Spaces:
Running
Running
File size: 9,056 Bytes
e2ab8a3 1bdfa3a e2ab8a3 e00ef83 1bdfa3a e2ab8a3 e00ef83 e2ab8a3 1bdfa3a e2ab8a3 e00ef83 e2ab8a3 e00ef83 e2ab8a3 e00ef83 e2ab8a3 e00ef83 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 |
import json
import uuid
import os
from pathlib import Path
from typing import Dict, Optional, Tuple, Iterator, List, Generator, Any
import struct
import requests
class StreamTracker:
def __init__(self):
self.has_content = False
def track(self, gen: Generator[str, None, None]) -> Generator[str, None, None]:
for item in gen:
if item:
self.has_content = True
yield item
def _get_proxies() -> Optional[Dict[str, str]]:
proxy = os.getenv("HTTP_PROXY", "").strip()
if proxy:
return {"http": proxy, "https": proxy}
return None
BASE_DIR = Path(__file__).resolve().parent
TEMPLATE_PATH = BASE_DIR / "templates" / "streaming_request.json"
def load_template() -> Tuple[str, Dict[str, str], Dict[str, Any]]:
data = json.loads(TEMPLATE_PATH.read_text(encoding="utf-8"))
url, headers, body = data
assert isinstance(url, str) and isinstance(headers, dict) and isinstance(body, dict)
return url, headers, body
def _merge_headers(as_log: Dict[str, str], bearer_token: str) -> Dict[str, str]:
headers = dict(as_log)
for k in list(headers.keys()):
kl = k.lower()
if kl in ("content-length","host","connection","transfer-encoding"):
headers.pop(k, None)
def set_header(name: str, value: str):
for key in list(headers.keys()):
if key.lower() == name.lower():
del headers[key]
headers[name] = value
set_header("Authorization", f"Bearer {bearer_token}")
set_header("amz-sdk-invocation-id", str(uuid.uuid4()))
return headers
def _parse_event_headers(raw: bytes) -> Dict[str, object]:
headers: Dict[str, object] = {}
i = 0
n = len(raw)
while i < n:
if i + 1 > n:
break
name_len = raw[i]
i += 1
if i + name_len + 1 > n:
break
name = raw[i : i + name_len].decode("utf-8", errors="ignore")
i += name_len
htype = raw[i]
i += 1
if htype == 0:
val = True
elif htype == 1:
val = False
elif htype == 2:
if i + 1 > n: break
val = raw[i]; i += 1
elif htype == 3:
if i + 2 > n: break
val = int.from_bytes(raw[i:i+2],"big",signed=True); i += 2
elif htype == 4:
if i + 4 > n: break
val = int.from_bytes(raw[i:i+4],"big",signed=True); i += 4
elif htype == 5:
if i + 8 > n: break
val = int.from_bytes(raw[i:i+8],"big",signed=True); i += 8
elif htype == 6:
if i + 2 > n: break
l = int.from_bytes(raw[i:i+2],"big"); i += 2
if i + l > n: break
val = raw[i:i+l]; i += l
elif htype == 7:
if i + 2 > n: break
l = int.from_bytes(raw[i:i+2],"big"); i += 2
if i + l > n: break
val = raw[i:i+l].decode("utf-8", errors="ignore"); i += l
elif htype == 8:
if i + 8 > n: break
val = int.from_bytes(raw[i:i+8],"big",signed=False); i += 8
elif htype == 9:
if i + 16 > n: break
import uuid as _uuid
val = str(_uuid.UUID(bytes=bytes(raw[i:i+16]))); i += 16
else:
break
headers[name] = val
return headers
class AwsEventStreamParser:
def __init__(self):
self._buf = bytearray()
def feed(self, data: bytes) -> List[Tuple[Dict[str, object], bytes]]:
if not data:
return []
self._buf.extend(data)
out: List[Tuple[Dict[str, object], bytes]] = []
while True:
if len(self._buf) < 12:
break
total_len, headers_len, _prelude_crc = struct.unpack(">I I I", self._buf[:12])
if total_len < 16 or headers_len > total_len:
self._buf.pop(0)
continue
if len(self._buf) < total_len:
break
msg = bytes(self._buf[:total_len])
del self._buf[:total_len]
headers_raw = msg[12:12+headers_len]
payload = msg[12+headers_len: total_len-4]
headers = _parse_event_headers(headers_raw)
out.append((headers, payload))
return out
def _try_decode_event_payload(payload: bytes) -> Optional[dict]:
