mcp-code-indexer 1.5.0__py3-none-any.whl → 1.6.1__py3-none-any.whl

mcp_code_indexer/__init__.py

@@ -6,7 +6,7 @@ intelligent codebase navigation through searchable file descriptions,
 token-aware overviews, and advanced merge capabilities.
 """
 
-__version__ = "1.5.0"
+__version__ = "1.6.0"
 __author__ = "MCP Code Indexer Contributors"
 __email__ = ""
 __license__ = "MIT"

mcp_code_indexer/git_hook_handler.py

@@ -22,6 +22,7 @@ from tenacity import retry, wait_exponential, stop_after_attempt, retry_if_excep
 from .database.database import DatabaseManager
 from .database.models import Project, FileDescription
 from .error_handler import ValidationError
+from .token_counter import TokenCounter
 
 
 class GitHookError(Exception):
@@ -59,11 +60,12 @@ class GitHookHandler:
         self.db_manager = db_manager
         self.cache_dir = cache_dir
         self.logger = logging.getLogger(__name__)
+        self.token_counter = TokenCounter()
         
         # Git hook specific settings
         self.config = {
             "model": os.getenv("MCP_GITHOOK_MODEL", self.OPENROUTER_MODEL),
-            "max_diff_size": 100000,  # Skip if diff larger than this
+            "max_diff_tokens": 136000,  # Skip if diff larger than this (in tokens)
             "timeout": 30,
             "temperature": 0.3,  # Lower temperature for consistent updates
         }
@@ -73,22 +75,43 @@ class GitHookHandler:
         if not self.api_key:
             raise GitHookError("OPENROUTER_API_KEY environment variable is required for git hook mode")
     
-    async def run_githook_mode(self) -> None:
+    async def run_githook_mode(
+        self, 
+        commit_hash: Optional[str] = None, 
+        commit_range: Optional[Tuple[str, str]] = None
+    ) -> None:
         """
         Run in git hook mode - analyze changes and update descriptions.
         
+        Args:
+            commit_hash: Process a specific commit by hash
+            commit_range: Process commits in range (start_hash, end_hash)
+        
         This is the main entry point for git hook functionality.
         """
         try:
             # Get git info from current directory
             project_info = await self._identify_project_from_git()
             
-            # Get git diff and commit message
-            git_diff = await self._get_git_diff()
-            commit_message = await self._get_commit_message()
+            # Get git diff and commit message based on mode
+            if commit_hash:
+                git_diff = await self._get_git_diff_for_commit(commit_hash)
+                commit_message = await self._get_commit_message_for_commit(commit_hash)
+            elif commit_range:
+                git_diff = await self._get_git_diff_for_range(commit_range[0], commit_range[1])
+                commit_message = await self._get_commit_messages_for_range(commit_range[0], commit_range[1])
+            else:
+                git_diff = await self._get_git_diff()
+                commit_message = await self._get_commit_message()
             
-            if not git_diff or len(git_diff) > self.config["max_diff_size"]:
-                self.logger.info(f"Skipping git hook update - diff too large or empty")
+            if not git_diff:
+                self.logger.info(f"Skipping git hook update - no git diff")
+                return
+                
+            # Check token count instead of character count
+            diff_tokens = self.token_counter.count_tokens(git_diff)
+            if diff_tokens > self.config["max_diff_tokens"]:
+                self.logger.info(f"Skipping git hook update - diff too large ({diff_tokens} tokens > {self.config['max_diff_tokens']} limit)")
                 return
             
