diff --git a/.github/workflows/reindex_on_merge.yml b/.github/workflows/reindex_on_merge.yml new file mode 100644 index 0000000..40005eb --- /dev/null +++ b/.github/workflows/reindex_on_merge.yml @@ -0,0 +1,42 @@ +# .github/workflows/reindex_on_merge.yml + +name: Re-index Repository on Merge (Self-Hosted) + +on: + pull_request: + types: [closed] + branches: + - main + workflow_dispatch: + +jobs: + reindex: + # This condition ensures the job only runs if the pull request was actually merged. + if: github.event.pull_request.merged == true || github.event_name == 'workflow_dispatch' + + # *** KEY CHANGE *** + # This tells GitHub to run this job on one of your self-hosted runners. + # You can also add labels to target specific servers, e.g., [self-hosted, linux, x64, my-app] + runs-on: inference-server + + steps: + # Step 1: Check out the repository's code + # This downloads the latest version of your 'main' branch into the runner's working directory. + - name: Checkout repository + uses: actions/checkout@v4 + + # Step 2: Run the indexing script + # This executes your 'create_index.py' script using the Python environment on your server. + # It assumes Python and all dependencies from requirements.txt are already installed on the server. + # The GITHUB_TOKEN is still passed securely to the script. + - name: Run indexing script + run: python create_index.py + env: + GITHUB_REPOSITORY: ${{ github.repository }} + GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} + # Optional: Specify the working directory if your bot lives in a subfolder + # working-directory: ./path/to/your/bot + + # The "Upload database artifact" step is no longer needed, as the database + # is now being written directly to a persistent location on your server. + diff --git a/create_index.py b/create_index.py new file mode 100644 index 0000000..d1292c7 --- /dev/null +++ b/create_index.py @@ -0,0 +1,185 @@ +import os +import logging +import torch +import gc # Import the garbage collection module +import chromadb +from chromadb.utils import embedding_functions +from dotenv import load_dotenv + +# Import the GitHubTool from its location +# Assuming it's in a 'tools' directory as per your other script +from tools.github_tool import GitHubTool + +# --- Configuration --- +# You can adjust these settings + +# If you have downloaded a model, provide the local path here. +# Otherwise, the model will be downloaded from Hugging Face. +# Example: EMBEDDING_MODEL_PATH = "/path/to/your/models/all-MiniLM-L6-v2" +EMBEDDING_MODEL_PATH = """C:\Models\embeddings\Qwen3-Embedding-0.6B""" + +# Path to store the local vector database +CHROMA_DB_PATH = "C:\Models\embeddings\embedding_result\chroma_db" +# Name of the collection within the database +CHROMA_COLLECTION_NAME = "github_repo" +# Files with these extensions will be indexed. Add any other text-based files you need. +# Excludes common binary/unwanted files. +INCLUDED_EXTENSIONS = ['.py', '.js', '.ts', '.md', '.txt', '.html', '.css', '.go', '.rs', '.java', '.c', '.h', '.cpp', '.sh', '.yaml', '.json'] + +# *** NEW: Intelligent Chunking Configuration *** +CHUNK_SIZE = 1000 # The target size for each text chunk in characters +CHUNK_OVERLAP = 200 # The number of characters to overlap between chunks +CHUNK_PROCESSING_BATCH_SIZE = 100 # The number of chunks to process in a single batch + +logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s') + +def get_all_repo_files(github_tool, repo_path=""): + """ + Recursively fetches all file paths from the GitHub repository. + """ + all_files = [] + try: + items = github_tool.execute("list_files", path=repo_path) + if isinstance(items, str) and items.startswith("Error"): + logging.error(f"Could not list files at path '{repo_path}': {items}") + return [] + + for item in items: + # Only index files with allowed extensions + if item['type'] == 'file' and any(item['name'].endswith(ext) for ext in INCLUDED_EXTENSIONS): + all_files.append(item['path']) + elif item['type'] == 'dir': + # It's a directory, so recurse into it + all_files.extend(get_all_repo_files(github_tool, repo_path=item['path'])) + + return all_files + except Exception as e: + logging.error(f"An unexpected error occurred while listing files at '{repo_path}': {e}") + return all_files + +def split_text(text: str) -> list[str]: + """ + Splits text into chunks of a specified size with overlap. + """ + chunks = [] + if text is None or not text.strip(): + return [] + + start = 0 + while start < len(text): + end = start + CHUNK_SIZE + chunks.append(text[start:end]) + start += CHUNK_SIZE - CHUNK_OVERLAP + return chunks + +def main(): + """ + Main function to initialize the database and index the GitHub repository. + """ + load_dotenv() + logging.info("Starting repository indexing process...") + + # 1. Initialize GitHub Tool + try: + github_repo = os.getenv("GITHUB_REPOSITORY") + github_token = os.getenv("GITHUB_TOKEN") + if not github_repo or not github_token: + raise ValueError("GITHUB_REPOSITORY and GITHUB_TOKEN environment variables are required.") + + github_tool = GitHubTool(repo=github_repo, token=github_token) + logging.info(f"Successfully initialized GitHubTool for repo: {github_repo}") + except Exception as e: + logging.fatal(f"Failed to initialize GitHubTool: {e}") + return + + # 2. Initialize ChromaDB and Embedding Model + client = chromadb.PersistentClient(path=CHROMA_DB_PATH) + device = "cuda" if torch.cuda.is_available() else "cpu" + logging.info(f"Using device: {device} for embedding model inference.") + model_location = EMBEDDING_MODEL_PATH + logging.info(f"Loading embedding model from: {model_location}") + + sentence_transformer_ef = embedding_functions.SentenceTransformerEmbeddingFunction( + model_name=model_location, + device=device + ) + + logging.info(f"Loading or creating Chroma collection: '{CHROMA_COLLECTION_NAME}'") + collection = client.get_or_create_collection( + name=CHROMA_COLLECTION_NAME, + embedding_function=sentence_transformer_ef, + metadata={"hnsw:space": "cosine"} + ) + + # 3. Fetch all file paths from the repository + logging.info("Fetching all file paths from the repository...") + file_paths = get_all_repo_files(github_tool) + if not file_paths: + logging.warning("No files found to index. Exiting.") + return + logging.info(f"Found {len(file_paths)} files to potentially index.") + + # 4. Process files and upsert to ChromaDB in chunk-based batches + + batch_documents = [] + batch_metadatas = [] + batch_ids = [] + + for i, file_path in enumerate(file_paths): + logging.info(f"Processing file {i+1}/{len(file_paths)}: {file_path}") + try: + content = github_tool.execute("read_file", path=file_path) + if not isinstance(content, str) or content.startswith("Error"): + logging.warning(f"Could not read or empty content for {file_path}. Skipping.") + continue + + # *** USE THE NEW, ROBUST CHUNKING METHOD *** + chunks = split_text(content) + + for chunk_index, chunk in enumerate(chunks): + # Add the processed chunk to the current batch + unique_id = f"{file_path}_{chunk_index}" + batch_documents.append(chunk) + batch_metadatas.append({"source": file_path, "chunk_index": chunk_index}) + batch_ids.append(unique_id) + + # If the batch reaches the desired size, upsert it to the database + if len(batch_documents) >= CHUNK_PROCESSING_BATCH_SIZE: + logging.info(f"Upserting batch of {len(batch_documents)} chunks...") + collection.upsert( + ids=batch_ids, + documents=batch_documents, + metadatas=batch_metadatas + ) + # Clear the batch lists to free up memory + batch_documents, batch_metadatas, batch_ids = [], [], [] + + # Force garbage collection and empty CUDA cache + logging.info("Cleaning up memory...") + gc.collect() + if device == 'cuda': + torch.cuda.empty_cache() + logging.info("Batch upserted and memory cleared.") + + except Exception as e: + logging.error(f"Error processing file {file_path}: {e}") + + # 5. Upsert any remaining documents after the loop finishes + if batch_documents: + logging.info(f"Upserting final batch of {len(batch_documents)} chunks...") + collection.upsert( + ids=batch_ids, + documents=batch_documents, + metadatas=batch_metadatas + ) + # Final cleanup + gc.collect() + if device == 'cuda': + torch.cuda.empty_cache() + logging.info("Final batch upserted.") + + logging.info("--- Indexing Complete ---") + logging.info(f"Total documents in collection: {collection.count()}") + +if __name__ == '__main__': + main() diff --git a/rag_inference_bot.py b/rag_inference_bot.py new file mode 100644 index 0000000..f005b8e --- /dev/null +++ b/rag_inference_bot.py @@ -0,0 +1,219 @@ +import logging +import chromadb +from chromadb.utils import embedding_functions +from inference_bot import InferenceBot # Correctly inherit from the ABC +from FlagEmbedding import FlagReranker +import argparse +import os +import importlib +import torch +from transformers import AutoTokenizer, AutoModelForCausalLM + + +# --- RAG Configuration --- +# Must match the settings in create_index.py +EMBEDDING_MODEL_NAME = """C:\Models\embeddings\Qwen3-Embedding-0.6B""" +CHROMA_DB_PATH = "C:\Models\embeddings\embedding_result\chroma_db" +CHROMA_COLLECTION_NAME = "github_repo" + +# Using a powerful open-source reranker model +RERANKER_MODEL_NAME = """C:\Models\embeddings\Qwen3-Reranker-0.6B""" + +# Number of initial results to fetch from the database before reranking +N_RESULTS_TO_RETRIEVE = 25 +# Number of final results to keep after reranking +N_RESULTS_TO_KEEP_AFTER_RERANK = 5 +# The minimum relevance score a result must have to be included in the final context +RERANKER_SCORE_THRESHOLD = 0.5 + +class RAGInferenceBot(InferenceBot): + def __init__(self): + """ + Initializes the RAG components, including a custom implementation + for the Qwen3-Reranker based on its model card. + """ + logging.info("Initializing RAG components...") + self._processing_status = {} + try: + # --- Embedding and Vector DB Initialization --- + self.chroma_client = chromadb.PersistentClient(path=CHROMA_DB_PATH) + self.embedding_function = embedding_functions.SentenceTransformerEmbeddingFunction( + model_name=EMBEDDING_MODEL_NAME + ) + self.collection = self.chroma_client.get_collection( + name=CHROMA_COLLECTION_NAME, + embedding_function=self.embedding_function + ) + logging.info("Successfully connected to ChromaDB collection for RAG.") + + # --- Custom Reranker Initialization (as per model card) --- + self.device = "cuda" if torch.cuda.is_available() else "cpu" + logging.info(f"Using device: {self.device} for Reranker model.") + + self.rerank_tokenizer = AutoTokenizer.from_pretrained(RERANKER_MODEL_NAME, padding_side='left') + self.rerank_model = AutoModelForCausalLM.from_pretrained(RERANKER_MODEL_NAME, torch_dtype=torch.float16).to(self.device).eval() + + # Manually set the padding token if it's missing + if self.rerank_tokenizer.pad_token is None: + logging.warning("Reranker tokenizer has no pad_token. Setting it to the eos_token.") + self.rerank_tokenizer.pad_token = self.rerank_tokenizer.eos_token + + # Get token IDs for score calculation + self.token_false_id = self.rerank_tokenizer.convert_tokens_to_ids("no") + self.token_true_id = self.rerank_tokenizer.convert_tokens_to_ids("yes") + self.max_length = 8192 + + # Define and pre-encode the special prefixes and suffixes from the model card + prefix_text = "<|im_start|>system\nJudge whether the Document meets the requirements based on the Query and the Instruct provided. Note that the answer can only be \"yes\" or \"no\".<|im_end|>\n<|im_start|>user\n" + suffix_text = "<|im_end|>\n<|im_start|>assistant\n\n\n\n\n" + self.prefix_tokens = self.rerank_tokenizer.encode(prefix_text, add_special_tokens=False) + self.suffix_tokens = self.rerank_tokenizer.encode(suffix_text, add_special_tokens=False) + + logging.info(f"Successfully initialized custom Reranker: {RERANKER_MODEL_NAME}") + + except Exception as e: + logging.fatal(f"Failed to initialize RAG components: {e}", exc_info=True) + raise + + # --- Implementation of all abstract methods from InferenceBot --- + + @property + def processing_status(self): + return self._processing_status + + def clear_conversation_history(self, user_id): + pass + + async def switch_model(self): + return "This bot only performs RAG lookups and has no swappable models." + + def set_processing_status(self, user_id, message_id): + self._processing_status[user_id] = {"processing": True, "message_id": message_id} + + def clear_processing_status(self, user_id): + if user_id in self._processing_status: + del self._processing_status[user_id] + + async def abort_processing(self, user_id): + if user_id in self.processing_status: + self.clear_processing_status(user_id) + return "Processing aborted." + else: + return "No active processing found to abort." + + def get_bot_status(self): + return f"RAG Bot is active.\nEmbedding Model: {os.path.basename(EMBEDDING_MODEL_NAME)}\nReranker Model: {os.path.basename(RERANKER_MODEL_NAME)}" + + async def start(self): + logging.info(f"{self.__class__.