refactor: code quality improvements, bug fixes, and test coverage

.gitignore

@@ -4,4 +4,5 @@ __pycache__/
 results/
 tests/config.py
 .vscode/
-settings.local.json
+settings.local.json
+*.stackdump

git_log_analysis/__init__.py

@@ -0,0 +1,11 @@
+"""
+git_log_analysis package public API
+
+Re-exports the most commonly used classes so callers can import directly
+from the package root instead of from the internal mcp sub-package.
+"""
+
+from git_log_analysis.mcp.service import GitLogService
+from git_log_analysis.mcp.query_engine import QueryParams
+
+__all__ = ["GitLogService", "QueryParams"]

git_log_analysis/cli/projects_command.py

@@ -2,7 +2,7 @@
 
 import argparse
 
-from git_log_analysis.mcp import GitLogService
+from git_log_analysis import GitLogService
 
 from .types import SubParser
 
@@ -53,24 +53,20 @@ def _projects_command(args):
 
         print(f"Available projects: {len(projects)}\n")
 
-        # Calculate commit count for each project
-        all_commits = service.load_commits()
-
-        # Count commits per project
-        project_stats = {}
-        for commit in all_commits:
-            project_stats[commit.project] = project_stats.get(commit.project, 0) + 1
+        # Get commit count per project (no file loading required)
+        project_commit_counts = service.project_commit_counts
+        total_commits = sum(project_commit_counts.values())
 
         # Display results
         print(f"{'Project':<20} {'Commits':>10}")
         print("-" * 32)
 
         for project_name in sorted(projects):
-            commit_count = project_stats.get(project_name, 0)
+            commit_count = project_commit_counts.get(project_name, 0)
             print(f"{project_name:<20} {commit_count:>10}")
 
         print("-" * 32)
-        print(f"{'Total':<20} {len(all_commits):>10}")
+        print(f"{'Total':<20} {total_commits:>10}")
 
     except FileNotFoundError as e:
         print(f"❌ Error: {e}")

git_log_analysis/cli/search_command.py

@@ -2,7 +2,7 @@
 
 import argparse
 
-from git_log_analysis.mcp import GitLogService, QueryParams
+from git_log_analysis import GitLogService, QueryParams
 
 from .formatters import get_formatter
 from .types import SubParser
@@ -52,7 +52,10 @@ def setup_search_parser(subparsers: SubParser):
     # Search filters
     search_group = parser.add_argument_group("Search Filters")
     search_group.add_argument(
-        "-k", "--keyword", default="", help="Keyword to search for (searches in commit messages)"
+        "-k",
+        "--keyword",
+        default="",
+        help="Keyword to search for (searches in commit messages)",
     )
     search_group.add_argument(
         "-s", "--start-date", help="Start date (YYYY-MM-DD format). Example: 2024-01-01"
@@ -82,7 +85,9 @@ def setup_search_parser(subparsers: SubParser):
         help="Output format (default: text)",
     )
     output_group.add_argument(
-        "-o", "--output", help="Save results to a file (outputs to stdout if not specified)"
+        "-o",
+        "--output",
+        help="Save results to a file (outputs to stdout if not specified)",
     )
 
     parser.set_defaults(func=_search_command)

git_log_analysis/git_diff_parser.py

@@ -1,8 +1,8 @@
 import re
 import json
-from datetime import datetime
-from typing import List, Dict, Optional, Any
 from dataclasses import dataclass, asdict
+from typing import Any
+from git_log_analysis.mcp.utils import parse_date_to_naive
 
 
 @dataclass
@@ -23,19 +23,19 @@ class DiffChunk:
     old_count: int
     new_start: int
     new_count: int
-    lines: List[DiffLine]
+    lines: list[DiffLine]
 
 
 @dataclass
 class FileChange:
     """File change information"""
 
-    old_file: Optional[str]
-    new_file: Optional[str]
-    file_mode: Optional[str]
+    old_file: str | None
+    new_file: str | None
+    file_mode: str | None
     change_type: str  # 'modified', 'added', 'deleted', 'renamed'
-    index: Optional[str]
-    chunks: List[DiffChunk]
+    index: str | None
+    chunks: list[DiffChunk]
 
 
 @dataclass
@@ -47,13 +47,13 @@ class CommitData:
     email: str
     date: str
     message: str
-    changes: List[FileChange]
+    changes: list[FileChange]
 
 
 class DiffParser:
     def extract_diff_section(
-        self, lines: List[str], start_index: int
-    ) -> tuple[List[str], int]:
+        self, lines: list[str], start_index: int
+    ) -> tuple[list[str], int]:
         """
         Extracts a diff section.
 
