317 lines
15 KiB
Python
317 lines
15 KiB
Python
import os, json, re, time, requests, sys, threading
|
||
|
||
try: import mykey
|
||
except: raise Exception('[ERROR] mykey.py not found, please copy mykey_template.py to mykey.py and fill your LLM backend.')
|
||
|
||
def get_config(name, default=""): return getattr(mykey, name, default)
|
||
|
||
sider_cookie = get_config("sider_cookie")
|
||
oai_apikey = get_config("oai_apikey")
|
||
oai_apibase = get_config("oai_apibase")
|
||
oai_model = get_config("oai_model")
|
||
google_api_key = get_config("google_api_key")
|
||
proxy = get_config("proxy", 'http://127.0.0.1:2082')
|
||
proxies = {"http": proxy, "https": proxy} if proxy else None
|
||
|
||
class SiderLLMSession:
|
||
def __init__(self, default_model="gemini-3.0-flash"):
|
||
from sider_ai_api import Session
|
||
self._core = Session(cookie=sider_cookie, proxies=proxies)
|
||
self.default_model = default_model
|
||
def ask(self, prompt, model=None, stream=False):
|
||
if model is None: model = self.default_model
|
||
if len(prompt) > 28000:
|
||
print(f"[Warn] Prompt too long ({len(prompt)} chars), truncating.")
|
||
prompt = prompt[-28000:]
|
||
full_text = self._core.chat(prompt, model, stream=False)
|
||
if stream: return iter([full_text]) # gen有奇怪的空回复或死循环行为，sider足够快
|
||
return full_text
|
||
|
||
class GeminiSession:
|
||
def __init__(self, api_key=None, default_model="gemini-2.0-flash-001", proxy=proxy):
|
||
self.api_key = api_key or google_api_key
|
||
if not self.api_key: raise ValueError("google_api_key 未配置或为空，请在 mykey.py 中设置")
|
||
self.default_model = default_model
|
||
self.proxies = {"http":proxy, "https":proxy} if proxy else None
|
||
def ask(self, prompt, model=None, stream=False):
|
||
if model is None: model = self.default_model
|
||
url = f"https://generativelanguage.googleapis.com/v1/models/{model}:generateContent?key={self.api_key}"
|
||
headers = {"Content-Type":"application/json"}
|
||
data = {"contents":[{"role":"user","parts":[{"text":prompt}]}]}
|
||
try:
|
||
kw = {"headers":headers, "json":data, "timeout":60, 'proxies': self.proxies}
|
||
r = requests.post(url, **kw)
|
||
except Exception as e:
|
||
return f"[GeminiError] request failed: {e}"
|
||
if r.status_code != 200:
|
||
body = r.text[:500].replace("\n"," ")
|
||
return f"[GeminiError] HTTP {r.status_code}: {body}"
|
||
try:
|
||
obj = r.json(); cands = obj.get("candidates") or []
|
||
if not cands: return "[GeminiError] empty candidates"
|
||
parts = (cands[0].get("content") or {}).get("parts") or []
|
||
full_text = "".join(p.get("text","") for p in parts)
|
||
except Exception as e:
|
||
return f"[GeminiError] invalid response format: {e}"
|
||
return iter([full_text]) if stream else full_text
|
||
|
||
class LLMSession:
|
||
def __init__(self, api_key=oai_apikey, api_base=oai_apibase, model=oai_model, context_win=12000):
|
||
self.api_key = api_key
|
||
self.api_base = api_base
|
||
self.raw_msgs = []
|
||
self.messages = []
|
||
self.context_win = context_win
|
||
self.default_model = model
|
||
self.lock = threading.Lock()
|
||
|
||
def raw_ask(self, messages, model=None, temperature=0.5):
|
||
if model is None: model = self.default_model
|
||
headers = {"Authorization": f"Bearer {self.api_key}", "Content-Type": "application/json", "Accept": "text/event-stream"}
|
||
payload = {"model": model, "messages": messages, "temperature": temperature, "stream": True}
|
||
try:
|
||
