ef2614a70a
- Introduced a new example script `chat_tui.py` that provides a full-screen Textual interface for interacting with FastGPT. - Implemented streaming chat updates, workflow logging, and modal handling for interactive nodes. - Enhanced FastGPT client with new streaming capabilities and structured event types for better interaction handling. - Normalized base URL handling in the client to prevent duplicate `/api` paths. - Added tests for streaming event parsing and interaction handling.
432 lines
15 KiB
Python
432 lines
15 KiB
Python
"""Interactive chat CLI with FastGPT streaming events.
|
|
|
|
Run from the examples directory with .env configured:
|
|
python chat_cli.py
|
|
python chat_cli.py --chat-id my_existing_conversation
|
|
|
|
This example supports:
|
|
- streaming text output
|
|
- workflow/tool event logs
|
|
- interactive FastGPT nodes (`userSelect` and `userInput`)
|
|
|
|
Type your message and press Enter. Type /quit to exit.
|
|
During an interactive prompt, type /cancel to stop that prompt locally.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import argparse
|
|
import json
|
|
import os
|
|
import sys
|
|
import uuid
|
|
from dataclasses import dataclass
|
|
from pathlib import Path
|
|
from typing import Any, Dict, List, Optional
|
|
|
|
from dotenv import load_dotenv
|
|
|
|
from fastgpt_client import ChatClient, FastGPTInteractiveEvent, iter_stream_events
|
|
|
|
load_dotenv(Path(__file__).with_name(".env"))
|
|
|
|
API_KEY = os.getenv("API_KEY")
|
|
BASE_URL = os.getenv("BASE_URL")
|
|
|
|
for stream in (sys.stdout, sys.stderr):
|
|
if hasattr(stream, "reconfigure"):
|
|
try:
|
|
stream.reconfigure(encoding="utf-8", errors="replace")
|
|
except Exception:
|
|
pass
|
|
|
|
|
|
@dataclass
|
|
class StreamTurnResult:
|
|
text: str
|
|
interactive: Optional[FastGPTInteractiveEvent] = None
|
|
|
|
|
|
def _write_text(text: str) -> None:
|
|
try:
|
|
sys.stdout.write(text)
|
|
except UnicodeEncodeError:
|
|
encoding = getattr(sys.stdout, "encoding", None) or "utf-8"
|
|
safe_text = text.encode(encoding, errors="replace").decode(encoding, errors="replace")
|
|
sys.stdout.write(safe_text)
|
|
sys.stdout.flush()
|
|
|
|
|
|
def _extract_text_from_event(kind: str, payload: Any) -> str:
|
|
if not isinstance(payload, dict):
|
|
return ""
|
|
|
|
if kind in {"answer", "fastAnswer"}:
|
|
text = payload.get("text")
|
|
if isinstance(text, str) and text:
|
|
return text
|
|
|
|
choices = payload.get("choices") if isinstance(payload.get("choices"), list) else []
|
|
if not choices:
|
|
return str(payload.get("text") or "")
|
|
|
|
first_choice = choices[0] if isinstance(choices[0], dict) else {}
|
|
delta = first_choice.get("delta") if isinstance(first_choice.get("delta"), dict) else {}
|
|
content = delta.get("content")
|
|
if isinstance(content, str) and content:
|
|
return content
|
|
|
|
message = first_choice.get("message") if isinstance(first_choice.get("message"), dict) else {}
|
|
message_content = message.get("content")
|
|
if isinstance(message_content, str) and message_content:
|
|
return message_content
|
|
|
|
return ""
|
|
|
|
|
|
def _nested_tool_payload(payload: Any) -> Dict[str, Any]:
|
|
if not isinstance(payload, dict):
|
|
return {}
|
|
nested = payload.get("tool")
|
|
if isinstance(nested, dict):
|
|
return nested
|
|
return {}
|
|
|
|
|
|
def _tool_name_from_event(payload: Any) -> str:
|
|
if not isinstance(payload, dict):
|
|
return "?"
|
|
nested = _nested_tool_payload(payload)
|
|
return str(payload.get("toolName") or nested.get("toolName") or payload.get("functionName") or nested.get("functionName") or "?")
