feat(tgapp): stream MarkdownV2 replies with draft fallback and file segments (#135)

This commit is contained in:
YooooEX
2026-04-24 09:01:49 +08:00
committed by GitHub
parent fd46d61543
commit f516c52a72

View File

@@ -1,91 +1,275 @@
import os, sys, re, threading, asyncio, queue as Q, socket, time
import os, sys, re, threading, asyncio, queue as Q, socket, time, random
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
_TEMP_DIR = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), 'temp')
from agentmain import GeneraticAgent
try:
from telegram import BotCommand
from telegram.constants import ChatType, MessageLimit, ParseMode
from telegram.ext import ApplicationBuilder, MessageHandler, filters, ContextTypes
from telegram.helpers import escape_markdown
from telegram.request import HTTPXRequest
except:
print("Please ask the agent install python-telegram-bot to use telegram module.")
sys.exit(1)
from chatapp_common import HELP_TEXT, TELEGRAM_MENU_COMMANDS, format_restore
from chatapp_common import (
FILE_HINT,
HELP_TEXT,
TELEGRAM_MENU_COMMANDS,
clean_reply,
extract_files,
format_restore,
split_text,
strip_files,
)
from continue_cmd import handle_frontend_command, reset_conversation
from llmcore import mykeys
agent = GeneraticAgent()
agent.verbose = False
agent.inc_out = True
ALLOWED = set(mykeys.get('tg_allowed_users', []))
_TAG_PATS = [r'<' + t + r'>.*?</' + t + r'>' for t in ('thinking', 'summary', 'tool_use')]
_TAG_PATS.append(r'<file_content>.*?</file_content>')
_DRAFT_HINT = "thinking..."
_STREAM_SUFFIX = ""
_STREAM_SEGMENT_LIMIT = max(1200, MessageLimit.MAX_TEXT_LENGTH - 256)
_QUEUE_WAIT_SECONDS = 1
_MD_TOKEN_RE = re.compile(
r"(`{3,})([A-Za-z0-9_+-]*)\n([\s\S]*?)\1"
r"|\[([^\]]+)\]\(([^)\n]+)\)"
r"|`([^`\n]+)`"
r"|\*\*([^\n]+?)\*\*"
r"|(?<!\*)\*(?!\*)([^\n]+?)(?<!\*)\*(?!\*)",
re.DOTALL,
)
def _clean(t):
for p in _TAG_PATS:
t = re.sub(p, '', t, flags=re.DOTALL)
return re.sub(r'\n{3,}', '\n\n', t).strip() or '...'
def _make_draft_id():
return random.randint(1, 2**31 - 1)
import html as _html
def _inline_md(s):
s = re.sub(r'\*\*(.+?)\*\*', r'<b>\1</b>', s)
s = re.sub(r'(?<!\*)\*(?!\*)(.+?)(?<!\*)\*(?!\*)', r'<i>\1</i>', s)
s = re.sub(r'`([^`]+)`', r'<code>\1</code>', s)
return s
def _to_html(t):
def _visible_segments(text):
text = (text or "").strip()
return split_text(text, _STREAM_SEGMENT_LIMIT) if text else []
def _resolve_files(paths):
files, seen = [], set()
for fpath in paths:
if not os.path.isabs(fpath):
fpath = os.path.join(_TEMP_DIR, fpath)
if fpath in seen or not os.path.exists(fpath):
continue
files.append(fpath)
seen.add(fpath)
return files
def _escape_pre(text):
return escape_markdown(text or "", version=2, entity_type="pre")
def _escape_code(text):
return escape_markdown(text or "", version=2, entity_type="code")
def _escape_link_target(text):
return escape_markdown(text or "", version=2, entity_type="text_link")
def _to_markdown_v2(text):
if not text:
return ""
parts, pos = [], 0
for m in re.finditer(r'(`{3,})(?:\w*\n)?([\s\S]*?)\1', t):
parts.append(_inline_md(_html.escape(t[pos:m.start()])))
parts.append('<pre><code>' + _html.escape(m.group(2)) + '</code></pre>')
pos = m.end()
parts.append(_inline_md(_html.escape(t[pos:])))
return ''.join(parts)
for match in _MD_TOKEN_RE.finditer(text):
parts.append(escape_markdown(text[pos:match.start()], version=2))
if match.group(1):
lang = re.sub(r"[^A-Za-z0-9_+-]", "", match.group(2) or "")
code = _escape_pre(match.group(3) or "")
header = f"```{lang}\n" if lang else "```\n"
parts.append(f"{header}{code}\n```")
elif match.group(4) is not None:
label = escape_markdown(match.group(4), version=2)
target = _escape_link_target(match.group(5))
parts.append(f"[{label}]({target})")
elif match.group(6) is not None:
parts.append(f"`{_escape_code(match.group(6))}`")
elif match.group(7) is not None:
parts.append(f"*{escape_markdown(match.group(7), version=2)}*")
elif match.group(8) is not None:
parts.append(f"_{escape_markdown(match.group(8), version=2)}_")
pos = match.end()
parts.append(escape_markdown(text[pos:], version=2))
return "".join(parts)
def _is_not_modified_error(exc):
return "not modified" in str(exc).lower()
class _TelegramStreamSession:
def __init__(self, root_msg):
self.root_msg = root_msg
