analyze_character_relations
Analyze relationships between characters in a play by specifying the corpus and play name. Identify connections and networks within dramatic texts using the Drama Corpora Project API.
Instructions
Analyze the character relationships in a play.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| corpus_name | Yes | ||
| play_name | Yes |
Implementation Reference
- dracor_mcp_fastmcp.py:443-542 (handler)The handler function for the 'analyze_character_relations' tool. It fetches play and character data from DraCor API, parses network CSV data to extract and analyze character relations by weight, attempts to get formal relations, sorts relations by strength, and returns a structured analysis including top/weakest relations, metrics, and play info.@mcp.tool() def analyze_character_relations(corpus_name: str, play_name: str) -> Dict: """Analyze the character relationships in a play.""" try: # Get play data play = api_request(f"corpora/{corpus_name}/plays/{play_name}") # Get character data characters = api_request(f"corpora/{corpus_name}/plays/{play_name}/characters") # Get network data in CSV format url = f"{DRACOR_API_BASE_URL}/corpora/{corpus_name}/plays/{play_name}/networkdata/csv" response = requests.get(url) response.raise_for_status() csv_data = response.text # Parse CSV data to extract relations relations = [] lines = csv_data.strip().split('\n') if len(lines) > 1: # Skip header headers = lines[0].split(',') for line in lines[1:]: parts = line.split(',') if len(parts) >= 4: source = parts[0].strip('"') target = parts[2].strip('"') weight = int(parts[3]) if parts[3].isdigit() else 0 # Find character names from IDs source_name = None target_name = None for char in characters: if char.get("id") == source: source_name = char.get("name") if char.get("id") == target: target_name = char.get("name") relations.append({ "source": source_name or source, "source_id": source, "target": target_name or target, "target_id": target, "weight": weight }) # Sort by weight to identify strongest relationships relations.sort(key=lambda x: x.get("weight", 0), reverse=True) # Try to get relations data if available try: relations_url = f"{DRACOR_API_BASE_URL}/corpora/{corpus_name}/plays/{play_name}/relations/csv" relations_response = requests.get(relations_url) formal_relations = [] if relations_response.status_code == 200: rel_lines = relations_response.text.strip().split('\n') if len(rel_lines) > 1: # Skip header for line in rel_lines[1:]: parts = line.split(',') if len(parts) >= 4: source = parts[0].strip('"') target = parts[2].strip('"') relation_type = parts[3].strip('"') # Find character names from IDs source_name = None target_name = None for char in characters: if char.get("id") == source: source_name = char.get("name") if char.get("id") == target: target_name = char.get("name") formal_relations.append({ "source": source_name or source, "target": target_name or target, "type": relation_type }) except: formal_relations = [] # Get metrics metrics = api_request(f"corpora/{corpus_name}/plays/{play_name}/metrics") return { "play": { "title": play.get("title"), "author": play.get("authors", [{}])[0].get("name") if play.get("authors") else None, "year": play.get("yearNormalized") }, "totalCharacters": len(characters), "totalRelations": len(relations), "strongestRelations": relations[:10], # Top 10 strongest relations "weakestRelations": relations[-10:] if len(relations) >= 10 else relations, # Bottom 10 "formalRelations": formal_relations, # Explicit relations if available "metrics": metrics } except Exception as e: return {"error": str(e)}
- dracor_mcp_fastmcp.py:443-443 (registration)The @mcp.tool() decorator registers the analyze_character_relations function as an MCP tool.@mcp.tool()