@@ -82,7 +82,7 @@ def extract_diff_section(
 
         return diff_lines, i - start_index
 
-    def parse_diff(self, diff_lines: List[str]) -> Optional[FileChange]:
+    def parse_diff(self, diff_lines: list[str]) -> FileChange | None:
         """Parses a diff."""
         if not diff_lines:
             return None
@@ -118,12 +118,16 @@ def parse_diff(self, diff_lines: List[str]) -> Optional[FileChange]:
                 change.index = line.replace("index ", "").strip()
 
             elif line.startswith("--- "):
-                old_path = line.replace("--- ", "").replace("a/", "")
+                old_path = line[4:]
+                if old_path.startswith("a/"):
+                    old_path = old_path[2:]
                 if old_path != "/dev/null":
                     change.old_file = old_path
 
             elif line.startswith("+++ "):
-                new_path = line.replace("+++ ", "").replace("b/", "")
+                new_path = line[4:]
+                if new_path.startswith("b/"):
+                    new_path = new_path[2:]
                 if new_path != "/dev/null":
                     change.new_file = new_path
 
@@ -174,10 +178,10 @@ class GitLogParser:
     """Git Log Diff parser"""
 
     def __init__(self):
-        self.commits: List[CommitData] = []
+        self.commits: list[CommitData] = []
         self.diff_parser = DiffParser()
 
-    def parse_file(self, file_path: str, encoding: str = "utf-8") -> List[CommitData]:
+    def parse_file(self, file_path: str, encoding: str = "utf-8") -> list[CommitData]:
         """Parses Git log from a file."""
         try:
             with open(file_path, "r", encoding=encoding) as f:
@@ -186,7 +190,7 @@ def parse_file(self, file_path: str, encoding: str = "utf-8") -> List[CommitData
         except Exception as e:
             raise Exception(f"Error parsing file: {e}")
 
-    def parse(self, content: str) -> List[CommitData]:
+    def parse(self, content: str) -> list[CommitData]:
         """Parses Git log content."""
         # CRLF -> LF
         normalized_content = content.replace("\r\n", "\n")
@@ -202,7 +206,7 @@ def parse(self, content: str) -> List[CommitData]:
 
         return self.commits
 
-    def _parse_commit(self, commit_section: str) -> Optional[CommitData]:
+    def _parse_commit(self, commit_section: str) -> CommitData | None:
         """Parses an individual commit section."""
         lines = commit_section.split("\n")
 
@@ -255,7 +259,7 @@ def _parse_commit(self, commit_section: str) -> Optional[CommitData]:
             changes=commit_data["changes"],
         )
 
-    def get_stats(self) -> Dict[str, Any]:
+    def get_stats(self) -> dict[str, Any]:
         """Returns statistics for parsed commits."""
         if not self.commits:
             return {}
@@ -286,9 +290,7 @@ def get_stats(self) -> Dict[str, Any]:
         dates = []
         for commit in self.commits:
             try:
-                # Parse "2023-12-11 08:05:23 +0900" format
-                date_part = commit.date.split(" +")[0]  # Remove timezone
-                dt = datetime.strptime(date_part, "%Y-%m-%d %H:%M:%S")
+                dt = parse_date_to_naive(commit.date)
                 dates.append(dt)
             except Exception as e:
                 raise Exception(f"Error parsing date: {e}")

git_log_analysis/git_monthly_log_generator.py

@@ -1,5 +1,4 @@
 import argparse
-import os
 import subprocess
 from datetime import datetime, timedelta
 from pathlib import Path
@@ -16,7 +15,6 @@ def _month_range(start: datetime, end: datetime):
 
 
 def generate_git_logs(repo_path: Path, start_year: int, start_month: int):
-    os.chdir(repo_path)
     if not (repo_path / ".git").exists():
         raise FileNotFoundError(f"'{repo_path}' is not a Git repository.")
     if start_year <= 0:
@@ -52,7 +50,7 @@ def generate_git_logs(repo_path: Path, start_year: int, start_month: int):
         ]
 
         with open(filepath, "w", encoding="utf-8") as f:
-            subprocess.run(git_log_cmd, stdout=f, check=True)
+            subprocess.run(git_log_cmd, stdout=f, cwd=repo_path, check=True)
 
     print(f"Done! Monthly logs are saved in {output_dir}.")
 

git_log_analysis/mcp/data_loader.py

@@ -4,7 +4,6 @@
 import sys
 import io
 from pathlib import Path
-from typing import Optional, Dict
 from dataclasses import dataclass
 
 utf8_stderr = io.TextIOWrapper(
@@ -50,8 +49,6 @@ def __init__(self, results_dir: str):
         if not self.results_dir.exists():
             raise FileNotFoundError(f"Directory not found: {results_dir}")
 