__name__} started.") + + # --- Core RAG Logic --- + + def _format_rerank_instruction(self, query, doc): + # Using the default instruction from the model card + instruction = 'Given a web search query, retrieve relevant passages that answer the query' + return f": {instruction}\n: {query}\n: {doc}" + + @torch.no_grad() + def _compute_rerank_scores(self, pairs: list[list[str, str]]): + """ + Custom score computation logic that follows the model card's example. + """ + # Format all pairs with the required instruction format + formatted_pairs = [self._format_rerank_instruction(query, doc) for query, doc in pairs] + + # Tokenize the formatted pairs + inputs = self.rerank_tokenizer( + formatted_pairs, padding=False, truncation='longest_first', + return_attention_mask=False, max_length=self.max_length - len(self.prefix_tokens) - len(self.suffix_tokens) + ) + + # Add special prefix and suffix tokens to each item in the batch + for i in range(len(inputs['input_ids'])): + inputs['input_ids'][i] = self.prefix_tokens + inputs['input_ids'][i] + self.suffix_tokens + + # Pad the batch to the same length + inputs = self.rerank_tokenizer.pad(inputs, padding=True, return_tensors="pt", max_length=self.max_length) + inputs = {key: inputs[key].to(self.device) for key in inputs} + + # Get model outputs (logits) + batch_scores = self.rerank_model(**inputs).logits[:, -1, :] + + # Calculate scores based on the probability of "yes" vs "no" + true_vector = batch_scores[:, self.token_true_id] + false_vector = batch_scores[:, self.token_false_id] + + scores = torch.stack([false_vector, true_vector], dim=1) + scores = torch.nn.functional.log_softmax(scores, dim=1) + + final_scores = scores[:, 1].exp().tolist() + return final_scores + + def _retrieve_and_rerank_context(self, query: str): + logging.info(f"RAG: Retrieving context for query: '{query}'") + if not query: return "" + try: + results = self.collection.query(query_texts=[query], n_results=N_RESULTS_TO_RETRIEVE) + initial_docs = results.get('documents', [[]])[0] + if not initial_docs: + logging.info("RAG: No initial documents found in vector search.") + return "No relevant context found in the knowledge base." + + logging.info(f"RAG: Retrieved {len(initial_docs)} initial documents from ChromaDB.") + + # Create pairs of [query, document] for reranking + rerank_pairs = [[query, doc] for doc in initial_docs] + + # Use our custom scoring function + scores = self._compute_rerank_scores(rerank_pairs) + + scored_docs = sorted(zip(scores, initial_docs, results['metadatas'][0]), key=lambda x: x[0], reverse=True) + + # Take the top N results + top_k_docs = scored_docs[:N_RESULTS_TO_KEEP_AFTER_RERANK] + + # *** NEW: Filter the top N results by the score threshold *** + final_docs = [doc for doc in top_k_docs if doc[0] >= RERANKER_SCORE_THRESHOLD] + + logging.info(f"RAG: Reranked and filtered down to {len(final_docs)} documents with score >= {RERANKER_SCORE_THRESHOLD}.") + + # If no documents meet the threshold, inform the user. + if not final_docs: + return "No highly relevant context found after filtering." + + context_lines = [] + for i, (score, doc, metadata) in enumerate(final_docs): + source = metadata.get('source', 'Unknown file') + chunk_index = metadata.get('chunk_index', 'N/A') + context_lines.append(f"--- Context {i+1} (Relevance: {score:.2f}) ---\nSource: {source}\n\n{doc}\n") + + return "\n".join(context_lines) + except Exception as e: + logging.error(f"RAG: Error during context retrieval/reranking: {e}", exc_info=True) + return "An error occurred while searching the knowledge base." + + async def handle_message(self, user_id, user_message): + context_response = self._retrieve_and_rerank_context(user_message) + return context_response + + +def main_rag(): + logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s') + + parser = argparse.ArgumentParser(description='RAG-Only Inference Bot') + parser.add_argument('--messenger', type=str, help='Messenger type (i.e. telegram)', required=True) + args = parser.parse_args() + + try: + bot = RAGInferenceBot() + + full_code_file = importlib.import_module(f'{args.messenger.lower()}_helper') + helper_class = getattr(full_code_file, f"{args.messenger.capitalize()}Helper") + helper = helper_class(bot) + helper.run() + + except Exception as e: + logging.fatal(f"An unexpected error occurred during bot initialization: {e}", exc_info=True) + +if __name__ == '__main__': + main_rag() diff --git a/requirements.txt b/requirements.txt index d76bbcd..6836b11 100644 --- a/requirements.txt +++ b/requirements.txt @@ -10,3 +10,9 @@ pytest-cov google-genai httpx==0.27.2 tiktoken +chromadb +sentence-transformers +transformers>=4.38 +torch --index-url https://download.pytorch.org/whl/cu121 +torchvision --index-url https://download.pytorch.org/whl/cu121 +torchaudio --index-url https://download.pytorch.org/whl/cu121 \ No newline at end of file diff --git a/tools/github_tool.py b/tools/github_tool.py index 034147c..8b6913c 100644 --- a/tools/github_tool.py +++ b/tools/github_tool.py @@ -1,4 +1,1275 @@ -# tools/github_tool.py\nfrom .base_tool import BaseTool\nimport requests\nimport os\nimport base64\nimport logging\n\nclass GitHubTool(BaseTool):\n def __init__(self, session=None, token=None, repo=None, base_url=None, initial_branch=\"main\", logger=None):\n self.base_url = base_url if base_url else \"https://api.github.com\"\n self._token = token if token else os.environ.get(\"GITHUB_TOKEN\")\n self._repo = repo if repo else os.environ.get(\"GITHUB_REPOSITORY\")\n \n if not self._token:\n # In a real scenario, might raise an error or operate in a degraded mode.\n # For this tool, token is essential.\n raise ValueError(\"GitHub token must be provided either as an argument or via GITHUB_TOKEN env var.\")\n if not self._repo:\n raise ValueError(\"GitHub repository (e.g., \'owner/repo\') must be provided either as an argument or via GITHUB_REPOSITORY env var.\")\n\n if session:\n self.session = session\n else:\n self.session = requests.Session()\n self.session.headers.update({\n \"Authorization\": f\"token {self._token}\",\n \"Accept\": \"application/vnd.github.v3+json\"\n })\n \n self.current_branch = initial_branch\n\n # Use provided logger or get a new one for the module\n # The application using this tool should configure the logging handlers and formatting.\n self.logger = logger if logger else logging.getLogger(__name__)\n # If no handlers are configured by the application, add a NullHandler\n # to prevent \"No handler found\" warnings if the tool logs something.\n if not self.logger.handlers:\n self.logger.addHandler(logging.NullHandler())\n\n def clear(self):\n if self.current_branch != \"main\":\n self._set_current_branch(\"main\")\n self.logger.info(f\"GitHubTool state cleared. Current branch is {self.current_branch}\")\n\n def get_functions(self):\n return [\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"read_file\",\n \"description\": \"Read a file from the repository\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"path\": {\"type\": \"string\", \"description\": \"Path to the file in the repository\"}\n },\n \"required\": [\"path\"]\n }\n },\n \"_tags\": [\"read\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"read_readme\",\n \"description\": \"Read the README.md file from the root of the repository\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {}\n }\n },\n \"_tags\": [\"read\", \"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"list_files\",\n \"description\": \"List files in a directory of the repository\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"path\": {\"type\": \"string\", \"description\": \"Path to the directory in the repository\"}\n },\n \"required\": [\"path\"]\n }\n },\n \"_tags\": [\"read\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"search_code\",\n \"description\": \"Search for code in the repository\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"query\": {\"type\": \"string\", \"description\": \"Search query\"}\n },\n \"required\": [\"query\"]\n }\n },\n \"_tags\": [\"read\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"create_branch\",\n \"description\": \"Create a new branch in the repository\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"branch_name\": {\"type\": \"string\", \"description\": \"Name of the new branch\"},\n \"base_branch\": {\"type\": \"string\", \"description\": \"Name of the base branch\", \"default\": \"main\"}\n },\n \"required\": [\"branch_name\"]\n }\n },\n \"_tags\": [\"write\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"commit_file\",\n \"description\": \"Commit a file to a branch (not main)\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"file_path\": {\"type\": \"string\", \"description\": \"Path to the file in the repository\"},\n \"commit_message\": {\"type\": \"string\", \"description\": \"Commit message\"},\n \"content\": {\"type\": \"string\", \"description\": \"Content of the file\"}\n },\n \"required\": [\"file_path\", \"commit_message\", \"content\"]\n }\n },\n \"_tags\": [\"write\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"create_pull_request\",\n \"description\": \"Create a pull request\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"title\": {\"type\": \"string\", \"description\": \"Title of the pull request\"},\n \"body\": {\"type\": \"string\", \"description\": \"Body of the pull request\"},\n \"base\": {\"type\": \"string\", \"description\": \"The name of the branch you want the changes pulled into\", \"default\": \"main\"}\n },\ - \"required\": [\"title\", \"body\"]\n }\n },\n \"_tags\": [\"write\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_commit_history\",\n \"description\": \"Get commit history for a file\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"file_path\": {\"type\": \"string\", \"description\": \"Path to the file in the repository\"},\n \"num_commits\": {\"type\": \"integer\", \"description\": \"Number of commits to retrieve\", \"default\": 10}\n },\n \"required\": [\"file_path\"]\n }\n },\n \"_tags\": [\"read\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"view_commit_details_for_file\",\n \"description\": \"View commit history and details for a specific file, including commit messages.\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"file_path\": {\"type\": \"string\", \"description\": \"Path to the file in the repository\"},\n \"num_commits\": {\"type\": \"integer\", \"description\": \"Number of commits to retrieve\", \"default\": 10}\n },\n \"required\": [\"file_path\"]\n }\n },\n \"_tags\": [\"read\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_branch_sha\",\n \"description\": \"Get the SHA of the latest commit on a branch\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"branch\": {\"type\": \"string\", \"description\": \"Name of the branch\"}\n },\n \"required\": [\"branch\"]\n }\n },\n \"_tags\": [\"read\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_current_branch\",\n \"description\": \"Get the name of the current branch\",\n \"parameters\": { \"type\": \"object\", \"properties\": {} }\n },\n \"_tags\": [\"read\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"set_current_branch\",\n \"description\": \"Set the current branch\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"branch_name\": {\"type\": \"string\", \"description\": \"Name of the branch to set as current\"}\n },\n \"required\": [\"branch_name\"]\n }\n },\n \"_tags\": [\"read\", \"write\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_file_at_commit\",\n \"description\": \"Get the contents of a file at a specific commit\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"file_path\": {\"type\": \"string\", \"description\": \"Path to the file in the repository\"},\n \"commit_sha\": {\"type\": \"string\", \"description\": \"SHA of the commit to retrieve the file from\"}\n },\n \"required\": [\"file_path\", \"commit_sha\"]\n }\n },\n \"_tags\": [\"read\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"list_branches\",\n \"description\": \"List all branches in the repository\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"per_page\": {\"type\": \"integer\", \"description\": \"Number of branches to return per page (max 100)\", \"default\": 100},\n \"all_pages\": {\"type\": \"boolean\", \"description\": \"Whether to fetch all pages of results\", \"default\": True}\n }\n }\n },\n \"_tags\": [\"read\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"approve_pull_request\",\n \"description\": \"Approve a pull request\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request\"}\n },\n \"required\": [\"pull_number\"]\n }\n },\n \"_tags\": [\"write\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"close_pull_request\",\n \"description\": \"Close a pull request\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request\"}\n },\n \"required\": [\"pull_number\"]\n }\n },\n \"_tags\": [\"write\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"merge_pull_request\",\n \"description\": \"Merge a pull request\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request\"},\n \"commit_title\": {\"type\": \"string\", \"description\": \"Title for the automatic commit message\", \"default\": \"Merge pull request\"},\n \"commit_message\": {\"type\": \"string\", \"description\": \"Extra detail to append to automatic commit message\", \"default\": \"\"},\n \"merge_method\": {\n \"type\": \"string\",\n \"description\": \"Merge method to use\",\n \"enum\": [\"merge\", \"squash\", \"rebase\"],\n \"default\": \"merge\"\n }\n },\n \"required\": [\"pull_number\"]\n }\n },\n \"_tags\": [\"write\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"delete_branch\",\n \"description\": \"Delete a branch\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"branch_name\": {\"type\": \"string\", \"description\": \"Name of the branch to delete\"}\n },\ - \"required\": [\"branch_name\"]\n }\n },\n \"_tags\": [\"write\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_issue_details\",\n \"description\": \"Get details of a specific issue\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"issue_number\": {\"type\": \"integer\", \"description\": \"The number of the issue\"}\n },\n \"required\": [\"issue_number\"]\n }\n },\n \"_tags\": [\"read\", \"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"create_issue\",\n \"description\": \"Create a new issue in the repository\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"title\": {\"type\": \"string\", \"description\": \"Title of the issue\"},\n \"body\": {\"type\": \"string\", \"description\": \"Body of the issue\"},\n \"labels\": {\n \"type\": \"array\",\n \"items\": {\"type\": \"string\"},\n \"description\": \"Labels to apply to the issue\"\n }\n },\n \"required\": [\"title\", \"body\"]\n }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"list_issues\",\n \"description\": \"List issues in the repository\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"state\": {\"type\": \"string\", \"enum\": [\"open\", \"closed\", \"all\"], \"default\": \"open\", \"description\": \"State of the issues to retrieve\"},\n \"per_page\": {\"type\": \"integer\", \"default\": 30, \"description\": \"Number of issues to return per page\"},\n \"page\": {\"type\": \"integer\", \"default\": 1, \"description\": \"Page number of the results to fetch\"}\n }\n }\n },\n \"_tags\": [\"read\", \"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"add_issue_comment\",\n \"description\": \"Add a comment to an issue\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"issue_number\": {\"type\": \"integer\", \"description\": \"The number of the issue\"},\n \"comment\": {\"type\": \"string\", \"description\": \"The comment to add to the issue\"}\n },\ - \"required\": [\"issue_number\", \"comment\"]\n }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_issue_comments\",\n \"description\": \"Get comments for an issue\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"issue_number\": {\"type\": \"integer\", \"description\": \"The number of the issue\"}\n },\n \"required\": [\"issue_number\"]\n }\n },\n \"_tags\": [\"read\", \"communicate\"]\n },\n { \n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_pull_request_general_comments\",\n \"description\": \"Get general comments posted on a pull request itself (not specific to file lines).