute_process_metrics(self, repo_path: str) -> Dict[str, Any]:
+        # Compute PyDriller process metrics for the repository.
+        metrics = {}
+        since_date = datetime.now() - timedelta(days=self.days_back)
+        to_date = datetime.now()
+        
+        try:
+            # ChangeSet metrics
+            cs = ChangeSet(repo_path, since=since_date, to=to_date)
+            metrics["change_set_max"] = cs.max()
+            metrics["change_set_avg"] = cs.avg()
+        except Exception as e:
+            colored_print(f"   Warning: ChangeSet metrics failed: {e}", "yellow")
+        
+        try:
+            # CodeChurn metrics
+            churn = CodeChurn(repo_path, since=since_date, to=to_date)
+            metrics["code_churn"] = churn.count()
+        except Exception as e:
+            colored_print(f"   Warning: CodeChurn metrics failed: {e}", "yellow")
+        
+        try:
+            # CommitsCount metrics
+            cc = CommitsCount(repo_path, since=since_date, to=to_date)
+            metrics["commits_per_file"] = cc.count()
+        except Exception as e:
+            colored_print(f"   Warning: CommitsCount metrics failed: {e}", "yellow")
+        
+        try:
+            # ContributorsCount metrics
+            contrib = ContributorsCount(repo_path, since=since_date, to=to_date)
+            metrics["contributors_per_file"] = contrib.count()
+        except Exception as e:
+            colored_print(f"   Warning: ContributorsCount metrics failed: {e}", "yellow")
+        
+        try:
+            # ContributorsExperience metrics
+            exp = ContributorsExperience(repo_path, since=since_date, to=to_date)
+            metrics["contributors_experience"] = exp.count()
+        except Exception as e:
+            colored_print(f"   Warning: ContributorsExperience metrics failed: {e}", "yellow")
+        
+        try:
+            # HistoryComplexity metrics
+            hc = HistoryComplexity(repo_path, since=since_date, to=to_date)
+            metrics["history_complexity"] = hc.count()
+        except Exception as e:
+            colored_print(f"   Warning: HistoryComplexity metrics failed: {e}", "yellow")
+        
+        try:
+            # HunksCount metrics
+            hunks = HunksCount(repo_path, since=since_date, to=to_date)
+            metrics["hunks_count"] = hunks.count()
+        except Exception as e:
+            colored_print(f"   Warning: HunksCount metrics failed: {e}", "yellow")
+        
+        try:
+            # LinesCount metrics
+            lines = LinesCount(repo_path, since=since_date, to=to_date)
+            metrics["lines_count"] = lines.count()
+        except Exception as e:
+            colored_print(f"   Warning: LinesCount metrics failed: {e}", "yellow")
+        
+        return metrics
+    
+    def analyze_repositories(self, urls: List[str]) -> List[RepositoryAnalysis]:
+        # Analyze multiple repositories from URLs.
+        results = []
+        
+        for i, url in enumerate(urls, 1):
+            colored_print(f"\n[{i}/{len(urls)}] Processing repository...", "cyan")
+            try:
+                result = self.analyze_repository(url)
+                results.append(result)
+            except Exception as e:
+                colored_print(f"   Error analyzing {url}: {e}", "red")
+                continue
+        
+        return results