-        self._projects: list[str]
-        self._summary_file_info_list: list[FileInfo]
         self._projects, self._summary_file_info_list = self._scan_projects()
 
     def _scan_projects(self) -> tuple[list[str], list[FileInfo]]:
@@ -75,12 +72,16 @@ def _scan_projects(self) -> tuple[list[str], list[FileInfo]]:
 
         projects: list[str] = []
         all_file_metadata: list[FileInfo] = []
+        self._project_commit_counts: dict[str, int] = {}
 
         for project_dir in sorted(subdirs):
             project_name = project_dir.name
-            file_metadata = self._scan_project_files(project_dir, project_name)
+            file_metadata, commit_count = self._scan_project_files(
+                project_dir, project_name
+            )
             all_file_metadata.extend(file_metadata)
             projects.append(project_name)
+            self._project_commit_counts[project_name] = commit_count
 
         logger.info(
             f"Scan complete: {len(all_file_metadata)} files found. Projects: {len(projects)}"
@@ -90,9 +91,10 @@ def _scan_projects(self) -> tuple[list[str], list[FileInfo]]:
 
     def _scan_project_files(
         self, project_dir: Path, project_name: str
-    ) -> list[FileInfo]:
+    ) -> tuple[list[FileInfo], int]:
         file_infos = []
         summary_files = sorted(project_dir.glob("*_summary.json"))
+        total_commit_count = 0
 
         for file_path in summary_files:
             date_match = re.search(r"(\d{4}-\d{2})", file_path.name)
@@ -102,12 +104,20 @@ def _scan_project_files(
                     path=file_path, project=project_name, year_month=year_month
                 )
                 file_infos.append(file_info)
+                try:
+                    with open(file_path, "r", encoding="utf-8") as f:
+                        commits_data = json.load(f)
+                    total_commit_count += len(commits_data)
+                except Exception as e:
+                    logger.warning(f"Cannot count commits in {file_path.name}: {e}")
             else:
                 logger.warning(f"Cannot parse date from file: {file_path.name}")
 
-        logger.info(f"[{project_name}] Summary: {len(summary_files)} files")
+        logger.info(
+            f"[{project_name}] Summary: {len(summary_files)} files, {total_commit_count} commits"
+        )
 
-        return file_infos
+        return file_infos, total_commit_count
 
     def _load_file_as_commits(self, file_path: Path, project: str) -> list[Commit]:
         try:
@@ -129,9 +139,9 @@ def _load_file_as_commits(self, file_path: Path, project: str) -> list[Commit]:
 
     def load_commits(
         self,
-        project: Optional[str] = None,
-        start_date: Optional[str] = None,
-        end_date: Optional[str] = None,
+        project: str | None = None,
+        start_date: str | None = None,
+        end_date: str | None = None,
     ) -> list[Commit]:
         """Loads required summary files and returns the list of commits
 
@@ -168,9 +178,9 @@ def load_commits(
 
     def _filter_summary_file_info_list(
         self,
-        project: Optional[str],
-        start_date: Optional[str],
-        end_date: Optional[str],
+        project: str | None,
+        start_date: str | None,
+        end_date: str | None,
     ) -> list[FileInfo]:
         filtered = []
 
@@ -191,7 +201,7 @@ def _filter_summary_file_info_list(
 
         return filtered
 
-    def _create_error_response(self, commit: Commit, error_msg: str) -> Dict:
+    def _create_error_response(self, commit: Commit, error_msg: str) -> dict:
         """Creates an error response dictionary"""
         return {
             "commit_metadata": commit,
@@ -200,8 +210,8 @@ def _create_error_response(self, commit: Commit, error_msg: str) -> Dict:
         }
 
     def load_commit_changes(
-        self, commit_hash: str, project: Optional[str] = None
-    ) -> Optional[Dict]:
+        self, commit_hash: str, project: str | None = None
+    ) -> dict | None:
         """Loads the changes for a specific commit by its hash.
 
         Returns:
@@ -220,7 +230,7 @@ def load_commit_changes(
         files_to_search = self._summary_file_info_list
         if project:
             files_to_search = [
-                f for f in files_to_search if f.project == project.lower()
+                f for f in files_to_search if f.project.lower() == project.lower()
             ]
 
         for file_info in files_to_search:
@@ -242,7 +252,9 @@ def load_commit_changes(
                         with open(changes_file, "r", encoding="utf-8") as f:
                             changes_map = json.load(f)
                     except Exception as e:
-                        logger.error(f"Failed to load changes file: {changes_file} - {e}")
+                        logger.error(
+                            f"Failed to load changes file: {changes_file} - {e}"
+                        )
                         return self._create_error_response(
                             commit, f"Error reading changes file: {e}"
                         )
@@ -257,3 +269,8 @@ def load_commit_changes(
     def projects(self) -> list[str]:
         """Returns the list of projects"""
         return self._projects or []
+
+    @property
+    def project_commit_counts(self) -> dict[str, int]:
+        """Returns the commit count per project (computed at scan time)"""
+        return self._project_commit_counts