\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request.\"}\n },\n \"required\": [\"pull_number\"]\n }\n },\n \"_tags\": [\"read\", \"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"create_project_board\",\n \"description\": \"Create a new project board\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"name\": {\"type\": \"string\", \"description\": \"Name of the project board\"},\n \"body\": {\"type\": \"string\", \"description\": \"Body of the project board\"}\n },\n \"required\": [\"name\"]\n }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"create_project_column\",\n \"description\": \"Create a new column in a project board\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"project_id\": {\"type\": \"integer\", \"description\": \"ID of the project board\"},\n \"column_name\": {\"type\": \"string\", \"description\": \"Name of the column\"}\n },\n \"required\": [\"project_id\", \"column_name\"]\n }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"create_project_card\",\n \"description\": \"Create a new card in a project column\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"column_id\": {\"type\": \"integer\", \"description\": \"ID of the project column\"},\n \"note\": {\"type\": \"string\", \"description\": \"Note for the project card\"}\n },\n \"required\": [\"column_id\", \"note\"]\n }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"move_project_card\",\n \"description\": \"Move a card to a new position\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"card_id\": {\"type\": \"integer\", \"description\": \"ID of the project card\"},\n \"position\": {\"type\": \"string\", \"description\": \"New position of the card\"},\n \"column_id\": {\"type\": \"integer\", \"description\": \"ID of the target column\"}\n },\n \"required\": [\"card_id\", \"position\", \"column_id\"]\n }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"link_issue_to_project_card\",\n \"description\": \"Link an issue or pull request to a project card\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"card_id\": {\"type\": \"integer\", \"description\": \"ID of the project card\"},\n \"content_id\": {\"type\": \"integer\", \"description\": \"ID of the issue or pull request\"},\n \"content_type\": {\"type\": \"string\", \"description\": \"Type of the content (Issue or PullRequest)\"}\n },\n \"required\": [\"card_id\", \"content_id\", \"content_type\"]\n }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"list_project_boards\",\n \"description\": \"List project boards associated with the repository\",\n \"parameters\": { \"type\": \"object\", \"properties\": {} }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"view_project_board_items\",\n \"description\": \"View items (columns and cards) in a specific project board\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"project_id\": {\"type\": \"integer\", \"description\": \"ID of the project board\"}\n },\n \"required\": [\"project_id\"]\n }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_pull_request_details\",\n \"description\": \"Get detailed information about a pull request\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request\"}\n },\n \"required\": [\"pull_number\"]\n }\n },\n \"_tags\": [\"read\", \"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_pull_request_diff\",\n \"description\": \"Get the diff of a pull request\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request\"}\n },\n \"required\": [\"pull_number\"]\n }\n },\n \"_tags\": [\"read\", \"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"get_pull_request_files\",\n \"description\": \"Get a list of files changed in a pull request, with their patch details\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request\"}\n },\n \"required\": [\"pull_number\"]\n }\n },\n \"_tags\": [\"read\", \"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"create_pull_request_review_comment\",\n \"description\": \"Add a comment to a specific line of a file in a pull request review\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request\"},\n \"body\": {\"type\": \"string\", \"description\": \"The text of the comment\"},\n \"commit_id\": {\"type\": \"string\", \"description\": \"The SHA of the commit to comment on\"},\n \"path\": {\"type\": \"string\", \"description\": \"The path to the file being commented on\"},\n \"position\": {\"type\": \"integer\", \"description\": \"The line index in the diff to comment on (starting at 1)\"},\n \"side\": {\"type\": \"string\", \"enum\": [\"LEFT\", \"RIGHT\"], \"description\": \"The side of the diff to comment on (LEFT for old file, RIGHT for new file)\", \"default\": \"RIGHT\"},\n \"start_line\": {\"type\": \"integer\", \"description\": \"The start line of the diff if commenting on a range\"},\n \"start_side\": {\"type\": \"string\", \"enum\": [\"LEFT\", \"RIGHT\"], \"description\": \"The side of the diff for the start line\"}\n },\n \"required\": [\"pull_number\", \"body\", \"commit_id\", \"path\", \"position\"]\n }\n },\n \"_tags\": [\"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"list_pull_request_review_comments\",\n \"description\": \"List comments on a pull request review\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request\"}\n },\n \"required\": [\"pull_number\"]\n }\n },\n \"_tags\": [\"read\", \"communicate\"]\n },\n {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"submit_pull_request_review\",\n \"description\": \"Submit a formal pull request review (APPROVE, REQUEST_CHANGES, COMMENT)\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"pull_number\": {\"type\": \"integer\", \"description\": \"The number of the pull request\"},\n \"event\": {\"type\": \"string\", \"enum\": [\"APPROVE\", \"REQUEST_CHANGES\", \"COMMENT\"], \"description\": \"The type of review event\"},\n \"body\": {\"type\": \"string\", \"description\": \"The body of the review (required for REQUEST_CHANGES, optional for others)\"}\n },\n \"required\": [\"pull_number\", \"event\"]\n }\n },\n \"_tags\": [\"communicate\"]\n }\n ]\n\n def execute(self, function_name, **kwargs):\n self.logger.info(f\"Executing GitHub Tool function: {function_name} with args: {kwargs}\")\n # Dispatch to the appropriate private method\n method_name = f\"_{function_name}\"\n if hasattr(self, method_name):\n method = getattr(self, method_name)\n try:\n return method(**kwargs) # Ensure only expected args are passed if method signature is strict\n except Exception as e:\n self.logger.error(f\"Error executing {method_name}: {e}\", exc_info=True)\n return f\"Error during {function_name} execution: {str(e)}\"\n else:\n error_message = f\"Unknown function: {function_name}\"\n self.logger.error(error_message)\n return error_message\n\n # Private methods for each function, using self.session for HTTP requests\n\n def _read_file(self, path):\n self.logger.info(f\"Reading file: {path} from branch: {self.current_branch}\")\n url = f\"{self.base_url}/repos/{self._repo}/contents/{path}\"\n response = self.session.get(url, params={\"ref\": self.current_branch})\n if response.status_code == 200:\n content = response.json()[\"content\"]\n decoded_content = base64.b64decode(content).decode(\'utf-8\')\n self.logger.info(f\"Successfully read file: {path}\")\n return decoded_content\n else:\n error_message = f\"Error reading file ({path}): {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _read_readme(self):\n self.logger.info(\"Reading README.md from the root of the repository.\")\n return self._read_file(\"README.md\")\n\n def _create_branch(self, branch_name, base_branch=\"main\"):\n self.logger.info(f\"Creating branch: {branch_name} from base: {base_branch}\")\n # Get SHA of base branch\n ref_url = f\"{self.base_url}/repos/{self._repo}/git/refs/heads/{base_branch}\"\n response_sha = self.session.get(ref_url)\n if response_sha.status_code != 200:\n error_message = f\"Error getting base branch SHA ({base_branch}): {response_sha.status_code} - {response_sha.text}\"\n self.logger.error(error_message)\n return error_message\n sha = response_sha.json()[\"object\"][\"sha\"]\n \n # Create new branch\n create_ref_url = f\"{self.base_url}/repos/{self._repo}/git/refs\"\n data = {\"ref\": f\"refs/heads/{branch_name}\", \"sha\": sha}\n response_create = self.session.post(create_ref_url, json=data)\n if response_create.status_code == 201:\n self.current_branch = branch_name\n success_message = f\"Branch \'{branch_name}\' created successfully from \'{base_branch}\' and set as current branch.\"\n self.logger.info(success_message)\n return success_message\n else:\n error_message = f\"Error creating branch \'{branch_name}\': {response_create.status_code} - {response_create.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _commit_file(self, file_path, content, commit_message):\n self.logger.info(f\"Committing file: {file_path} to branch: {self.current_branch} with message: \'{commit_message}\'\")\n if self.current_branch == \"main\":\n error_message = \"Action directly to main branch is not allowed. Please create and switch to a new branch first.\"\n self.logger.warning(error_message)\n return error_message\n\n url = f\"{self.base_url}/repos/{self._repo}/contents/{file_path}\"\n encoded_content = base64.b64encode(content.encode(\'utf-8\')).decode(\'utf-8\')\n data = {\n \"message\": commit_message,\n \"content\": encoded_content,\n \"branch\": self.current_branch\n }\n\n # Check if file exists to get its SHA for update\n self.logger.info(f\"Checking if file \'{file_path}\' exists on branch \'{self.current_branch}\'\")\n get_response = self.session.get(url, params={\"ref\": self.current_branch})\n if get_response.status_code == 200:\n data[\"sha\"] = get_response.json()[\"sha\"]\n self.logger.info(f\"File \'{file_path}\' exists, will update.\")\n elif get_response.status_code == 404:\n self.logger.info(f\"File \'{file_path}\' does not exist, will create.\")\n else:\n error_message = f\"Error checking file existence for \'{file_path}\': {get_response.status_code} - {get_response.text}\"\n self.logger.error(error_message)\n return error_message\n\n response = self.session.put(url, json=data)\n if response.status_code in [200, 201]: # 200 for update, 201 for create\n commit_sha = response.json().get(\"commit\", {}).get(\"sha\", \"N/A\")\n success_message = f\"File \'{file_path}\' committed successfully to branch \'{self.current_branch}\'. Commit SHA: {commit_sha}\"\n self.logger.info(success_message)\n return success_message\n else:\n error_message = f\"Error committing file \'{file_path}\': {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _create_pull_request(self, title, body, base=\"main\"):\n self.logger.info(f\"Creating pull request: \'{title}\' from branch \'{self.current_branch}\' to \'{base}\'\")\n if self.current_branch == base:\n error_message = f\"Cannot create a pull request from branch \'{self.current_branch}\' to itself (\'{base}\').\"\n self.logger.warning(error_message)\n return error_message\n \n url = f\"{self.base_url}/repos/{self._repo}/pulls\"\n data = {\"title\": title, \"body\": body, \"head\": self.current_branch, \"base\": base}\n response = self.session.post(url, json=data)\n if response.status_code == 201:\n pr_html_url = response.json().get(\"html_url\", \"N/A\")\n pr_number = response.json().get(\"number\", \"N/A\")\n success_message = f\"Pull request \'{title}\' created successfully: {pr_html_url} (Number: {pr_number})\"\n self.logger.info(success_message)\n return success_message\n else:\n error_message = f\"Error creating pull request: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _get_branch_sha(self, branch):\n self.logger.info(f\"Getting SHA for branch: {branch}\")\n url = f\"{self.base_url}/repos/{self._repo}/git/refs/heads/{branch}\"\n response = self.session.get(url)\n if response.status_code == 200:\n sha = response.json()[\"object\"][\"sha\"]\n self.logger.info(f\"SHA for branch \'{branch}\' is {sha}\")\n return sha\n else:\n error_message = f\"Error getting SHA for branch \'{branch}\': {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _list_files(self, path):\n self.logger.info(f\"Listing files in path: \'{path}\' on branch: \'{self.current_branch}\'\")\n url = f\"{self.base_url}/repos/{self._repo}/contents/{path.strip(\'/\')}\" # Ensure no leading/trailing slashes for consistency\n response = self.session.get(url, params={\"ref\": self.current_branch})\n if response.status_code == 200:\n items = response.json()\n results = []\n if isinstance(items, list): # It\'s a directory listing\n for item in items:\n results.append({\"name\": item[\"name\"], \"type\": item[\"type\"], \"path\": item[\"path\"]})\n elif isinstance(items, dict) and \'type\' in items: # It\'s a single file response\n results.append({\"name\": items[\"name\"], \"type\": items[\"type\"], \"path\": items[\"path\"]})\n self.logger.info(f\"Successfully listed {len(results)} items in \'{path}\'.