             # Fetch current state
@@ -224,6 +247,97 @@ class GitHookHandler:
         except subprocess.CalledProcessError:
             # If no commits exist yet, return empty string
             return ""
+
+    async def _get_git_diff_for_commit(self, commit_hash: str) -> str:
+        """
+        Get git diff for a specific commit.
+        
+        Args:
+            commit_hash: The commit hash to analyze
+            
+        Returns:
+            Git diff content as string
+        """
+        try:
+            # Get diff for the specific commit compared to its parent
+            diff_result = await self._run_git_command([
+                "diff", "--no-color", "--no-ext-diff", f"{commit_hash}~1..{commit_hash}"
+            ])
+            return diff_result
+            
+        except subprocess.CalledProcessError:
+            # If parent doesn't exist (first commit), diff against empty tree
+            try:
+                diff_result = await self._run_git_command([
+                    "diff", "--no-color", "--no-ext-diff", "4b825dc642cb6eb9a060e54bf8d69288fbee4904", commit_hash
+                ])
+                return diff_result
+            except subprocess.CalledProcessError as e:
+                raise GitHookError(f"Failed to get git diff for commit {commit_hash}: {e}")
+
+    async def _get_git_diff_for_range(self, start_hash: str, end_hash: str) -> str:
+        """
+        Get git diff for a range of commits.
+        
+        Args:
+            start_hash: Starting commit hash (exclusive)
+            end_hash: Ending commit hash (inclusive)
+            
+        Returns:
+            Git diff content as string
+        """
+        try:
+            diff_result = await self._run_git_command([
+                "diff", "--no-color", "--no-ext-diff", f"{start_hash}..{end_hash}"
+            ])
+            return diff_result
+        except subprocess.CalledProcessError as e:
+            raise GitHookError(f"Failed to get git diff for range {start_hash}..{end_hash}: {e}")
+
+    async def _get_commit_message_for_commit(self, commit_hash: str) -> str:
+        """
+        Get the commit message for a specific commit.
+        
+        Args:
+            commit_hash: The commit hash
+            
+        Returns:
+            Commit message as string
+        """
+        try:
+            message_result = await self._run_git_command([
+                "log", "-1", "--pretty=%B", commit_hash
+            ])
+            return message_result.strip()
+        except subprocess.CalledProcessError as e:
+            raise GitHookError(f"Failed to get commit message for {commit_hash}: {e}")
+
+    async def _get_commit_messages_for_range(self, start_hash: str, end_hash: str) -> str:
+        """
+        Get commit messages for a range of commits.
+        
+        Args:
+            start_hash: Starting commit hash (exclusive)
+            end_hash: Ending commit hash (inclusive)
+            
+        Returns:
+            Combined commit messages as string
+        """
+        try:
+            # Get all commit messages in the range
+            message_result = await self._run_git_command([
+                "log", "--pretty=%B", f"{start_hash}..{end_hash}"
+            ])
+            
+            # Clean up and format the messages
+            messages = message_result.strip()
+            if messages:
+                return f"Combined commit messages for range {start_hash}..{end_hash}:\n\n{messages}"
+            else:
+                return f"No commits found in range {start_hash}..{end_hash}"
+                
+        except subprocess.CalledProcessError as e:
+            raise GitHookError(f"Failed to get commit messages for range {start_hash}..{end_hash}: {e}")
     
     def _extract_changed_files(self, git_diff: str) -> List[str]:
         """

mcp_code_indexer/main.py

@@ -82,8 +82,11 @@ def parse_arguments() -> argparse.Namespace:
     
     parser.add_argument(
         "--githook",
-        action="store_true",
-        help="Git hook mode: auto-update descriptions based on git diff using OpenRouter API"
+        nargs="*",
+        metavar="COMMIT_HASH",
+        help="Git hook mode: auto-update descriptions based on git diff using OpenRouter API. "
+             "Usage: --githook (current changes), --githook HASH (specific commit), "
+             "--githook HASH1 HASH2 (commit range from HASH1 to HASH2)"
     )
     
     parser.add_argument(
@@ -462,12 +465,16 @@ async def handle_githook(args: argparse.Namespace) -> None:
         from .database.database import DatabaseManager
         from .git_hook_handler import GitHookHandler
         
+        # Process commit hash arguments
+        commit_hashes = args.githook if args.githook else []
+        
         logger.info("Starting git hook execution", extra={
             "structured_data": {
                 "args": {
                     "db_path": str(args.db_path),
                     "cache_dir": str(args.cache_dir),
-                    "token_limit": args.token_limit
+                    "token_limit": args.token_limit,
+                    "commit_hashes": commit_hashes
                 }
             }
         })
@@ -497,7 +504,17 @@ async def handle_githook(args: argparse.Namespace) -> None:
         
         # Run git hook analysis
         logger.info("Starting git hook analysis")
-        await git_handler.run_githook_mode()
+        if len(commit_hashes) == 0:
+            # Process current staged changes
+            await git_handler.run_githook_mode()
+        elif len(commit_hashes) == 1:
+            # Process specific commit
+            await git_handler.run_githook_mode(commit_hash=commit_hashes[0])
+        elif len(commit_hashes) == 2:
+            # Process commit range
+            await git_handler.run_githook_mode(commit_range=(commit_hashes[0], commit_hashes[1]))
+        else:
+            raise ValueError("--githook accepts 0, 1, or 2 commit hashes")
         logger.info("Git hook analysis completed successfully")
         
     except Exception as e:
@@ -682,15 +699,18 @@ def generate_project_markdown(project, branch, overview, files, logger):
     
     markdown_lines = []
     