try:
return json.loads(payload.decode("utf-8"))
except Exception:
return None
def _extract_text_from_event(ev: dict) -> Optional[str]:
for key in ("assistantResponseEvent","assistantMessage","message","delta","data"):
if key in ev and isinstance(ev[key], dict):
inner = ev[key]
if isinstance(inner.get("content"), str) and inner.get("content"):
return inner["content"]
if isinstance(ev.get("content"), str) and ev.get("content"):
return ev["content"]
for list_key in ("chunks","content"):
if isinstance(ev.get(list_key), list):
buf = []
for item in ev[list_key]:
if isinstance(item, dict):
if isinstance(item.get("content"), str):
buf.append(item["content"])
elif isinstance(item.get("text"), str):
buf.append(item["text"])
elif isinstance(item, str):
buf.append(item)
if buf:
return "".join(buf)
for k in ("text","delta","payload"):
v = ev.get(k)
if isinstance(v, str) and v:
return v
return None
def openai_messages_to_text(messages: List[Dict[str, Any]]) -> str:
lines: List[str] = []
for m in messages:
role = m.get("role","user")
content = m.get("content","")
if isinstance(content, list):
parts = []
for seg in content:
if isinstance(seg, dict) and isinstance(seg.get("text"), str):
parts.append(seg["text"])
elif isinstance(seg, str):
parts.append(seg)
content = "\n".join(parts)
elif not isinstance(content, str):
content = str(content)
lines.append(f"{role}:\n{content}")
return "\n\n".join(lines)
def inject_history(body_json: Dict[str, Any], history_text: str) -> None:
try:
cur = body_json["conversationState"]["currentMessage"]["userInputMessage"]
content = cur.get("content","")
if isinstance(content, str):
cur["content"] = content.replace("你好，你必须讲个故事", history_text)
except Exception:
pass
def inject_model(body_json: Dict[str, Any], model: Optional[str]) -> None:
if not model:
return
try:
body_json["conversationState"]["currentMessage"]["userInputMessage"]["modelId"] = model
except Exception:
pass
def send_chat_request(access_token: str, messages: List[Dict[str, Any]], model: Optional[str] = None, stream: bool = False, timeout: Tuple[int,int] = (15,300)) -> Tuple[Optional[str], Optional[Generator[str, None, None]], bool]:
url, headers_from_log, body_json = load_template()
headers_from_log["amz-sdk-invocation-id"] = str(uuid.uuid4())
try:
body_json["conversationState"]["conversationId"] = str(uuid.uuid4())
except Exception:
pass
history_text = openai_messages_to_text(messages)
inject_history(body_json, history_text)
inject_model(body_json, model)
payload_str = json.dumps(body_json, ensure_ascii=False)
headers = _merge_headers(headers_from_log, access_token)
session = requests.Session()
proxies = _get_proxies()
resp = session.post(url, headers=headers, data=payload_str, stream=True, timeout=timeout, proxies=proxies)
if resp.status_code >= 400:
try:
err = resp.text
except Exception:
err = f"HTTP {resp.status_code}"
raise requests.HTTPError(f"Upstream error {resp.status_code}: {err}", response=resp)
parser = AwsEventStreamParser()
tracker = StreamTracker()
def _iter_text() -> Generator[str, None, None]:
for chunk in resp.iter_content(chunk_size=None):
if not chunk:
continue
events = parser.feed(chunk)
for _ev_headers, payload in events:
parsed = _try_decode_event_payload(payload)
if parsed is not None:
text = _extract_text_from_event(parsed)
if isinstance(text, str) and text:
yield text
else:
try:
txt = payload.decode("utf-8", errors="ignore")
if txt:
yield txt
except Exception:
pass
if stream:
return None, tracker.track(_iter_text()), tracker
else:
buf = []
for t in tracker.track(_iter_text()):
buf.append(t)
return "".join(buf), None, tracker |