\")\n return results\n elif response.status_code == 404:\n self.logger.warning(f\"Path \'{path}\' not found on branch \'{self.current_branch}\'.\")\n return f\"Error: Path \'{path}\' not found.\"\n else:\n error_message = f\"Error listing files in \'{path}\': {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _search_code(self, query):\n self.logger.info(f\"Searching code with query: \'{query}\' in repo: \'{self._repo}\'\")\n url = f\"{self.base_url}/search/code\"\n params = {\"q\": f\"{query} repo:{self._repo}\"}\n response = self.session.get(url, params=params)\n if response.status_code == 200:\n search_results = response.json().get(\"items\", [])\n results = [{\"path\": item[\"path\"], \"url\": item[\"html_url\"]} for item in search_results]\n self.logger.info(f\"Code search for \'{query}\' found {len(results)} items.\")\n return results\n else:\n error_message = f\"Error searching code for \'{query}\': {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _get_commit_history(self, file_path, num_commits=10):\n self.logger.info(f\"Getting last {num_commits} commit(s) for file: \'{file_path}\' on branch \'{self.current_branch}\'\")\n url = f\"{self.base_url}/repos/{self._repo}/commits\"\n params = {\"path\": file_path, \"sha\": self.current_branch, \"per_page\": num_commits}\n response = self.session.get(url, params=params)\n if response.status_code == 200:\n commits_data = response.json()\n commits = [{\n \"sha\": commit[\"sha\"],\n \"message\": commit[\"commit\"][\"message\"],\n \"author\": commit[\"commit\"][\"author\"][\"name\"],\n \"date\": commit[\"commit\"][\"author\"][\"date\"]\n } for commit in commits_data]\n self.logger.info(f\"Successfully retrieved {len(commits)} commit(s) for \'{file_path}\'.\")\n return commits\n else:\n error_message = f\"Error getting commit history for \'{file_path}\': {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _view_commit_details_for_file(self, file_path, num_commits=10):\n # This function is essentially the same as get_commit_history based on its description.\n self.logger.info(f\"Viewing commit details for file \'{file_path}\' (last {num_commits} commits) - using _get_commit_history.\")\n return self._get_commit_history(file_path, num_commits)\n\n def _get_current_branch(self):\n self.logger.info(f\"Current branch is: {self.current_branch}\")\n return self.current_branch\n\n def _set_current_branch(self, branch_name):\n self.logger.info(f\"Attempting to set current branch to: {branch_name}\")\n # Check if branch exists by trying to get its SHA\n sha_info = self._get_branch_sha(branch_name)\n if isinstance(sha_info, str) and sha_info.startswith(\"Error getting SHA\"): # Crude check for error string\n error_message = f\"Cannot set current branch: Branch \'{branch_name}\' not found or error accessing it. Details: {sha_info}\"\n self.logger.warning(error_message)\n return error_message \n \n self.current_branch = branch_name\n success_message = f\"Current branch set to: {self.current_branch}\"\n self.logger.info(success_message)\n return success_message\n\n def _get_file_at_commit(self, file_path, commit_sha):\n self.logger.info(f\"Getting file \'{file_path}\' at commit SHA: {commit_sha}\")\n url = f\"{self.base_url}/repos/{self._repo}/contents/{file_path}\"\n response = self.session.get(url, params={\"ref\": commit_sha})\n if response.status_code == 200:\n content = response.json()[\"content\"]\n decoded_content = base64.b64decode(content).decode(\'utf-8\')\n self.logger.info(f\"Successfully retrieved file \'{file_path}\' at commit {commit_sha}.\")\n return decoded_content\n else:\n error_message = f\"Error reading file \'{file_path}\' at commit {commit_sha}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _list_branches(self, per_page=100, all_pages=True):\n self.logger.info(f\"Listing branches for repo \'{self._repo}\'. Per_page={per_page}, All_pages={all_pages}\")\n url = f\"{self.base_url}/repos/{self._repo}/branches\"\n params = {\"per_page\": min(per_page, 100)} # Respect GitHub API limit\n branches_list = []\n page = 1\n while url:\n self.logger.debug(f\"Fetching page {page} from {url} with params {params if page==1 else {}}\")\n response = self.session.get(url, params=params if page == 1 else None) # params only for first page if paginating via links\n if response.status_code != 200:\n error_message = f\"Error listing branches: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n \n current_page_branches = [branch[\"name\"] for branch in response.json()]\n branches_list.extend(current_page_branches)\n self.logger.debug(f\"Fetched {len(current_page_branches)} branches on page {page}.\")\n\n if not all_pages or not response.links.get(\"next\"):\n break\n url = response.links[\"next\"][\"url\"]\n page += 1\n params = {} # Clear params for subsequent calls using a link that includes them\n\n self.logger.info(f\"Successfully listed {len(branches_list)} branches.\")\n return branches_list\n\n def _approve_pull_request(self, pull_number):\n self.logger.info(f\"Approving pull request #{pull_number}\")\n url = f\"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/reviews\"\n data = {\"event\": \"APPROVE\"}\n response = self.session.post(url, json=data)\n if response.status_code == 200:\n success_message = f\"Pull request #{pull_number} approved successfully.\"\n self.logger.info(success_message)\n return success_message\n else:\n error_message = f\"Error approving pull request #{pull_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _close_pull_request(self, pull_number):\n self.logger.info(f\"Closing pull request #{pull_number}\")\n url = f\"{self.base_url}/repos/{self._repo}/pulls/{pull_number}\"\n data = {\"state\": \"closed\"}\n response = self.session.patch(url, json=data) # Use PATCH for update\n if response.status_code == 200:\n success_message = f\"Pull request #{pull_number} closed successfully.\"\n self.logger.info(success_message)\n return success_message\n else:\n error_message = f\"Error closing pull request #{pull_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _merge_pull_request(self, pull_number, commit_title=\"Merge pull request\", commit_message=\"\", merge_method=\"merge\"):\n self.logger.info(f\"Merging pull request #{pull_number} using method \'{merge_method}\'\")\n url = f\"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/merge\"\n data = {\"commit_title\": commit_title, \"commit_message\": commit_message, \"merge_method\": merge_method}\n response = self.session.put(url, json=data)\n if response.status_code == 200:\n success_message = f\"Pull request #{pull_number} merged successfully.\"\n self.logger.info(success_message)\n return success_message\n elif response.status_code == 405: # Method Not Allowed (e.g., PR not mergeable)\n error_message = f\"Error merging pull request #{pull_number}: Not mergeable. {response.json().get(\'message\', response.text)}\"\n self.logger.warning(error_message)\n return error_message\n elif response.status_code == 409: # Conflict\n error_message = f\"Error merging pull request #{pull_number}: Merge conflict. {response.json().get(\'message\', response.text)}\"\n self.logger.warning(error_message)\n return error_message\n else:\n error_message = f\"Error merging pull request #{pull_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _delete_branch(self, branch_name):\n self.logger.info(f\"Deleting branch: {branch_name}\")\n if branch_name == \"main\" or (hasattr(self, \'default_branch\') and branch_name == self.default_branch) :\n # Add a check for a configurable default branch if necessary\n error_message = f\"Cannot delete protected branch: {branch_name}\"\n self.logger.warning(error_message)\n return error_message\n\n url = f\"{self.base_url}/repos/{self._repo}/git/refs/heads/{branch_name}\"\n response = self.session.delete(url)\n if response.status_code == 204:\n success_message = f\"Branch \'{branch_name}\' deleted successfully.\"\n self.logger.info(success_message)\n if self.current_branch == branch_name:\n self.current_branch = \"main\" # Or some other default\n self.logger.info(f\"Current branch was {branch_name}, reset to {self.current_branch}.\")\n return success_message\n else:\n error_message = f\"Error deleting branch \'{branch_name}\': {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _get_issue_details(self, issue_number):\n self.logger.info(f\"Getting details for issue #{issue_number}\")\n url = f\"{self.base_url}/repos/{self._repo}/issues/{issue_number}\"\n response = self.session.get(url)\n if response.status_code == 200:\n self.logger.info(f\"Successfully retrieved details for issue #{issue_number}.\")\n return response.json() # Return raw JSON data for now\n else:\n error_message = f\"Error getting details for issue #{issue_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _create_issue(self, title, body, labels=None):\n self.logger.info(f\"Creating new issue with title: \'{title}\'\")\n url = f\"{self.base_url}/repos/{self._repo}/issues\"\n data = {\"title\": title, \"body\": body}\n if labels: # Ensure labels is a list of strings\n data[\"labels\"] = labels if isinstance(labels, list) else [labels]\n \n response = self.session.post(url, json=data)\n if response.status_code == 201:\n issue_html_url = response.json().get(\"html_url\", \"N/A\")\n issue_number = response.json().get(\"number\", \"N/A\")\n success_message = f\"Issue \'{title}\' created successfully: {issue_html_url} (Number: {issue_number})\"\n self.logger.info(success_message)\n return success_message\n else:\n error_message = f\"Error creating issue \'{title}\': {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _list_issues(self, state=\"open\", per_page=30, page=1):\n self.logger.info(f\"Listing issues with state: {state}, per_page: {per_page}, page: {page}\")\n url = f\"{self.base_url}/repos/{self._repo}/issues\"\n params = {\"state\": state, \"per_page\": per_page, \"page\": page}\n response = self.session.get(url, params=params)\n if response.status_code == 200:\n issues_data = response.json()\n self.logger.info(f\"Successfully listed {len(issues_data)} issues.\")\n # Return a summary or full data based on needs\n return [{ \"title\": i[\"title\"], \"number\": i[\"number\"], \"state\": i[\"state\"], \"url\": i[\"html_url\"] } for i in issues_data]\n else:\n error_message = f\"Error listing issues: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _add_issue_comment(self, issue_number, comment):\n self.logger.info(f\"Adding comment to issue #{issue_number}: \'{comment[:50]}...\'\")\n url = f\"{self.base_url}/repos/{self._repo}/issues/{issue_number}/comments\"\n data = {\"body\": comment}\n response = self.session.post(url, json=data)\n if response.status_code == 201:\n comment_html_url = response.json().get(\"html_url\", \"N/A\")\n success_message = f\"Comment added to issue #{issue_number} successfully: {comment_html_url}\"\n self.logger.info(success_message)\n return success_message\n else:\n error_message = f\"Error adding comment to issue #{issue_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _get_issue_comments(self, issue_number):\n self.logger.info(f\"Getting comments for issue #{issue_number}\")\n url = f\"{self.base_url}/repos/{self._repo}/issues/{issue_number}/comments\"\n response = self.session.get(url)\n if response.status_code == 200:\n comments_data = response.json()\n self.logger.info(f\"Successfully retrieved {len(comments_data)} comments for issue #{issue_number}.\")\n # Return summary or full data\n return [{ \"user\": c[\"user\"][\"login\"], \"body\": c[\"body\"], \"created_at\": c[\"created_at\"] } for c in comments_data]\n else:\n error_message = f\"Error getting comments for issue #{issue_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _get_pull_request_general_comments(self, pull_number):\n self.logger.info(f\"Getting general comments for pull request #{pull_number}\")\n # In GitHub API, PR comments (general, not review comments on lines) are issue comments.\n # The PR is also an issue, so use the issue comments endpoint.\n return self._get_issue_comments(issue_number=pull_number)\n\n def _create_project_board(self, name, body=None):\n self.logger.info(f\"Creating project board: \'{name}\'\")\n url = f\"{self.base_url}/repos/{self._repo}/projects\"\n headers = self.session.headers.copy() # Get existing session headers\n headers[\"Accept\"] = \"application/vnd.github.inertia-preview+json\" # Required for Projects API\n data = {\"name\": name}\n if body: data[\"body\"] = body\n response = self.session.post(url, headers=headers, json=data)\n if response.status_code == 201:\n project_data = response.json()\n success_message = f\"Project board \'{name}\' created successfully with ID: {project_data[\'id\']}\"\n self.logger.info(success_message)\n return project_data # Return full project data\n else:\n error_message = f\"Error creating project board \'{name}\': {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _create_project_column(self, project_id, column_name):\n self.logger.info(f\"Creating column \'{column_name}\' for project ID: {project_id}\")\n url = f\"{self.base_url}/projects/{project_id}/columns\"\n headers = self.session.headers.copy()\n headers[\"Accept\"] = \"application/vnd.github.inertia-preview+json\"\n data = {\"name\": column_name}\n response = self.session.post(url, headers=headers, json=data)\n if response.status_code == 201:\n column_data = response.json()\n success_message = f\"Column \'{column_name}\' created successfully for project {project_id} with ID: {column_data[\'id\']}\"\n self.logger.info(success_message)\n return column_data\n else:\n error_message = f\"Error creating column \'{column_name}\' for project {project_id}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _create_project_card(self, column_id, note=None, content_id=None, content_type=None):\n self.logger.info(f\"Creating card in column ID: {column_id}\")\n url = f\"{self.base_url}/projects/columns/{column_id}/cards\"\n headers = self.session.headers.copy()\n headers[\"Accept\"] = \"application/vnd.github.inertia-preview+json\"\n data = {}\n if note:\n data[\"note\"] = note\n if content_id and content_type:\n data[\"content_id\"] = content_id\n data[\"content_type\"] = content_type\n elif (content_id and not content_type) or (not content_id and content_type):\n err = \"Both content_id and content_type must be provided to link content to a project card.