|
|
|
|
|
|
def _normalize_option(raw_option: Any, index: int) -> Optional[Dict[str, str]]:
|
|
if isinstance(raw_option, str):
|
|
value = raw_option.strip()
|
|
if not value:
|
|
return None
|
|
return {"id": f"option_{index}", "label": value, "value": value, "description": ""}
|
|
|
|
if not isinstance(raw_option, dict):
|
|
return None
|
|
|
|
label = str(raw_option.get("label") or raw_option.get("value") or raw_option.get("id") or "").strip()
|
|
value = str(raw_option.get("value") or raw_option.get("label") or raw_option.get("id") or "").strip()
|
|
option_id = str(raw_option.get("id") or value or f"option_{index}").strip()
|
|
if not label and not value:
|
|
return None
|
|
return {
|
|
"id": option_id or f"option_{index}",
|
|
"label": label or value,
|
|
"value": value or label,
|
|
"description": str(raw_option.get("description") or "").strip(),
|
|
}
|
|
|
|
|
|
def _resolve_option_token(token: str, options: List[Dict[str, str]]) -> Optional[str]:
|
|
normalized = token.strip()
|
|
if not normalized:
|
|
return None
|
|
|
|
if normalized.isdigit():
|
|
index = int(normalized) - 1
|
|
if 0 <= index < len(options):
|
|
return options[index]["value"]
|
|
|
|
lowered = normalized.lower()
|
|
for option in options:
|
|
if lowered in {
|
|
option["id"].lower(),
|
|
option["label"].lower(),
|
|
option["value"].lower(),
|
|
}:
|
|
return option["value"]
|
|
return None
|
|
|
|
|
|
def _interactive_prompt_text(payload: Dict[str, Any], default_text: str) -> str:
|
|
params = payload.get("params") if isinstance(payload.get("params"), dict) else {}
|
|
return str(
|
|
payload.get("prompt")
|
|
or payload.get("title")
|
|
or payload.get("text")
|
|
or payload.get("description")
|
|
or params.get("description")
|
|
or default_text
|
|
).strip()
|
|
|
|
|
|
def _prompt_user_select(event: FastGPTInteractiveEvent) -> Optional[str]:
|
|
payload = event.data
|
|
params = payload.get("params") if isinstance(payload.get("params"), dict) else {}
|
|
prompt_text = _interactive_prompt_text(payload, "Please select an option")
|
|
multiple = bool(params.get("multiple") or payload.get("multiple"))
|
|
raw_options = params.get("userSelectOptions") if isinstance(params.get("userSelectOptions"), list) else []
|
|
options = [item for index, raw in enumerate(raw_options, start=1) if (item := _normalize_option(raw, index))]
|
|
|
|
print()
|
|
print(f"[INTERACTIVE] {prompt_text}")
|
|
for index, option in enumerate(options, start=1):
|
|
print(f" {index}. {option['label']}")
|
|
if option["description"]:
|
|
print(f" {option['description']}")
|
|
|
|
if not options:
|
|
print("No selectable options were provided by FastGPT.")
|
|
return None
|
|
|
|
while True:
|
|
hint = "comma-separated indexes/values" if multiple else "an index or value"
|
|
raw_input_value = input(f"Select {hint} (/cancel to stop): ").strip()
|
|
if not raw_input_value:
|
|
print("Selection is required.")
|
|
continue
|
|
if raw_input_value.lower() == "/cancel":
|
|
return None
|
|
|
|
tokens = [part.strip() for part in raw_input_value.split(",")] if multiple else [raw_input_value]
|
|
selected_values: List[str] = []
|
|
invalid_tokens: List[str] = []
|
|
for token in tokens:
|
|
resolved = _resolve_option_token(token, options)
|
|
if resolved is None:
|
|
invalid_tokens.append(token)
|
|
continue
|
|
selected_values.append(resolved)
|
|
|
|
if invalid_tokens:
|
|
print(f"Invalid option(s): {', '.join(invalid_tokens)}")
|
|
continue
|
|
if not selected_values:
|
|
print("Selection is required.")