with requests.post(f"{self.api_base}/chat/completions",
|
||
headers=headers, json=payload, stream=True, timeout=(5, 60)) as r:
|
||
r.raise_for_status()
|
||
buffer = ''
|
||
for line in r.iter_lines():
|
||
line = line.decode("utf-8")
|
||
if not line or not line.startswith("data:"): continue
|
||
data = line[5:].lstrip()
|
||
if data == "[DONE]": break
|
||
obj = json.loads(data)
|
||
ch = (obj.get("choices") or [{}])[0]
|
||
finish_reason = ch.get("finish_reason")
|
||
delta = (ch.get("delta") or {}).get("content")
|
||
if delta:
|
||
yield delta
|
||
buffer += delta
|
||
if '</tool_use>' in buffer[-30:]: break
|
||
if finish_reason: break
|
||
except Exception as e:
|
||
yield f"Error: {str(e)}"
|
||
|
||
def make_messages(self, raw_list, omit_images=True):
|
||
messages = []
|
||
for msg in raw_list:
|
||
if omit_images and msg['image']:
|
||
messages.append({"role": msg['role'], "content": "[Image omitted, if you needed it, ask me]\n" + msg['prompt']})
|
||
elif not omit_images and msg['image']:
|
||
messages.append({"role": msg['role'], "content": [
|
||
{"type": "image_url", "image_url": {"url": f"data:image/png;base64,{msg['image']}"}},
|
||
{"type": "text", "text": msg['prompt']} ]})
|
||
else:
|
||
messages.append({"role": msg['role'], "content": msg['prompt']})
|
||
return messages
|
||
|
||
def summary_history(self, model=None):
|
||
if model is None: model = self.default_model
|
||
with self.lock:
|
||
keep = 0; tok = 0
|
||
for m in reversed(self.raw_msgs):
|
||
l = len(str(m))//4
|
||
if tok + l > self.context_win*0.2: break
|
||
tok += l; keep += 1
|
||
keep = max(2, keep)
|
||
old, self.raw_msgs = self.raw_msgs[:-keep], self.raw_msgs[-keep:]
|
||
if len(old) == 0: old = self.raw_msgs; self.raw_msgs = []
|
||
p = "Summarize prev summary and prev conversations into compact memory (facts/decisions/constraints/open questions). Do NOT restate long schemas. The new summary should less than 1000 tokens. Permit dropping non-important things.\n"
|
||
messages = self.make_messages(old, omit_images=True)
|
||
messages += [{"role":"user", "content":p}]
|
||
msg_lens = [1000 if isinstance(m["content"], list) else len(str(m["content"]))//4 for m in messages]
|
||
summary = ''.join(list(self.raw_ask(messages, model, temperature=0.1)))
|
||
print('[Debug] Summary length:', len(summary)//4, '; Orig context lengths:', str(msg_lens))
|
||
if not summary.startswith("Error:"):
|
||
self.raw_msgs.insert(0, {"role":"assistant", "prompt":"Prev summary:\n"+summary, "image":None})
|
||
else: self.raw_msgs = old + self.raw_msgs # 不做了，下次再做
|
||
|
||
def ask(self, prompt, model=None, image_base64=None, stream=False):
|
||
if model is None: model = self.default_model
|
||
def _ask_gen():
|
||
content = ''
|
||
with self.lock:
|
||
self.raw_msgs.append({"role": "user", "prompt": prompt, "image": image_base64})
|
||
messages = self.make_messages(self.raw_msgs[:-1], omit_images=True)
|
||
messages += self.make_messages([self.raw_msgs[-1]], omit_images=False)
|
||
msg_lens = [1000 if isinstance(m["content"], list) else len(str(m["content"]))//4 for m in messages]
|
||
total_len = sum(msg_lens) # estimate token count
|
||
gen = self.raw_ask(messages, model)
|
||
for chunk in gen:
|
||
content += chunk; yield chunk
|
||
if not content.startswith("Error:"):
|
||
self.raw_msgs.append({"role": "assistant", "prompt": content, "image": None})
|
||
if total_len > 5000: print(f"[Debug] Whole context length {total_len} {str(msg_lens)}.")
