Method _run

src/hyperagent/tools/nav_tools.py:335–467

Opens the specified file and returns its content. Args: relative_file_path (str): The relative path to the file. max_new_line (int, optional): The maximum number of lines to include in the returned content. Defaults to 500. Returns: str:

(self, relative_file_path: str, start_line: Optional[int] = None, end_line: Optional[int] = None, keywords: Optional[List[str]] = [], preview_size: int = 8, max_num_result: int = 5, semantic_query: str = "")

        self.model = AutoModel.from_pretrained('jinaai/jina-embeddings-v2-base-code', trust_remote_code=True)
    
    def _run(self, relative_file_path: str, start_line: Optional[int] = None, end_line: Optional[int] = None, keywords: Optional[List[str]] = [], preview_size: int = 8, max_num_result: int = 5, semantic_query: str = ""):
        """
        Opens the specified file and returns its content.

        Args:
            relative_file_path (str): The relative path to the file.
            max_new_line (int, optional): The maximum number of lines to include in the returned content. Defaults to 500.

        Returns:
            str: The content of the file.

        """
        if len(keywords) == 0 and start_line is None and end_line is None:
            return "Please specify the keyword or start and end line to view the content of the file."
        
        abs_path = find_matching_abs_path(self.path, relative_file_path)
        # abs_path = os.path.join(self.path, relative_file_path)
        source = open(abs_path, "r").read()
        lines = source.split("\n")

        cos_sim = lambda a,b: (a @ b.T) / (norm(a)*norm(b))

        def chunk_list(lst, chunk_size):
            return [lst[i:i + chunk_size] for i in range(0, len(lst), chunk_size)]
        
        if semantic_query != "":
            embeddings = self.modle.encode(
                [
                    "\n".join(segment_lst) for segment_lst in chunk_list(lines, 50)
                ]
            )

            query = self.model.encode([semantic_query])
            similarities = []
            for emb in embeddings:
                similarities.append(cos_sim(query, emb))
            
            sorted_idx = sorted(range(len(similarities)), key=lambda x: similarities[x], reverse=True)

            line_ranges = sorted_idx[:max_num_result]
            line_ranges = [(i*50, (i+1)*50) for i in line_ranges]
            returned_source = []
            for start, end in line_ranges:
                expanded_source = "\n".join(lines[start:end])
                expanded_source = add_num_line(expanded_source, start+1)
                returned_source.append(expanded_source)
            
            semantic_query_result =  "\n--------------\n".join(returned_source)

        try:
            if start_line is not None and end_line is not None and len(keywords) == 0:
                if end_line - start_line > 90:
                    return f"The number of lines to show is limited at 90, the requested number of lines is {end_line - start_line}, please specify the start and end line again or using keyword instead. For example {start_line}:{start_line+90}"
                
                if start_line > len(lines):
                    return f"Invalid start line, the start line is greater than the total number of lines in the file, the total number of lines in the file is {len(lines)}"
                
                source = "\n".join(lines[start_line:end_line])