|
|
continue
|
|
if not multiple:
|
|
return selected_values[0]
|
|
return ", ".join(selected_values)
|
|
|
|
|
|
def _prompt_user_input(event: FastGPTInteractiveEvent) -> Optional[str]:
|
|
payload = event.data
|
|
params = payload.get("params") if isinstance(payload.get("params"), dict) else {}
|
|
prompt_text = _interactive_prompt_text(payload, "Please provide the requested input")
|
|
form_fields = params.get("inputForm") if isinstance(params.get("inputForm"), list) else []
|
|
|
|
print()
|
|
print(f"[INTERACTIVE] {prompt_text}")
|
|
|
|
if not form_fields:
|
|
value = input("Input (/cancel to stop): ").strip()
|
|
if value.lower() == "/cancel":
|
|
return None
|
|
return value
|
|
|
|
values: Dict[str, Any] = {}
|
|
for index, field in enumerate(form_fields, start=1):
|
|
if not isinstance(field, dict):
|
|
continue
|
|
|
|
name = str(field.get("key") or field.get("name") or f"field_{index}").strip() or f"field_{index}"
|
|
label = str(field.get("label") or field.get("name") or name).strip() or name
|
|
placeholder = str(field.get("placeholder") or "").strip()
|
|
default_value = field.get("defaultValue", field.get("default"))
|
|
required = bool(field.get("required"))
|
|
|
|
prompt_label = label
|
|
if placeholder:
|
|
prompt_label = f"{prompt_label} ({placeholder})"
|
|
if default_value not in (None, ""):
|
|
prompt_label = f"{prompt_label} [{default_value}]"
|
|
if not required:
|
|
prompt_label = f"{prompt_label} [optional]"
|
|
prompt_label = f"{prompt_label}: "
|
|
|
|
while True:
|
|
raw_input_value = input(prompt_label).strip()
|
|
if raw_input_value.lower() == "/cancel":
|
|
return None
|
|
if not raw_input_value and default_value not in (None, ""):
|
|
raw_input_value = str(default_value)
|
|
if raw_input_value or not required:
|
|
values[name] = raw_input_value
|
|
break
|
|
print("This field is required.")
|
|
|
|
return json.dumps(values, ensure_ascii=False)
|
|
|
|
|
|
def prompt_interactive(event: FastGPTInteractiveEvent) -> Optional[str]:
|
|
if event.interaction_type == "userInput":
|
|
return _prompt_user_input(event)
|
|
return _prompt_user_select(event)
|
|
|
|
|
|
def stream_reply(client: ChatClient, messages: List[Dict[str, Any]], chat_id: str) -> StreamTurnResult:
|
|
"""Stream a single FastGPT request and stop when interactive input is required."""
|
|
|
|
response = client.create_chat_completion(
|
|
messages=messages,
|
|
stream=True,
|
|
detail=True,
|
|
chatId=chat_id,
|
|
)
|
|
response.raise_for_status()
|
|
|
|
full_content: List[str] = []
|
|
interactive_event: Optional[FastGPTInteractiveEvent] = None
|
|
printed_text = False
|
|
|
|
try:
|
|
for event in iter_stream_events(response):
|
|
if event.kind in {"data", "answer", "fastAnswer"}:
|
|
content = _extract_text_from_event(event.kind, event.data)
|
|
if content:
|
|
printed_text = True
|
|
full_content.append(content)
|
|
_write_text(content)
|
|
continue
|
|
|
|
if event.kind == "flowNodeStatus":
|
|
if printed_text:
|
|
print()
|
|
printed_text = False
|
|
if isinstance(event.data, dict):
|
|
status = str(event.data.get("status") or "?")