|
||
if total_len > self.context_win:
|
||
yield '[NextWillSummary]'
|
||
threading.Thread(target=self.summary_history, daemon=True).start()
|
||
if stream: return _ask_gen()
|
||
return ''.join(list(_ask_gen()))
|
||
|
||
|
||
class MockFunction:
|
||
def __init__(self, name, arguments):
|
||
self.name = name
|
||
self.arguments = arguments
|
||
|
||
class MockToolCall:
|
||
def __init__(self, name, args):
|
||
arg_str = json.dumps(args, ensure_ascii=False) if isinstance(args, dict) else args
|
||
self.function = MockFunction(name, arg_str)
|
||
|
||
class MockResponse:
|
||
def __init__(self, thinking, content, tool_calls, raw):
|
||
self.thinking = thinking # 存放 <thinking> 内部的思维过程
|
||
self.content = content # 存放去除标签后的纯文本回复
|
||
self.tool_calls = tool_calls # 存放 MockToolCall 列表 或 None
|
||
self.raw = raw
|
||
def __repr__(self):
|
||
return f"<MockResponse thinking={bool(self.thinking)}, content='{self.content}', tools={bool(self.tool_calls)}>"
|
||
|
||
class ToolClient:
|
||
def __init__(self, backends, auto_save_tokens=False):
|
||
if isinstance(backends, list): self.backends = backends
|
||
else: self.backends = [backends]
|
||
self.backend = self.backends[0]
|
||
self.auto_save_tokens = auto_save_tokens
|
||
self.last_tools = ''
|
||
self.total_cd_tokens = 0
|
||
|
||
def chat(self, messages, tools=None):
|
||
full_prompt = self._build_protocol_prompt(messages, tools)
|
||
print("Full prompt length:", len(full_prompt), 'chars')
|
||
with open('model_responses.txt', 'a', encoding='utf-8', errors="replace") as f:
|
||
f.write(f"=== Prompt ===\n{full_prompt}\n")
|
||
gen = self.backend.ask(full_prompt, stream=True)
|
||
raw_text = ''; summarytag = '[NextWillSummary]'
|
||
for chunk in gen:
|
||
raw_text += chunk;
|
||
if chunk != summarytag: yield chunk
|
||
print('Complete response received.')
|
||
if raw_text.endswith(summarytag):
|
||
self.last_tools = ''; raw_text = raw_text[:-len(summarytag)]
|
||
with open('model_responses.txt', 'a', encoding='utf-8', errors="replace") as f:
|
||
f.write(f"=== Response ===\n{raw_text}\n\n")
|
||
return self._parse_mixed_response(raw_text)
|
||
|
||
def _build_protocol_prompt(self, messages, tools):
|
||
system_content = next((m['content'] for m in messages if m['role'].lower() == 'system'), "")
|
||
history_msgs = [m for m in messages if m['role'].lower() != 'system']
|
||
|
||
# 构造工具描述
|
||
tool_instruction = ""
|
||
if tools:
|
||
tools_json = json.dumps(tools, ensure_ascii=False, separators=(',', ':'))
|
||
tool_instruction = f"""
|
||
### 交互协议 (必须严格遵守)
|
||
请按照以下步骤思考并行动，标签之间需要回车换行：
|
||
1. **思考**: 在 `<thinking>` 标签中先进行思考，分析现状和策略。
|
||
2. **总结**: 在 `<summary>` 中输出*极为简短*的高度概括的单行（<30字）物理快照，包括上次工具调用结果获取的新信息+本次工具调用意图和预期。此内容将进入长期工作记忆，记录关键信息，严禁输出无实际信息增量的描述。
|
||
3. **行动**: 如果需要调用工具，请在回复正文之后输出一个 **<tool_use>块**，然后结束，我会稍后给你返回<tool_result>块。
|
||
格式: ```<tool_use>\n{{"name": "工具名", "arguments": {{参数}}}}\n</tool_use>\n```
|
||
|
||
### 可用工具库
|
||
{tools_json}
|
||
"""
|
||
if self.auto_save_tokens and self.last_tools == tools_json:
|
||
tool_instruction = "\n### 交互协议保持不变，沿用之前的协议和工具库。\n"
|
||
else:
|
||
self.total_cd_tokens = 0
|
||
self.last_tools = tools_json
|
||
|
||
prompt = ""
|
||
if system_content: prompt += f"=== SYSTEM ===\n{system_content}\n"
|
||
prompt += f"{tool_instruction}\n\n"
|
||
for m in history_msgs:
|
||
role = "USER" if m['role'] == 'user' else "ASSISTANT"
|
||
prompt += f"=== {role} ===\n{m['content']}\n\n"
|
||
