Kris
05abecc02b
该提交实现了支持企业微信、微信公众号、个人微信桥接三种模式的完整微信渠道适配器，包含以下核心模块： 1. 基础认证与配置相关：auth_adapter、config_reload、setup_contract等 2. 消息处理与格式转换：format、attachment_adapter、outbound_adapter等 3. 多模式客户端支持：wecom/mp子模块，包含加解密、消息收发能力 4. 辅助能力：限速器、防抖、会话绑定、事件映射、模板渲染等 5. 扩展能力：二维码登录、消息读取、特权用户、心跳监控等 实现了完整的微信生态对接能力，支持消息收发、事件处理、API调用限流、配置热重载等功能。
85 lines
2.4 KiB
Python
85 lines
2.4 KiB
Python
from __future__ import annotations
|
||
|
||
from typing import Literal
|
||
|
||
NATURAL_BOUNDARIES = [
|
||
"\n\n",
|
||
"\n",
|
||
"。",
|
||
"！",
|
||
"？",
|
||
"；",
|
||
"：",
|
||
". ",
|
||
"! ",
|
||
"? ",
|
||
]
|
||
|
||
|
||
class WeChatChunker:
|
||
def chunk_text(self, text: str, limit: int = 2048) -> list[str]:
|
||
if len(text) <= limit:
|
||
return [text]
|
||
|
||
chunks: list[str] = []
|
||
remaining = text
|
||
while len(remaining) > limit:
|
||
cut = self._find_boundary(remaining, limit)
|
||
if cut <= 0:
|
||
cut = limit
|
||
chunks.append(remaining[:cut])
|
||
remaining = remaining[cut:]
|
||
if remaining:
|
||
chunks.append(remaining)
|
||
return chunks
|
||
|
||
def chunk_markdown(self, text: str, limit: int = 2048) -> list[str]:
|
||
if len(text) <= limit:
|
||
return [text]
|
||
|
||
chunks: list[str] = []
|
||
remaining = text
|
||
code_block = False
|
||
|
||
while len(remaining) > limit:
|
||
if "```" in remaining[:limit]:
|
||
fence_idx = remaining.find("```")
|
||
if fence_idx >= 0:
|
||
code_block = not code_block
|
||
if code_block:
|
||
close_idx = remaining.find("```", fence_idx + 3)
|
||
block_end = close_idx + 3 if close_idx >= 0 else limit
|
||
if block_end <= limit:
|
||
cut = block_end
|
||
else:
|
||
cut = limit
|
||
else:
|
||
cut = self._find_boundary(remaining, limit)
|
||
chunks.append(remaining[:cut])
|
||
remaining = remaining[cut:]
|
||
continue
|
||
|
||
cut = self._find_boundary(remaining, limit)
|
||
if cut <= 0:
|
||
cut = limit
|
||
chunks.append(remaining[:cut])
|
||
remaining = remaining[cut:]
|
||
|
||
if remaining:
|
||
chunks.append(remaining)
|
||
return chunks
|
||
|
||
def chunk(self, text: str, limit: int, mode: Literal["text", "markdown"] = "text") -> list[str]:
|
||
if mode == "markdown":
|
||
return self.chunk_markdown(text, limit)
|
||
return self.chunk_text(text, limit)
|
||
|
||
@staticmethod
|
||
def _find_boundary(text: str, limit: int) -> int:
|
||
best = -1
|
||
for boundary in NATURAL_BOUNDARIES:
|
||
idx = text.rfind(boundary, 0, limit)
|
||
if idx > best:
|
||
best = idx
|
||
return best + 1 if best >= 0 else 0
|