|
|
node_name = str(event.data.get("nodeName") or event.data.get("name") or event.data.get("node_id") or "Unknown node")
|
|
print(f"[FLOW] {status}: {node_name}")
|
|
else:
|
|
print(f"[FLOW] {event.data}")
|
|
continue
|
|
|
|
if event.kind == "flowResponses":
|
|
if printed_text:
|
|
print()
|
|
printed_text = False
|
|
if isinstance(event.data, dict):
|
|
module_name = str(event.data.get("moduleName") or event.data.get("nodeName") or "Unknown module")
|
|
print(f"[FLOW] response from: {module_name}")
|
|
elif isinstance(event.data, list):
|
|
print(f"[FLOW] response details: {len(event.data)} module record(s)")
|
|
else:
|
|
print(f"[FLOW] response details: {event.data}")
|
|
continue
|
|
|
|
if event.kind == "toolCall":
|
|
if printed_text:
|
|
print()
|
|
printed_text = False
|
|
tool_name = _tool_name_from_event(event.data)
|
|
print(f"[TOOL] Calling: {tool_name}")
|
|
continue
|
|
|
|
if event.kind == "toolParams":
|
|
if printed_text:
|
|
print()
|
|
printed_text = False
|
|
print(f"[TOOL] Params: {event.data}")
|
|
continue
|
|
|
|
if event.kind == "toolResponse":
|
|
if printed_text:
|
|
print()
|
|
printed_text = False
|
|
print(f"[TOOL] Response: {event.data}")
|
|
continue
|
|
|
|
if event.kind == "updateVariables":
|
|
if printed_text:
|
|
print()
|
|
printed_text = False
|
|
if isinstance(event.data, dict):
|
|
print(f"[VARS] Updated: {event.data.get('variables') or event.data}")
|
|
else:
|
|
print(f"[VARS] Updated: {event.data}")
|
|
continue
|
|
|
|
if event.kind == "interactive":
|
|
if printed_text:
|
|
print()
|
|
interactive_event = event
|
|
break
|
|
|
|
if event.kind == "error":
|
|
if printed_text:
|
|
print()
|
|
message = str(event.data.get("message") or event.data.get("error") or "Unknown FastGPT error")
|
|
raise RuntimeError(message)
|
|
|
|
if event.kind == "done":
|
|
break
|
|
finally:
|
|
response.close()
|
|
|
|
return StreamTurnResult(text="".join(full_content), interactive=interactive_event)
|
|
|
|
|
|
def _parse_args() -> argparse.Namespace:
|
|
parser = argparse.ArgumentParser(description="FastGPT interactive chat CLI")
|
|
parser.add_argument(
|
|
"--chat-id",
|
|
dest="chat_id",
|
|
help="Reuse an existing FastGPT chatId. Defaults to a random chat_tui_* value.",
|
|
)
|
|
return parser.parse_args()
|
|
|
|
|
|
def main() -> None:
|
|
args = _parse_args()
|
|
|
|
if not API_KEY or not BASE_URL:
|
|
print("Set API_KEY and BASE_URL in .env (see .env.example)")
|
|
sys.exit(1)
|
|
|
|
chat_id = args.chat_id or f"chat_tui_{uuid.uuid4().hex[:12]}"
|
|
print("FastGPT Chat (interactive streaming). Type /quit to exit.\n")
|
|
print(f"Using chatId: {chat_id}\n")
|
|
|
|
with ChatClient(api_key=API_KEY, base_url=BASE_URL) as client:
|
|
while True:
|
|
try:
|
|
user_input = input("You: ").strip()
|
|
except (EOFError, KeyboardInterrupt):
|
|
print("\nBye.")
|
|
break
|
|
|
|
if not user_input:
|
|
continue
|
|
if user_input.lower() in {"/quit", "/exit", "/q"}:
|
|
print("Bye.")
|
|
break
|
|
|
|
pending_messages = [{"role": "user", "content": user_input}]
|
|
assistant_parts: List[str] = []
|
|
|
|
print("Assistant: ", end="", flush=True)
|
|
while True:
|
|
try:
|
|
turn_result = stream_reply(client, pending_messages, chat_id)
|
|
except Exception as exc:
|
|
print(f"\nError: {exc}")
|
|
break
|
|
|
|
if turn_result.text:
|
|
assistant_parts.append(turn_result.text)
|
|
|
|
if turn_result.interactive is None:
|
|
print()
|
|
break
|
|
|
|
follow_up = prompt_interactive(turn_result.interactive)
|
|
if follow_up is None:
|
|
print("[INTERACTIVE] Prompt cancelled locally.")
|
|
break
|
|
|
|
pending_messages = [{"role": "user", "content": follow_up}]
|
|
print("Assistant (resume): ", end="", flush=True)
|
|
|
|
if assistant_parts:
|
|
print("-" * 40)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|