self.total_cd_tokens += len(m['content'])
|
||
|
||
if self.total_cd_tokens > 9000: self.last_tools = ''
|
||
|
||
prompt += "=== ASSISTANT ===\n"
|
||
return prompt
|
||
|
||
def _parse_mixed_response(self, text):
|
||
remaining_text = text
|
||
thinking = ''
|
||
think_pattern = r"<thinking>(.*?)</thinking>"
|
||
think_match = re.search(think_pattern, text, re.DOTALL)
|
||
|
||
if think_match:
|
||
thinking = think_match.group(1).strip()
|
||
remaining_text = re.sub(think_pattern, "", remaining_text, flags=re.DOTALL)
|
||
|
||
tool_calls = None
|
||
tool_pattern = r"<tool_use>(.*?)</tool_use>"
|
||
tool_match = re.search(tool_pattern, remaining_text, re.DOTALL)
|
||
|
||
json_str = ""
|
||
if tool_match:
|
||
json_str = tool_match.group(1).strip()
|
||
remaining_text = re.sub(tool_pattern, "", remaining_text, flags=re.DOTALL)
|
||
elif '<tool_use>' in remaining_text:
|
||
weaktoolstr = remaining_text.split('<tool_use>')[-1].strip()
|
||
json_str = weaktoolstr if weaktoolstr.endswith('}') else ''
|
||
if json_str == '' and '```' in weaktoolstr and weaktoolstr.split('```')[0].strip().endswith('}'):
|
||
json_str = weaktoolstr.split('```')[0].strip()
|
||
remaining_text = remaining_text.replace('<tool_use>'+weaktoolstr, "")
|
||
|
||
if json_str:
|
||
try:
|
||
data = tryparse(json_str)
|
||
func_name = data.get('name') or data.get('function') or data.get('tool')
|
||
args = data.get('arguments') or data.get('args') or data.get('params') or data.get('parameters')
|
||
if args is None: args = data
|
||
if func_name: tool_calls = [MockToolCall(func_name, args)]
|
||
except json.JSONDecodeError:
|
||
print("[Warn] Failed to parse tool_use JSON:", json_str)
|
||
remaining_text += f"[Warning] JSON 解析失败，模型输出了无效的 JSON."
|
||
except Exception as e:
|
||
print("[Error] Exception during tool_use parsing:", str(e), data)
|
||
|
||
content = remaining_text.strip()
|
||
if not content: content = ""
|
||
return MockResponse(thinking, content, tool_calls, text)
|
||
|
||
def tryparse(json_str):
|
||
try: return json.loads(json_str)
|
||
except:
|
||
return json.loads(json_str[:-1])
|
||
|
||
if __name__ == "__main__":
|
||
import sys, os
|
||
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
||
try:
|
||
import mykey
|
||
except ImportError:
|
||
class MockMyKey: pass
|
||
mykey = MockMyKey()
|
||
|
||
sider_cookie = get_config("sider_cookie")
|
||
oai_apikey = get_config("oai_apikey")
|
||
oai_apibase = get_config("oai_apibase")
|
||
oai_model = get_config("oai_model")
|
||
google_api_key = get_config("google_api_key")
|
||
|
||
llmclient = ToolClient(GeminiSession(api_key=google_api_key, proxy='127.0.0.1:2082').ask)
|
||
#llmclient = ToolClient(LLMSession(api_key=oai_apikey, api_base=oai_apibase, model=oai_model).ask)
|
||
#llmclient = ToolClient(SiderLLMSession().ask)
|
||
def get_final(gen):
|
||
try:
|
||
while True: print('mid:', next(gen))
|
||
except StopIteration as e:
|
||
return e.value
|
||
|
||
response = get_final(llmclient.chat(
|
||
messages=[{"role": "user", "content": "我的IP是多少"}],
|
||
tools=[{"name": "get_ip", "parameters": {}}]
|
||
))
|
||
print(f"思考: {response.thinking}")
|
||
if response.tool_calls:
|
||
cmd = response.tool_calls[0]
|
||
print(f"调用: {cmd.function.name} 参数: {cmd.function.arguments}")
|
||
|
||
response = get_final(llmclient.chat(
|
||
messages=[{"role": "user", "content": "<tool_result>10.176.45.12</tool_result>"}]
|
||
))
|
||
print(response.content) |