-    # Project header
-    markdown_lines.append(f"# {project.name}")
+    # Project header with sentence case
+    project_name = project.name.title() if project.name.islower() else project.name
+    markdown_lines.append(f"# {project_name}")
     markdown_lines.append("")
     
     # Project metadata
     if project.remote_origin:
         markdown_lines.append(f"**Repository:** {project.remote_origin}")
+        markdown_lines.append("")
     if project.upstream_origin:
         markdown_lines.append(f"**Upstream:** {project.upstream_origin}")
+        markdown_lines.append("")
     markdown_lines.append(f"**Branch:** {branch}")
     markdown_lines.append("")
     
@@ -768,8 +788,8 @@ async def main() -> None:
     """Main entry point for the MCP server."""
     args = parse_arguments()
     
-    # Handle git hook command
-    if args.githook:
+    # Handle git hook command  
+    if args.githook is not None:
         await handle_githook(args)
         return
     

{mcp_code_indexer-1.5.0.dist-info → mcp_code_indexer-1.6.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mcp-code-indexer
-Version: 1.5.0
+Version: 1.6.1
 Summary: MCP server that tracks file descriptions across codebases, enabling AI agents to efficiently navigate and understand code through searchable summaries and token-aware overviews.
 Author: MCP Code Indexer Contributors
 Maintainer: MCP Code Indexer Contributors
@@ -57,8 +57,8 @@ Dynamic: requires-python
 
 # MCP Code Indexer 🚀
 
-[![PyPI version](https://badge.fury.io/py/mcp-code-indexer.svg?3)](https://badge.fury.io/py/mcp-code-indexer)
-[![Python](https://img.shields.io/pypi/pyversions/mcp-code-indexer.svg?3)](https://pypi.org/project/mcp-code-indexer/)
+[![PyPI version](https://badge.fury.io/py/mcp-code-indexer.svg?5)](https://badge.fury.io/py/mcp-code-indexer)
+[![Python](https://img.shields.io/pypi/pyversions/mcp-code-indexer.svg?5)](https://pypi.org/project/mcp-code-indexer/)
 [![License](https://img.shields.io/badge/License-MIT-blue.svg)](https://opensource.org/licenses/MIT)
 
 A production-ready **Model Context Protocol (MCP) server** that revolutionizes how AI agents navigate and understand codebases. Instead of repeatedly scanning files, agents get instant access to intelligent descriptions, semantic search, and context-aware recommendations.
@@ -246,6 +246,28 @@ The server provides **11 powerful MCP tools** for intelligent codebase managemen
 
 💡 **Pro Tip**: Always start with `check_codebase_size` to get personalized recommendations for navigating your specific codebase.
 
+## 🔗 Git Hook Integration
+
+Keep your codebase documentation automatically synchronized with automated analysis on every commit, rebase, or merge:
+
+```bash
+# Analyze current staged changes
+mcp-code-indexer --githook
+
+# Analyze a specific commit
+mcp-code-indexer --githook abc123def
+
+# Analyze a commit range (perfect for rebases)
+mcp-code-indexer --githook abc123 def456
+```
+
+**🎯 Perfect for**:
+- **Automated documentation** that never goes stale
+- **Rebase-aware analysis** that handles complex git operations
+- **Zero-effort maintenance** with background processing
+
+See the **[Git Hook Setup Guide](docs/git-hook-setup.md)** for complete installation instructions including post-commit, post-merge, and post-rewrite hooks.
+
 ## 🏗️ Architecture Highlights
 