\"\n self.logger.warning(err)\n return err\n \n if not data:\n return \"Error: Card must have a note or content to link.\"\n\n response = self.session.post(url, headers=headers, json=data)\n if response.status_code == 201:\n card_data = response.json()\n success_message = f\"Card created successfully in column {column_id} with ID: {card_data[\'id\']}\"\n self.logger.info(success_message)\n return card_data\n else:\n error_message = f\"Error creating card in column {column_id}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _move_project_card(self, card_id, position, column_id=None):\n self.logger.info(f\"Moving card ID: {card_id} to position: {position}\" + (f\" in column ID: {column_id}\" if column_id else \"\"))\n url = f\"{self.base_url}/projects/columns/cards/{card_id}/moves\"\n headers = self.session.headers.copy()\n headers[\"Accept\"] = \"application/vnd.github.inertia-preview+json\"\n data = {\"position\": position}\n if column_id:\n data[\"column_id\"] = column_id\n \n response = self.session.post(url, headers=headers, json=data)\n if response.status_code == 201: # Successful move returns 201 with empty body\n success_message = f\"Card {card_id} moved successfully to position {position}\" + (f\" in column {column_id}\" if column_id else \".\")\n self.logger.info(success_message)\n return success_message # Return success message as body is empty\n else:\n error_message = f\"Error moving card {card_id}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n # _link_issue_to_project_card is effectively handled by _create_project_card if content_id and content_type are passed.\n # The API used to have a separate link endpoint, but now it is part of card creation/update.\n # For updating an existing card to link an issue, one would PATCH the card\'s content_id/content_type.\n # Let\'s assume the function intends to update an existing card if it\'s a separate function.\n # However, the provided API spec for `link_issue_to_project_card` uses PATCH on card_id, so let\'s implement that.\n def _link_issue_to_project_card(self, card_id, content_id, content_type):\n self.logger.info(f\"Linking content_id {content_id} (type: {content_type}) to card_id {card_id}\")\n url = f\"{self.base_url}/projects/cards/{card_id}\" # Note: API docs suggest /projects/columns/cards/{card_id} or /projects/cards/{card_id}\n # Using /projects/cards/{card_id} as it seems more general for card update.\n headers = self.session.headers.copy()\n headers[\"Accept\"] = \"application/vnd.github.inertia-preview+json\"\n data = {\"content_id\": content_id, \"content_type\": content_type}\n\n response = self.session.patch(url, headers=headers, json=data)\n if response.status_code == 200:\n updated_card = response.json()\n success_message = f\"{content_type} {content_id} linked to card {card_id} successfully.\"\n self.logger.info(success_message)\n return updated_card\n else:\n error_message = f\"Error linking {content_type} {content_id} to card {card_id}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _list_project_boards(self):\n self.logger.info(f\"Listing project boards for repo: {self._repo}\")\n url = f\"{self.base_url}/repos/{self._repo}/projects\"\n headers = self.session.headers.copy()\n headers[\"Accept\"] = \"application/vnd.github.inertia-preview+json\"\n response = self.session.get(url, headers=headers)\n if response.status_code == 200:\n projects_data = response.json()\n self.logger.info(f\"Successfully listed {len(projects_data)} project boards.\")\n return projects_data\n else:\n error_message = f\"Error listing project boards: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _view_project_board_items(self, project_id):\n self.logger.info(f\"Viewing items for project ID: {project_id}\")\n columns_url = f\"{self.base_url}/projects/{project_id}/columns\"\n headers = self.session.headers.copy()\n headers[\"Accept\"] = \"application/vnd.github.inertia-preview+json\"\n \n columns_response = self.session.get(columns_url, headers=headers)\n if columns_response.status_code != 200:\n error_message = f\"Error fetching columns for project {project_id}: {columns_response.status_code} - {columns_response.text}\"\n self.logger.error(error_message)\n return error_message\n \n columns_data = columns_response.json()\n project_items = []\n for column in columns_data:\n column_info = {\"id\": column[\"id\"], \"name\": column[\"name\"], \"cards\": []}\n cards_url = column[\"cards_url\"]\n cards_response = self.session.get(cards_url, headers=headers)\n if cards_response.status_code == 200:\n column_info[\"cards\"] = cards_response.json()\n else:\n self.logger.error(f\"Error fetching cards for column {column[\'id\']}(\'{column[\'name\']}\'): {cards_response.status_code} - {cards_response.text}\")\n column_info[\"cards\"] = \"Error fetching cards\"\n project_items.append(column_info)\n \n self.logger.info(f\"Successfully retrieved items for project ID: {project_id}.\")\n return project_items\n\n def _get_pull_request_details(self, pull_number):\n self.logger.info(f\"Getting details for PR #{pull_number}\")\n url = f\"{self.base_url}/repos/{self._repo}/pulls/{pull_number}\"\n response = self.session.get(url)\n if response.status_code == 200:\n self.logger.info(f\"Successfully retrieved details for PR #{pull_number}.\")\n return response.json()\n else:\n error_message = f\"Error getting details for PR #{pull_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _get_pull_request_diff(self, pull_number):\n self.logger.info(f\"Getting diff for PR #{pull_number}\")\n url = f\"{self.base_url}/repos/{self._repo}/pulls/{pull_number}\"\n diff_headers = self.session.headers.copy()\n diff_headers[\"Accept\"] = \"application/vnd.github.diff\"\n response = self.session.get(url, headers=diff_headers)\n if response.status_code == 200:\n self.logger.info(f\"Successfully retrieved diff for PR #{pull_number}.\")\n return response.text\n else:\n error_message = f\"Error getting diff for PR #{pull_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _get_pull_request_files(self, pull_number):\n self.logger.info(f\"Getting files for PR #{pull_number}\")\n url = f\"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/files\"\n response = self.session.get(url)\n if response.status_code == 200:\n self.logger.info(f\"Successfully retrieved files for PR #{pull_number}.\")\n return response.json()\n else:\n error_message = f\"Error getting files for PR #{pull_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _create_pull_request_review_comment(self, pull_number, body, commit_id, path, position, side=\"RIGHT\", start_line=None, start_side=None):\n self.logger.info(f\"Creating review comment on PR #{pull_number}, file \'{path}\', position {position}\")\n url = f\"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/comments\"\n data = {\"body\": body, \"commit_id\": commit_id, \"path\": path, \"position\": position, \"side\": side}\n if start_line is not None: data[\"start_line\"] = start_line\n if start_side is not None: data[\"start_side\"] = start_side\n \n response = self.session.post(url, json=data)\n if response.status_code == 201:\n comment_url = response.json().get(\"html_url\", \"N/A\")\n success_message = f\"Review comment created successfully on PR #{pull_number}: {comment_url}\"\n self.logger.info(success_message)\n return success_message\n else:\n error_message = f\"Error creating review comment on PR #{pull_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _list_pull_request_review_comments(self, pull_number):\n self.logger.info(f\"Listing review comments for PR #{pull_number}\")\n url = f\"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/comments\"\n response = self.session.get(url)\n if response.status_code == 200:\n self.logger.info(f\"Successfully retrieved review comments for PR #{pull_number}.\")\n return response.json()\n else:\n error_message = f\"Error listing review comments for PR #{pull_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n\n def _submit_pull_request_review(self, pull_number, event, body=None):\n self.logger.info(f\"Submitting \'{event}\' review for PR #{pull_number}\")\n url = f\"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/reviews\"\n data = {\"event\": event.upper()} # Ensure event is uppercase as per API\n if body: data[\"body\"] = body\n \n response = self.session.post(url, json=data)\n if response.status_code == 200:\n review_url = response.json().get(\"html_url\", \"N/A\")\n success_message = f\"Review ({event}) submitted successfully for PR #{pull_number}: {review_url}\"\n self.logger.info(success_message)\n return success_message\n else:\n error_message = f\"Error submitting review for PR #{pull_number}: {response.status_code} - {response.text}\"\n self.logger.error(error_message)\n return error_message\n \ No newline at end of file +# tools/github_tool.py +from .base_tool import BaseTool +import requests +import os +import base64 +import logging + +class GitHubTool(BaseTool): + def __init__(self, session=None, token=None, repo=None, base_url=None, initial_branch="main", logger=None): + self.base_url = base_url if base_url else "https://api.github.com" + self._token = token if token else os.environ.get("GITHUB_TOKEN") + self._repo = repo if repo else os.environ.get("GITHUB_REPOSITORY") + + if not self._token: + # In a real scenario, might raise an error or operate in a degraded mode. + # For this tool, token is essential. + raise ValueError("GitHub token must be provided either as an argument or via GITHUB_TOKEN env var.") + if not self._repo: + raise ValueError("GitHub repository (e.g., 'owner/repo') must be provided either as an argument or via GITHUB_REPOSITORY env var.") + + if session: + self.session = session + else: + self.session = requests.Session() + self.session.headers.update({ + "Authorization": f"token {self._token}", + "Accept": "application/vnd.github.v3+json" + }) + + self.current_branch = initial_branch + + # Use provided logger or get a new one for the module + # The application using this tool should configure the logging handlers and formatting. + self.logger = logger if logger else logging.getLogger(__name__) + # If no handlers are configured by the application, add a NullHandler + # to prevent "No handler found" warnings if the tool logs something. + if not self.logger.handlers: + self.logger.addHandler(logging.NullHandler()) + + def clear(self): + if self.current_branch != "main": + self._set_current_branch("main") + self.logger.info(f"GitHubTool state cleared. Current branch is {self.current_branch}") + + def get_functions(self): + return [ + { + "type": "function", + "function": { + "name": "read_file", + "description": "Read a file from the repository", + "parameters": { + "type": "object", + "properties": { + "path": {"type": "string", "description": "Path to the file in the repository"} + }, + "required": ["path"] + } + }, + "_tags": ["read"] + }, + { + "type": "function", + "function": { + "name": "read_readme", + "description": "Read the README.md file from the root of the repository", + "parameters": { + "type": "object", + "properties": {} + } + }, + "_tags": ["read", "communicate"] + }, + { + "type": "function", + "function": { + "name": "list_files", + "description": "List files in a directory of the repository", + "parameters": { + "type": "object", + "properties": { + "path": {"type": "string", "description": "Path to the directory in the repository"} + }, + "required": ["path"] + } + }, + "_tags": ["read"] + }, + { + "type": "function", + "function": { + "name": "search_code", + "description": "Search for code in the repository", + "parameters": { + "type": "object", + "properties": { + "query": {"type": "string", "description": "Search query"} + }, + "required": ["query"] + } + }, + "_tags": ["read"] + }, + { + "type": "function", + "function": { + "name": "create_branch", + "description": "Create a new branch in the repository", + "parameters": { + "type": "object", + "properties": { + "branch_name": {"type": "string", "description": "Name of the new branch"}, + "base_branch": {"type": "string", "description": "Name of the base branch", "default": "main"} + }, + "required": ["branch_name"] + } + }, + "_tags": ["write"] + }, + { + "type": "function", + "function": { + "name": "commit_file", + "description": "Commit a file to a branch (not main)", + "parameters": { + "type": "object", + "properties": { + "file_path": {"type": "string", "description": "Path to the file in the repository"}, + "commit_message": {"type": "string", "description": "Commit message"}, + "content": {"type": "string", "description": "Content of the file"} + }, + "required": ["file_path", "commit_message", "content"] + } + }, + "_tags": ["write"] + }, + { + "type": "function", + "function": { + "name": "create_pull_request", + "description": "Create a pull request", + "parameters": { + "type": "object", + "properties": { + "title": {"type": "string", "description": "Title of the pull request"}, + "body": {"type": "string", "description": "Body of the pull request"}, + "base": {"type": "string", "description": "The name of the branch you want the changes pulled into", "default": "main"} + }, + "required": ["title", "body"] + } + }, + "_tags": ["write"] + }, + { + "type": "function", + "function": { + "name": "get_commit_history", + "description": "Get commit history for a file", + "parameters": { + "type": "object", + "properties": { + "file_path": {"type": "string", "description": "Path to the file in the repository"}, + "num_commits": {"type": "integer", "description": "Number of commits to retrieve", "default": 