self.private_chat = getattr(getattr(root_msg, "chat", None), "type", "") == ChatType.PRIVATE
self.can_use_draft = self.private_chat
self.draft_id = _make_draft_id()
self.live_msg = None
self.raw_text = ""
self.files = []
self.sent_segments = 0
self.active_display = ""
async def prime(self):
if self.can_use_draft and await self._send_draft(_DRAFT_HINT):
self.active_display = _DRAFT_HINT
return
await self._upsert_live_message(_DRAFT_HINT)
self.active_display = _DRAFT_HINT
async def add_chunk(self, chunk):
if not chunk:
return
self.raw_text += chunk
await self._refresh(done=False, send_files=False)
async def finalize(self, full_text=None, send_files=True):
if full_text is not None:
self.raw_text = full_text
await self._refresh(done=True, send_files=send_files)
async def finish_with_notice(self, notice):
if self.raw_text.strip():
await self.finalize(send_files=False)
await self._reply_text(notice)
return
if self.live_msg is not None:
await self._edit_text(self.live_msg, notice)
self.live_msg = None
self.active_display = ""
return
await self._reply_text(notice)
self.active_display = ""
async def _refresh(self, done, send_files):
cleaned = clean_reply(self.raw_text) if self.raw_text.strip() else ""
self.files = _resolve_files(extract_files(cleaned))
body = strip_files(cleaned)
if done and not body and self.files:
body = "已生成附件"
elif done and not body:
body = "..."
segments = _visible_segments(body)
finalized_target = len(segments) if done else max(len(segments) - 1, 0)
while self.sent_segments < finalized_target:
await self._finalize_segment(segments[self.sent_segments])
self.sent_segments += 1
if done:
if send_files:
await self._send_files()
return
active_text = segments[-1] if segments else _DRAFT_HINT
await self._stream_active(active_text)
async def _stream_active(self, text):
display = (text or _DRAFT_HINT).strip() or _DRAFT_HINT
if display != _DRAFT_HINT:
display = display + _STREAM_SUFFIX
if display == self.active_display:
return
if self.can_use_draft and await self._send_draft(display):
self.active_display = display
return
await self._upsert_live_message(display)
self.active_display = display
async def _finalize_segment(self, text):
final_text = (text or "").strip() or "..."
if self.live_msg is not None:
await self._edit_text(self.live_msg, final_text)
self.live_msg = None
else:
await self._reply_text(final_text)
self.active_display = ""
if self.can_use_draft:
self.draft_id = _make_draft_id()
async def _send_files(self):
for fpath in self.files:
if fpath.lower().endswith((".png", ".jpg", ".jpeg", ".gif", ".webp")):
try:
with open(fpath, "rb") as fp:
await self.root_msg.reply_photo(fp)
except Exception: pass
else:
try:
with open(fpath, "rb") as fp:
await self.root_msg.reply_document(fp)
except Exception: pass
async def _send_draft(self, text):
try:
await self.root_msg.reply_text_draft(
self.draft_id,
_to_markdown_v2(text),
parse_mode=ParseMode.MARKDOWN_V2,
)
return True
except Exception as exc:
if _is_not_modified_error(exc):
return True
print(f"[TG draft fallback] {type(exc).__name__}: {exc}", flush=True)
self.can_use_draft = False
self.draft_id = _make_draft_id()
return False
async def _reply_text(self, text):
markdown = _to_markdown_v2(text)
try:
return await self.root_msg.reply_text(markdown, parse_mode=ParseMode.MARKDOWN_V2)
except Exception as exc:
if _is_not_modified_error(exc):
return None
return await self.root_msg.reply_text(text)
async def _edit_text(self, msg, text):
markdown = _to_markdown_v2(text)
try:
updated = await msg.edit_text(markdown, parse_mode=ParseMode.MARKDOWN_V2)
except Exception as exc:
if _is_not_modified_error(exc):
return msg
updated = await msg.edit_text(text)
return updated if hasattr(updated, "edit_text") else msg
async def _upsert_live_message(self, text):
if self.live_msg is None:
self.live_msg = await self._reply_text(text)
else:
self.live_msg = await self._edit_text(self.live_msg, text)
async def _stream(dq, msg):
last_text = ""
while True:
await asyncio.sleep(3)
item = None
try:
while True: item = dq.get_nowait()
except Q.Empty: pass
if item is None: continue
raw = item.get("done") or item.get("next", "")
done = "done" in item
show = _clean(raw)
if len(show) > 4000:
# freeze current msg, start a new one
try: msg = await msg.reply_text("(continued...)")