 ### Performance Optimized

{mcp_code_indexer-1.5.0.dist-info → mcp_code_indexer-1.6.1.dist-info}/RECORD

@@ -1,10 +1,10 @@
-mcp_code_indexer/__init__.py,sha256=MkcEc96Vu59HUZk8IWIwS4psMwejpOuIvqik9_crg-w,473
+mcp_code_indexer/__init__.py,sha256=DKmuPpw8URDyojC4T1bttK8kGMI9lv_-B9UYP8dqb90,473
 mcp_code_indexer/__main__.py,sha256=4Edinoe0ug43hobuLYcjTmGp2YJnlFYN4_8iKvUBJ0Q,213
 mcp_code_indexer/error_handler.py,sha256=cNSUFFrGBMLDv4qa78c7495L1wSl_dXCRbzCJOidx-Q,11590
 mcp_code_indexer/file_scanner.py,sha256=ctXeZMROgDThEtjzsANTK9TbK-fhTScMBd4iyuleBT4,11734
-mcp_code_indexer/git_hook_handler.py,sha256=u_C5zVSDFkagBufgqOPgZHh3ZgvUuDzfqm0BLmCE2Fo,21253
+mcp_code_indexer/git_hook_handler.py,sha256=qd6y48PDnwnmErzkGyQGILuD3wQe3hbKEN0i6kxdIX4,25884
 mcp_code_indexer/logging_config.py,sha256=yCGQD-xx9oobS-YctOFcaE1Q3iiuOj2E6cTfKHbh_wc,7358
-mcp_code_indexer/main.py,sha256=e4Peuus4kZt0lc986pCtOz3Kk194rGN6cwsvxTqde28,30512
+mcp_code_indexer/main.py,sha256=7k00hj2C1CxTDDErbq2Ee072MbvqEAsRqrMHH-w1oM0,31538
 mcp_code_indexer/merge_handler.py,sha256=lJR8eVq2qSrF6MW9mR3Fy8UzrNAaQ7RsI2FMNXne3vQ,14692
 mcp_code_indexer/token_counter.py,sha256=WrifOkbF99nWWHlRlhCHAB2KN7qr83GOHl7apE-hJcE,8460
 mcp_code_indexer/data/stop_words_english.txt,sha256=7Zdd9ameVgA6tN_zuXROvHXD4hkWeELVywPhb7FJEkw,6343
@@ -17,9 +17,9 @@ mcp_code_indexer/server/__init__.py,sha256=16xMcuriUOBlawRqWNBk6niwrvtv_JD5xvI36
 mcp_code_indexer/server/mcp_server.py,sha256=bRl5JTFWlQ0MrIulkts6fDws6kPqfy6kKoQdenMOk04,61290
 mcp_code_indexer/tiktoken_cache/9b5ad71b2ce5302211f9c61530b329a4922fc6a4,sha256=Ijkht27pm96ZW3_3OFE-7xAPtR0YyTWXoRO8_-hlsqc,1681126
 mcp_code_indexer/tools/__init__.py,sha256=m01mxML2UdD7y5rih_XNhNSCMzQTz7WQ_T1TeOcYlnE,49
-mcp_code_indexer-1.5.0.dist-info/licenses/LICENSE,sha256=JN9dyPPgYwH9C-UjYM7FLNZjQ6BF7kAzpF3_4PwY4rY,1086
-mcp_code_indexer-1.5.0.dist-info/METADATA,sha256=yTD1vbKGaJ8RGmFL4SsRYZj1mwB4Wls5_FHz2YwISTM,16721
-mcp_code_indexer-1.5.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-mcp_code_indexer-1.5.0.dist-info/entry_points.txt,sha256=8HqWOw1Is7jOP1bvIgaSwouvT9z_Boe-9hd4NzyJOhY,68
-mcp_code_indexer-1.5.0.dist-info/top_level.txt,sha256=yKYCM-gMGt-cnupGfAhnZaoEsROLB6DQ1KFUuyKx4rw,17
-mcp_code_indexer-1.5.0.dist-info/RECORD,,
+mcp_code_indexer-1.6.1.dist-info/licenses/LICENSE,sha256=JN9dyPPgYwH9C-UjYM7FLNZjQ6BF7kAzpF3_4PwY4rY,1086
+mcp_code_indexer-1.6.1.dist-info/METADATA,sha256=Fa4NYHH1gbZoF_aloOCDVWF7khgSqSU2ppYYZC_uOEI,17453
+mcp_code_indexer-1.6.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+mcp_code_indexer-1.6.1.dist-info/entry_points.txt,sha256=8HqWOw1Is7jOP1bvIgaSwouvT9z_Boe-9hd4NzyJOhY,68
+mcp_code_indexer-1.6.1.dist-info/top_level.txt,sha256=yKYCM-gMGt-cnupGfAhnZaoEsROLB6DQ1KFUuyKx4rw,17
+mcp_code_indexer-1.6.1.dist-info/RECORD,,