10} + }, + "required": ["file_path"] + } + }, + "_tags": ["read"] + }, + { + "type": "function", + "function": { + "name": "view_commit_details_for_file", + "description": "View commit history and details for a specific file, including commit messages.", + "parameters": { + "type": "object", + "properties": { + "file_path": {"type": "string", "description": "Path to the file in the repository"}, + "num_commits": {"type": "integer", "description": "Number of commits to retrieve", "default": 10} + }, + "required": ["file_path"] + } + }, + "_tags": ["read"] + }, + { + "type": "function", + "function": { + "name": "get_branch_sha", + "description": "Get the SHA of the latest commit on a branch", + "parameters": { + "type": "object", + "properties": { + "branch": {"type": "string", "description": "Name of the branch"} + }, + "required": ["branch"] + } + }, + "_tags": ["read"] + }, + { + "type": "function", + "function": { + "name": "get_current_branch", + "description": "Get the name of the current branch", + "parameters": {"type": "object", "properties": {}} + }, + "_tags": ["read"] + }, + { + "type": "function", + "function": { + "name": "set_current_branch", + "description": "Set the current branch", + "parameters": { + "type": "object", + "properties": { + "branch_name": {"type": "string", "description": "Name of the branch to set as current"} + }, + "required": ["branch_name"] + } + }, + "_tags": ["read", "write"] + }, + { + "type": "function", + "function": { + "name": "get_file_at_commit", + "description": "Get the contents of a file at a specific commit", + "parameters": { + "type": "object", + "properties": { + "file_path": {"type": "string", "description": "Path to the file in the repository"}, + "commit_sha": {"type": "string", "description": "SHA of the commit to retrieve the file from"} + }, + "required": ["file_path", "commit_sha"] + } + }, + "_tags": ["read"] + }, + { + "type": "function", + "function": { + "name": "list_branches", + "description": "List all branches in the repository", + "parameters": { + "type": "object", + "properties": { + "per_page": {"type": "integer", "description": "Number of branches to return per page (max 100)", "default": 100}, + "all_pages": {"type": "boolean", "description": "Whether to fetch all pages of results", "default": True} + } + } + }, + "_tags": ["read"] + }, + { + "type": "function", + "function": { + "name": "approve_pull_request", + "description": "Approve a pull request", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request"} + }, + "required": ["pull_number"] + } + }, + "_tags": ["write"] + }, + { + "type": "function", + "function": { + "name": "close_pull_request", + "description": "Close a pull request", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request"} + }, + "required": ["pull_number"] + } + }, + "_tags": ["write"] + }, + { + "type": "function", + "function": { + "name": "merge_pull_request", + "description": "Merge a pull request", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request"}, + "commit_title": {"type": "string", "description": "Title for the automatic commit message", "default": "Merge pull request"}, + "commit_message": {"type": "string", "description": "Extra detail to append to automatic commit message", "default": ""}, + "merge_method": { + "type": "string", + "description": "Merge method to use", + "enum": ["merge", "squash", "rebase"], + "default": "merge" + } + }, + "required": ["pull_number"] + } + }, + "_tags": ["write"] + }, + { + "type": "function", + "function": { + "name": "delete_branch", + "description": "Delete a branch", + "parameters": { + "type": "object", + "properties": { + "branch_name": {"type": "string", "description": "Name of the branch to delete"} + }, + "required": ["branch_name"] + } + }, + "_tags": ["write"] + }, + { + "type": "function", + "function": { + "name": "get_issue_details", + "description": "Get details of a specific issue", + "parameters": { + "type": "object", + "properties": { + "issue_number": {"type": "integer", "description": "The number of the issue"} + }, + "required": ["issue_number"] + } + }, + "_tags": ["read", "communicate"] + }, + { + "type": "function", + "function": { + "name": "create_issue", + "description": "Create a new issue in the repository", + "parameters": { + "type": "object", + "properties": { + "title": {"type": "string", "description": "Title of the issue"}, + "body": {"type": "string", "description": "Body of the issue"}, + "labels": { + "type": "array", + "items": {"type": "string"}, + "description": "Labels to apply to the issue" + } + }, + "required": ["title", "body"] + } + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "list_issues", + "description": "List issues in the repository", + "parameters": { + "type": "object", + "properties": { + "state": {"type": "string", "enum": ["open", "closed", "all"], "default": "open", "description": "State of the issues to retrieve"}, + "per_page": {"type": "integer", "default": 30, "description": "Number of issues to return per page"}, + "page": {"type": "integer", "default": 1, "description": "Page number of the results to fetch"} + } + } + }, + "_tags": ["read", "communicate"] + }, + { + "type": "function", + "function": { + "name": "add_issue_comment", + "description": "Add a comment to an issue", + "parameters": { + "type": "object", + "properties": { + "issue_number": {"type": "integer", "description": "The number of the issue"}, + "comment": {"type": "string", "description": "The comment to add to the issue"} + }, + "required": ["issue_number", "comment"] + } + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "get_issue_comments", + "description": "Get comments for an issue", + "parameters": { + "type": "object", + "properties": { + "issue_number": {"type": "integer", "description": "The number of the issue"} + }, + "required": ["issue_number"] + } + }, + "_tags": ["read", "communicate"] + }, + { + "type": "function", + "function": { + "name": "get_pull_request_general_comments", + "description": "Get general comments posted on a pull request itself (not specific to file lines).", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request."} + }, + "required": ["pull_number"] + } + }, + "_tags": ["read", "communicate"] + }, + { + "type": "function", + "function": { + "name": "create_project_board", + "description": "Create a new project board", + "parameters": { + "type": "object", + "properties": { + "name": {"type": "string", "description": "Name of the project board"}, + "body": {"type": "string", "description": "Body of the project board"} + }, + "required": ["name"] + } + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "create_project_column", + "description": "Create a new column in a project board", + "parameters": { + "type": "object", + "properties": { + "project_id": {"type": "integer", "description": "ID of the project board"}, + "column_name": {"type": "string", "description": "Name of the column"} + }, + "required": ["project_id", "column_name"] + } + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "create_project_card", + "description": "Create a new card in a project column", + "parameters": { + "type": "object", + "properties": { + "column_id": {"type": "integer", "description": "ID of the project column"}, + "note": {"type": "string", "description": "Note for the project card"} + }, + "required": ["column_id", "note"] + } + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "move_project_card", + "description": "Move a card to a new position", + "parameters": { + "type": "object", + "properties": { + "card_id": {"type": "integer", "description": "ID of the project card"}, + "position": {"type": "string", "description": "New position of the card"}, + "column_id": {"type": "integer", "description": "ID of the target column"} + }, + "required": ["card_id", "position", "column_id"] + } + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "link_issue_to_project_card", + "description": "Link an issue or pull request to a project card", + "parameters": { + "type": "object", + "properties": { + "card_id": {"type": "integer", "description": "ID of the project card"}, + "content_id": {"type": "integer", "description": "ID of the issue or pull request"}, + "content_type": {"type": "string", "description": "Type of the content (Issue or PullRequest)"} + }, + "required": ["card_id", "content_id", "content_type"] + } + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "list_project_boards", + "description": "List project boards associated with the repository", + "parameters": {"type": "object", "properties": {}} + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "view_project_board_items", + "description": "View items (columns and cards) in a specific project board", + "parameters": { + "type": "object", + "properties": { + "project_id": {"type": "integer", "description": "ID of the project board"} + }, + "required": ["project_id"] + } + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "get_pull_request_details", + "description": "Get detailed information about a pull request", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request"} + }, + "required": ["pull_number"] + } + }, + "_tags": ["read", "communicate"] + }, + { + "type": "function", + "function": { + "name": "get_pull_request_diff", + "description": "Get the diff of a pull request", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request"} + }, + "required": ["pull_number"] + } + }, + "_tags": ["read", "communicate"] + }, + { + "type": "function", + "function": { + "name": "get_pull_request_files", + "description": "Get a list of files changed in a pull request, with their patch details", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request"} + }, + "required": ["pull_number"] + } + }, + "_tags": ["read", "communicate"] + }, + { + "type": "function", + "function": { + "name": "create_pull_request_review_comment", + "description": "Add a comment to a specific line of a file in a pull request review", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request"}, + "body": {"type": "string", "description": "The text of the comment"}, + "commit_id": {"type": "string", "description": "The SHA of the commit to comment on"}, + "path": {"type": "string", "description": "The path to the file being commented on"}, + "position": {"type": "integer", "description": "The line index in the diff to comment on (starting at 1)"}, + "side": {"type": "string", "enum": ["LEFT", "RIGHT"], "description": "The side of the diff to comment on (LEFT for old file, RIGHT for new file)", "default": "RIGHT"}, + "start_line": {"type": "integer", "description": "The start line of the diff if commenting on a range"}, + "start_side": {"type": "string", "enum": ["LEFT", "RIGHT"], "description": "The side of the diff for the start line"} + }, + "required": ["pull_number", "body", "commit_id", "path", "position"] + } + }, + "_tags": ["communicate"] + }, + { + "type": "function", + "function": { + "name": "list_pull_request_review_comments", + "description": "List comments on a pull request review", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request"} + }, + "required": ["pull_number"] + } + }, + "_tags": ["read", "communicate"] + }, + { + "type": "function", + "function": { + "name": "submit_pull_request_review", + "description": "Submit a formal pull request review (APPROVE, REQUEST_CHANGES, COMMENT)", + "parameters": { + "type": "object", + "properties": { + "pull_number": {"type": "integer", "description": "The number of the pull request"}, + "event": {"type": "string", "enum": ["APPROVE", "REQUEST_CHANGES", "COMMENT"], "description": "The type of review event"}, + "body": {"type": "string", "description": "The body of the review (required for REQUEST_CHANGES, optional for others)"} + }, + "required": ["pull_number", "event"] + } + }, + "_tags": ["communicate"] + } + ] + + def execute(self, function_name, **kwargs): + self.logger.info(f"Executing GitHub Tool function: {function_name} with args: {kwargs}") + # Dispatch to the appropriate private method + method_name = f"_{function_name}" + if hasattr(self, method_name): + method = getattr(self, method_name) + try: + return method(**kwargs) # Ensure only expected args are passed if method signature is strict + except Exception as e: + self.logger.error(f"Error executing {method_name}: {e}", exc_info=True) + return f"Error during {function_name} execution: {str(e)}" + else: + error_message = f"Unknown function: {function_name}" + self.logger.error(error_message) + return error_message + + # Private methods for each function, using self.session for HTTP requests + + def _read_file(self, path): + self.logger.info(f"Reading file: {path} from branch: {self.current_branch}") + url = f"{self.base_url}/repos/{self._repo}/contents/{path}" + response = self.session.get(url, params={"ref": self.current_branch}) + if response.status_code == 200: + content = response.json()["content"] + decoded_content = base64.b64decode(content).decode('utf-8') + self.logger.info(f"Successfully read file: {path}") + return decoded_content + else: + error_message = f"Error reading file ({path}): {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _read_readme(self): + self.logger.info("Reading README.md from the root of the repository.") + return self._read_file("README.md") + + def _create_branch(self, branch_name, base_branch="main"): + self.logger.info(f"Creating branch: {branch_name} from base: {base_branch}") + # Get SHA of base branch + ref_url = f"{self.base_url}/repos/{self._repo}/git/refs/heads/{base_branch}" + response_sha = self.session.get(ref_url) + if response_sha.status_code != 200: + error_message = f"Error getting base branch SHA ({base_branch}): {response_sha.status_code} - {response_sha.text}" + self.logger.error(error_message) + return error_message + sha = response_sha.json()["object"]["sha"] + + # Create new branch + create_ref_url = f"{self.base_url}/repos/{self._repo}/git/refs" + data = {"ref": f"refs/heads/{branch_name}", "sha": sha} + response_create = self.session.post(create_ref_url, json=data) + if