except Exception: pass
last_text = ""
show = show[-3900:]
display = show if done else show + ""
if display != last_text:
try: await msg.edit_text(_to_html(display), parse_mode='HTML')
except Exception:
try: await msg.edit_text(display)
except Exception: pass
last_text = display
if done:
files = re.findall(r'\[FILE:([^\]]+)\]', show[-1000:])
for fpath in files:
if not os.path.isabs(fpath): fpath = os.path.join(_TEMP_DIR, fpath)
if os.path.exists(fpath):
if fpath.lower().endswith(('.png','.jpg','.jpeg','.gif','.webp')):
try: await msg.reply_photo(open(fpath,'rb'))
except Exception: pass
else:
try: await msg.reply_document(open(fpath,'rb'))
except Exception: pass
show = re.sub(r'\[FILE:[^\]]+\]', '', show)
if show.strip():
try: await msg.edit_text(_to_html(show), parse_mode='HTML')
except Exception:
try: await msg.edit_text(show)
except Exception: pass
break
session = _TelegramStreamSession(msg)
await session.prime()
try:
while True:
try:
first = await asyncio.to_thread(dq.get, True, _QUEUE_WAIT_SECONDS)
except Q.Empty:
continue
items = [first]
try:
while True:
items.append(dq.get_nowait())
except Q.Empty:
pass
done_item = next((item for item in items if "done" in item), None)
if done_item is not None:
await session.finalize(done_item.get("done", ""))
break
chunk = "".join(item.get("next", "") for item in items if item.get("next"))
if chunk:
await session.add_chunk(chunk)
except asyncio.CancelledError:
await session.finish_with_notice("⏹️ 已停止")
except Exception as exc:
print(f"[TG stream error] {type(exc).__name__}: {exc}", flush=True)
await session.finish_with_notice(f"❌ 输出失败: {exc}")
def _normalized_command(text):
parts = (text or '').strip().split(None, 1)
parts = (text or "").strip().split(None, 1)
if not parts:
return ''
head = parts[0].lower()
@@ -105,10 +289,9 @@ async def handle_msg(update, ctx):
uid = update.effective_user.id
if ALLOWED and uid not in ALLOWED:
return await update.message.reply_text("no")
msg = await update.message.reply_text("thinking...")
prompt = f"If you need to show files to user, use [FILE:filepath] in your response.\n\n{update.message.text}"
prompt = f"{FILE_HINT}\n\n{update.message.text}"
dq = agent.put_task(prompt, source="telegram")
task = asyncio.create_task(_stream(dq, msg))
task = asyncio.create_task(_stream(dq, update.message))
ctx.user_data['stream_task'] = task
async def cmd_abort(update, ctx):
@@ -139,9 +322,8 @@ async def handle_photo(update, ctx):
await file.download_to_drive(os.path.join(_TEMP_DIR, fpath))
caption = update.message.caption
prompt = f"[TIPS] 收到图片temp/{fpath}\n{caption}" if caption else f"[TIPS] 收到图片temp/{fpath},请等待下一步指令"
msg = await update.message.reply_text("thinking...")
dq = agent.put_task(prompt, source="telegram")
task = asyncio.create_task(_stream(dq, msg))
task = asyncio.create_task(_stream(dq, update.message))
ctx.user_data['stream_task'] = task
async def handle_command(update, ctx):