response_create.status_code == 201: + self.current_branch = branch_name + success_message = f"Branch '{branch_name}' created successfully from '{base_branch}' and set as current branch." + self.logger.info(success_message) + return success_message + else: + error_message = f"Error creating branch '{branch_name}': {response_create.status_code} - {response_create.text}" + self.logger.error(error_message) + return error_message + + def _commit_file(self, file_path, content, commit_message): + self.logger.info(f"Committing file: {file_path} to branch: {self.current_branch} with message: '{commit_message}'") + if self.current_branch == "main": + error_message = "Action directly to main branch is not allowed. Please create and switch to a new branch first." + self.logger.warning(error_message) + return error_message + + url = f"{self.base_url}/repos/{self._repo}/contents/{file_path}" + encoded_content = base64.b64encode(content.encode('utf-8')).decode('utf-8') + data = { + "message": commit_message, + "content": encoded_content, + "branch": self.current_branch + } + + # Check if file exists to get its SHA for update + self.logger.info(f"Checking if file '{file_path}' exists on branch '{self.current_branch}'") + get_response = self.session.get(url, params={"ref": self.current_branch}) + if get_response.status_code == 200: + data["sha"] = get_response.json()["sha"] + self.logger.info(f"File '{file_path}' exists, will update.") + elif get_response.status_code == 404: + self.logger.info(f"File '{file_path}' does not exist, will create.") + else: + error_message = f"Error checking file existence for '{file_path}': {get_response.status_code} - {get_response.text}" + self.logger.error(error_message) + return error_message + + response = self.session.put(url, json=data) + if response.status_code in [200, 201]: # 200 for update, 201 for create + commit_sha = response.json().get("commit", {}).get("sha", "N/A") + success_message = f"File '{file_path}' committed successfully to branch '{self.current_branch}'. Commit SHA: {commit_sha}" + self.logger.info(success_message) + return success_message + else: + error_message = f"Error committing file '{file_path}': {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _create_pull_request(self, title, body, base="main"): + self.logger.info(f"Creating pull request: '{title}' from branch '{self.current_branch}' to '{base}'") + if self.current_branch == base: + error_message = f"Cannot create a pull request from branch '{self.current_branch}' to itself ('{base}')." + self.logger.warning(error_message) + return error_message + + url = f"{self.base_url}/repos/{self._repo}/pulls" + data = {"title": title, "body": body, "head": self.current_branch, "base": base} + response = self.session.post(url, json=data) + if response.status_code == 201: + pr_html_url = response.json().get("html_url", "N/A") + pr_number = response.json().get("number", "N/A") + success_message = f"Pull request '{title}' created successfully: {pr_html_url} (Number: {pr_number})" + self.logger.info(success_message) + return success_message + else: + error_message = f"Error creating pull request: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _get_branch_sha(self, branch): + self.logger.info(f"Getting SHA for branch: {branch}") + url = f"{self.base_url}/repos/{self._repo}/git/refs/heads/{branch}" + response = self.session.get(url) + if response.status_code == 200: + sha = response.json()["object"]["sha"] + self.logger.info(f"SHA for branch '{branch}' is {sha}") + return sha + else: + error_message = f"Error getting SHA for branch '{branch}': {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _list_files(self, path): + self.logger.info(f"Listing files in path: '{path}' on branch: '{self.current_branch}'") + url = f"{self.base_url}/repos/{self._repo}/contents/{path.strip('/')}" # Ensure no leading/trailing slashes for consistency + response = self.session.get(url, params={"ref": self.current_branch}) + if response.status_code == 200: + items = response.json() + results = [] + if isinstance(items, list): # It's a directory listing + for item in items: + results.append({"name": item["name"], "type": item["type"], "path": item["path"]}) + elif isinstance(items, dict) and 'type' in items: # It's a single file response + results.append({"name": items["name"], "type": items["type"], "path": items["path"]}) + self.logger.info(f"Successfully listed {len(results)} items in '{path}'.") + return results + elif response.status_code == 404: + self.logger.warning(f"Path '{path}' not found on branch '{self.current_branch}'.") + return f"Error: Path '{path}' not found." + else: + error_message = f"Error listing files in '{path}': {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _search_code(self, query): + self.logger.info(f"Searching code with query: '{query}' in repo: '{self._repo}'") + url = f"{self.base_url}/search/code" + params = {"q": f"{query} repo:{self._repo}"} + response = self.session.get(url, params=params) + if response.status_code == 200: + search_results = response.json().get("items", []) + results = [{"path": item["path"], "url": item["html_url"]} for item in search_results] + self.logger.info(f"Code search for '{query}' found {len(results)} items.") + return results + else: + error_message = f"Error searching code for '{query}': {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _get_commit_history(self, file_path, num_commits=10): + self.logger.info(f"Getting last {num_commits} commit(s) for file: '{file_path}' on branch '{self.current_branch}'") + url = f"{self.base_url}/repos/{self._repo}/commits" + params = {"path": file_path, "sha": self.current_branch, "per_page": num_commits} + response = self.session.get(url, params=params) + if response.status_code == 200: + commits_data = response.json() + commits = [{ + "sha": commit["sha"], + "message": commit["commit"]["message"], + "author": commit["commit"]["author"]["name"], + "date": commit["commit"]["author"]["date"] + } for commit in commits_data] + self.logger.info(f"Successfully retrieved {len(commits)} commit(s) for '{file_path}'.") + return commits + else: + error_message = f"Error getting commit history for '{file_path}': {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _view_commit_details_for_file(self, file_path, num_commits=10): + # This function is essentially the same as get_commit_history based on its description. + self.logger.info(f"Viewing commit details for file '{file_path}' (last {num_commits} commits) - using _get_commit_history.") + return self._get_commit_history(file_path, num_commits) + + def _get_current_branch(self): + self.logger.info(f"Current branch is: {self.current_branch}") + return self.current_branch + + def _set_current_branch(self, branch_name): + self.logger.info(f"Attempting to set current branch to: {branch_name}") + # Check if branch exists by trying to get its SHA + sha_info = self._get_branch_sha(branch_name) + if isinstance(sha_info, str) and sha_info.startswith("Error getting SHA"): # Crude check for error string + error_message = f"Cannot set current branch: Branch '{branch_name}' not found or error accessing it. Details: {sha_info}" + self.logger.warning(error_message) + return error_message + + self.current_branch = branch_name + success_message = f"Current branch set to: {self.current_branch}" + self.logger.info(success_message) + return success_message + + def _get_file_at_commit(self, file_path, commit_sha): + self.logger.info(f"Getting file '{file_path}' at commit SHA: {commit_sha}") + url = f"{self.base_url}/repos/{self._repo}/contents/{file_path}" + response = self.session.get(url, params={"ref": commit_sha}) + if response.status_code == 200: + content = response.json()["content"] + decoded_content = base64.b64decode(content).decode('utf-8') + self.logger.info(f"Successfully retrieved file '{file_path}' at commit {commit_sha}.") + return decoded_content + else: + error_message = f"Error reading file '{file_path}' at commit {commit_sha}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _list_branches(self, per_page=100, all_pages=True): + self.logger.info(f"Listing branches for repo '{self._repo}'. Per_page={per_page}, All_pages={all_pages}") + url = f"{self.base_url}/repos/{self._repo}/branches" + params = {"per_page": min(per_page, 100)} # Respect GitHub API limit + branches_list = [] + page = 1 + while url: + self.logger.debug(f"Fetching page {page} from {url} with params {params if page==1 else {}}") + response = self.session.get(url, params=params if page == 1 else None) # params only for first page if paginating via links + if response.status_code != 200: + error_message = f"Error listing branches: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + current_page_branches = [branch["name"] for branch in response.json()] + branches_list.extend(current_page_branches) + self.logger.debug(f"Fetched {len(current_page_branches)} branches on page {page}.") + + if not all_pages or not response.links.get("next"): + break + url = response.links["next"]["url"] + page += 1 + params = {} # Clear params for subsequent calls using a link that includes them + + self.logger.info(f"Successfully listed {len(branches_list)} branches.") + return branches_list + + def _approve_pull_request(self, pull_number): + self.logger.info(f"Approving pull request #{pull_number}") + url = f"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/reviews" + data = {"event": "APPROVE"} + response = self.session.post(url, json=data) + if response.status_code == 200: + success_message = f"Pull request #{pull_number} approved successfully." + self.logger.info(success_message) + return success_message + else: + error_message = f"Error approving pull request #{pull_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _close_pull_request(self, pull_number): + self.logger.info(f"Closing pull request #{pull_number}") + url = f"{self.base_url}/repos/{self._repo}/pulls/{pull_number}" + data = {"state": "closed"} + response = self.session.patch(url, json=data) # Use PATCH for update + if response.status_code == 200: + success_message = f"Pull request #{pull_number} closed successfully." + self.logger.info(success_message) + return success_message + else: + error_message = f"Error closing pull request #{pull_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _merge_pull_request(self, pull_number, commit_title="Merge pull request", commit_message="", merge_method="merge"): + self.logger.info(f"Merging pull request #{pull_number} using method '{merge_method}'") + url = f"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/merge" + data = {"commit_title": commit_title, "commit_message": commit_message, "merge_method": merge_method} + response = self.session.put(url, json=data) + if response.status_code == 200: + success_message = f"Pull request #{pull_number} merged successfully." + self.logger.info(success_message) + return success_message + elif response.status_code == 405: # Method Not Allowed (e.g., PR not mergeable) + error_message = f"Error merging pull request #{pull_number}: Not mergeable. {response.json().get('message', response.text)}" + self.logger.warning(error_message) + return error_message + elif response.status_code == 409: # Conflict + error_message = f"Error merging pull request #{pull_number}: Merge conflict. {response.json().get('message', response.text)}" + self.logger.warning(error_message) + return error_message + else: + error_message = f"Error merging pull request #{pull_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _delete_branch(self, branch_name): + self.logger.info(f"Deleting branch: {branch_name}") + if branch_name == "main" or (hasattr(self, 'default_branch') and branch_name == self.default_branch): + # Add a check for a configurable default branch if necessary + error_message = f"Cannot delete protected branch: {branch_name}" + self.logger.warning(error_message) + return error_message + + url = f"{self.base_url}/repos/{self._repo}/git/refs/heads/{branch_name}" + response = self.session.delete(url) + if response.status_code == 204: + success_message = f"Branch '{branch_name}' deleted successfully." + self.logger.info(success_message) + if self.current_branch == branch_name: + self.current_branch = "main" # Or some other default + self.logger.info(f"Current branch was {branch_name}, reset to {self.current_branch}.") + return success_message + else: + error_message = f"Error deleting branch '{branch_name}': {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _get_issue_details(self, issue_number): + self.logger.info(f"Getting details for issue #{issue_number}") + url = f"{self.base_url}/repos/{self._repo}/issues/{issue_number}" + response = self.session.get(url) + if response.status_code == 200: + self.logger.info(f"Successfully retrieved details for issue #{issue_number}.") + return response.json() # Return raw JSON data for now + else: + error_message = f"Error getting details for issue #{issue_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _create_issue(self, title, body, labels=None): + self.logger.info(f"Creating new issue with title: '{title}'") + url = f"{self.base_url}/repos/{self._repo}/issues" + data = {"title": title, "body": body} + if labels: # Ensure labels is a list of strings + data["labels"] = labels if isinstance(labels, list) else [labels] + + response = self.session.post(url, json=data) + if response.status_code == 201: + issue_html_url = response.json().get("html_url", "N/A") + issue_number = response.json().get("number", "N/A") + success_message = f"Issue '{title}' created successfully: {issue_html_url} (Number: {issue_number})" + self.logger.info(success_message) + return success_message + else: + error_message = f"Error creating issue '{title}': {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _list_issues(self, state="open", per_page=30, page=1): + self.logger.info(f"Listing issues with state: {state}, per_page: {per_page}, page: {page}") + url = f"{self.base_url}/repos/{self._repo}/issues" + params = {"state": state, "per_page": per_page, "page": page} + response = self.session.get(url, params=params) + if response.status_code == 200: + issues_data = response.json() + self.logger.info(f"Successfully listed {len(issues_data)} issues.") + # Return a summary or full data based on needs + return [{"title": i["title"], "number": i["number"], "state": i["state"], "url": i["html_url"]} for i in issues_data] + else: + error_message = f"Error listing issues: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _add_issue_comment(self, issue_number, comment): + self.logger.info(f"Adding comment to issue #{issue_number}: '{comment[:50]}...'") + url = f"{self.base_url}/repos/{self._repo}/issues/{issue_number}/comments" + data = {"body": comment} + response = self.session.post(url, json=data) + if response.status_code == 201: + comment_html_url = response.json().get("html_url", "N/A") + success_message = f"Comment added to issue #{issue_number} successfully: {comment_html_url}" + self.logger.info(success_message) + return success_message + else: + error_message = f"Error adding comment to issue #{issue_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _get_issue_comments(self, issue_number): + self.logger.info(f"Getting comments for issue #{issue_number}") + url = f"{self.base_url}/repos/{self._repo}/issues/{issue_number}/comments" + response = self.session.get(url) + if response.status_code == 200: + comments_data = response.json() + self.logger.info(f"Successfully retrieved {len(comments_data)} comments for issue #{issue_number}.") + # Return summary or full data + return [{"user": c["user"]["login"], "body": c["body"], "created_at": c["created_at"]} for c in comments_data] + else: + error_message = f"Error getting comments for issue #{issue_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _get_pull_request_general_comments(self, pull_number): + self.logger.info(f"Getting general comments for pull request #{pull_number}") + # In GitHub API, PR comments (general, not review comments on lines) are issue comments. + # The PR is also an issue, so use the issue comments endpoint. + return self._get_issue_comments(issue_number=pull_number) + + def _create_project_board(self, name, body=None): + self.logger.info(f"Creating project board: '{name}'") + url = f"{self.base_url}/repos/{self._repo}/projects" + headers = self.session.headers.copy() # Get existing session headers + headers["Accept"] = "application/vnd.github.inertia-preview+json" # Required for Projects API + data = {"name": name} + if body: + data["body"] = body + response = self.session.post(url, headers=headers, json=data) + if response.status_code == 201: + project_data = response.json() + success_message = f"Project board '{name}' created successfully with ID: {project_data['id']}" + self.logger.info(success_message) + return project_data # Return full project data + else: + error_message = f"Error creating project board '{name}': {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _create_project_column(self, project_id, column_name): + self.logger.info(f"Creating column '{column_name}' for project ID: {project_id}") + url = f"{self.base_url}/projects/{project_id}/columns" + headers = self.session.headers.copy() + headers["Accept"] = "application/vnd.github.inertia-preview+json" + data = {"name": column_name} + response = self.session.post(url, headers=headers, json=data) + if response.status_code == 201: + column_data = response.json() + success_message = f"Column '{column_name}' created successfully for project {project_id} with ID: {column_data['id']}" + self.logger.info(success_message) + return column_data + else: + error_message = f"Error creating column '{column_name}' for project {project_id}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _create_project_card(self, column_id, note=None, content_id=None, content_type=None): + self.logger.info(f"Creating card in column ID: {column_id}") + url = f"{self.base_url}/projects/columns/{column_id}/cards" + headers = self.session.headers.copy() + headers["Accept"] = "application/vnd.github.inertia-preview+json" + data = {} + if note: + data["note"] = note + if content_id and content_type: + data["content_id"] = content_id + data["content_type"] = content_type + elif (content_id and not content_type) or (not content_id and content_type): + err = "Both content_id and content_type must be provided to link content to a project card." + self.logger.warning(err) + return err + + if not data: + return "Error: Card must have a note or content to link." + + response = self.session.post(url, headers=headers, json=data) + if response.status_code == 201: + card_data = response.json() + success_message = f"Card created successfully in column {column_id} with ID: {card_data['id']}" + self.logger.info(success_message) + return card_data + else: + error_message = f"Error creating card in column {column_id}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _move_project_card(self, card_id, position, column_id=None): + self.logger.info(f"Moving card ID: {card_id} to position: {position}" + (f" in column ID: {column_id}" if column_id else "")) + url = f"{self.base_url}/projects/columns/cards/{card_id}/moves" + headers = self.session.headers.copy() + headers["Accept"] = "application/vnd.github.inertia-preview+json" + data = {"position": position} + if column_id: + data["column_id"] = column_id + + response = self.session.post(url, headers=headers, json=data) + if response.status_code == 201: # Successful move returns 201 with empty body + success_message = f"Card {card_id} moved successfully to position {position}" + (f" in column {column_id}" if column_id else ".") + self.logger.info(success_message) + return success_message # Return success message as body is empty + else: + error_message = f"Error moving card {card_id}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + # _link_issue_to_project_card is effectively handled by _create_project_card if content_id and content_type are passed. + # The API used to have a separate link endpoint, but now it is part of card creation/update. + # For updating an existing card to link an issue, one would PATCH the card's content_id/content_type. + # Let's assume the function intends to update an existing card if it's a separate function. + # However, the provided API spec for `link_issue_to_project_card` uses PATCH on card_id, so let's implement that. + def _link_issue_to_project_card(self, card_id, content_id, content_type): + self.logger.info(f"Linking content_id {content_id} (type: {content_type}) to card_id {card_id}") + url = f"{self.base_url}/projects/cards/{card_id}" # Note: API docs suggest /projects/columns/cards/{card_id} or /projects/cards/{card_id} + # Using /projects/cards/{card_id} as it seems more general for card update. + headers = self.session.headers.copy() + headers["Accept"] = "application/vnd.github.inertia-preview+json" + data = {"content_id": content_id, "content_type": content_type} + + response = self.session.patch(url, headers=headers, json=data) + if response.status_code == 200: + updated_card = response.json() + success_message = f"{content_type} {content_id} linked to card {card_id} successfully." + self.logger.info(success_message) + return updated_card + else: + error_message = f"Error linking {content_type} {content_id} to card {card_id}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _list_project_boards(self): + self.logger.info(f"Listing project boards for repo: {self._repo}") + url = f"{self.base_url}/repos/{self._repo}/projects" + headers = self.session.headers.copy() + headers["Accept"] = "application/vnd.github.inertia-preview+json" + response = self.session.get(url, headers=headers) + if response.status_code == 200: + projects_data = response.json() + self.logger.info(f"Successfully listed {len(projects_data)} project boards.") + return projects_data + else: + error_message = f"Error listing project boards: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _view_project_board_items(self, project_id): + self.logger.info(f"Viewing items for project ID: {project_id}") + columns_url = f"{self.base_url}/projects/{project_id}/columns" + headers = self.session.headers.copy() + headers["Accept"] = "application/vnd.github.inertia-preview+json" + + columns_response = self.session.get(columns_url, headers=headers) + if columns_response.status_code != 200: + error_message = f"Error fetching columns for project {project_id}: {columns_response.status_code} - {columns_response.text}" + self.logger.error(error_message) + return error_message + + columns_data = columns_response.json() + project_items = [] + for column in columns_data: + column_info = {"id": column["id"], "name": column["name"], "cards": []} + cards_url = column["cards_url"] + cards_response = self.session.get(cards_url, headers=headers) + if cards_response.status_code == 200: + column_info["cards"] = cards_response.json() + else: + self.logger.error(f"Error fetching cards for column {column['id']}('{column['name']}'): {cards_response.status_code} - {cards_response.text}") + column_info["cards"] = "Error fetching cards" + project_items.append(column_info) + + self.logger.info(f"Successfully retrieved items for project ID: {project_id}.") + return project_items + + def _get_pull_request_details(self, pull_number): + self.logger.info(f"Getting details for PR #{pull_number}") + url = f"{self.base_url}/repos/{self._repo}/pulls/{pull_number}" + response = self.session.get(url) + if response.status_code == 200: + self.logger.info(f"Successfully retrieved details for PR #{pull_number}.") + return response.json() + else: + error_message = f"Error getting details for PR #{pull_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _get_pull_request_diff(self, pull_number): + self.logger.info(f"Getting diff for PR #{pull_number}") + url = f"{self.base_url}/repos/{self._repo}/pulls/{pull_number}" + diff_headers = self.session.headers.copy() + diff_headers["Accept"] = "application/vnd.github.diff" + response = self.session.get(url, headers=diff_headers) + if response.status_code == 200: + self.logger.info(f"Successfully retrieved diff for PR #{pull_number}.") + return response.text + else: + error_message = f"Error getting diff for PR #{pull_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _get_pull_request_files(self, pull_number): + self.logger.info(f"Getting files for PR #{pull_number}") + url = f"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/files" + response = self.session.get(url) + if response.status_code == 200: + self.logger.info(f"Successfully retrieved files for PR #{pull_number}.") + return response.json() + else: + error_message = f"Error getting files for PR #{pull_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _create_pull_request_review_comment(self, pull_number, body, commit_id, path, position, side="RIGHT", start_line=None, start_side=None): + self.logger.info(f"Creating review comment on PR #{pull_number}, file '{path}', position {position}") + url = f"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/comments" + data = {"body": body, "commit_id": commit_id, "path": path, "position": position, "side": side} + if start_line is not None: + data["start_line"] = start_line + if start_side is not None: + data["start_side"] = start_side + + response = self.session.post(url, json=data) + if response.status_code == 201: + comment_url = response.json().get("html_url", "N/A") + success_message = f"Review comment created successfully on PR #{pull_number}: {comment_url}" + self.logger.info(success_message) + return success_message + else: + error_message = f"Error creating review comment on PR #{pull_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _list_pull_request_review_comments(self, pull_number): + self.logger.info(f"Listing review comments for PR #{pull_number}") + url = f"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/comments" + response = self.session.get(url) + if response.status_code == 200: + self.logger.info(f"Successfully retrieved review comments for PR #{pull_number}.") + return response.json() + else: + error_message = f"Error listing review comments for PR #{pull_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message + + def _submit_pull_request_review(self, pull_number, event, body=None): + self.logger.info(f"Submitting '{event}' review for PR #{pull_number}") + url = f"{self.base_url}/repos/{self._repo}/pulls/{pull_number}/reviews" + data = {"event": event.upper()} # Ensure event is uppercase as per API + if body: + data["body"] = body + + response = self.session.post(url, json=data) + if response.status_code == 200: + review_url = response.json().get("html_url", "N/A") + success_message = f"Review ({event}) submitted successfully for PR #{pull_number}: {review_url}" + self.logger.info(success_message) + return success_message + else: + error_message = f"Error submitting review for PR #{pull_number}: {response.status_code} - {response.text}" + self.logger.error(error_message) + return error_message \ No newline at end of file diff --git a/tools/standalone_llm_tool.py b/tools/standalone_llm_tool.py index 41d8c0d..6df2e0b 100644 --- a/tools/standalone_llm_tool.py +++ b/tools/standalone_llm_tool.py @@ -3,6 +3,7 @@ import os import json import logging from openai import OpenAI +import re import urllib.request import urllib.error @@ -82,10 +83,12 @@ class StandaloneLLMTool(BaseTool): headers={'Content-Type': 'text/plain; charset=utf-8', 'User-Agent': 'DualAICopilot/0.1'}, method='POST' ) - with urllib.request.urlopen(req, timeout=500) as response: + with urllib.request.urlopen(req, timeout=3600) as response: if response.status == 200: response_data = response.read().decode('utf-8') logging.info(f"Received response from external copilot: {response_data[:100]}...") + # Remove content within tags + response_data = re.sub(r".*?", "", response_data, flags=re.DOTALL) return response_data else: error_message = f"External copilot at {self.copilot_url} returned an error: {response.status} {response.reason}"