Kris
a2aa782b86
新增Slack适配器全套核心模块，包括消息处理流水线、会话管理、配置适配、权限控制等完整功能： 1. 新增语音、视觉相关的TTS和图像分析导出接口 2. 实现消息预处理、路由、线程上下文处理的完整流水线 3. 新增账号管理、缓存机制、房间上下文提取功能 4. 支持Webhook和Socket Mode两种事件接收方式 5. 实现权限白名单、审批配对、自动状态管理功能 6. 新增配置迁移、作用域校验、重连策略等辅助模块
131 lines
3.6 KiB
Python
131 lines
3.6 KiB
Python
from __future__ import annotations
|
|
|
|
from dataclasses import dataclass
|
|
from enum import StrEnum
|
|
|
|
|
|
class ChunkMode(StrEnum):
|
|
LENGTH = "length"
|
|
NEWLINE = "newline"
|
|
|
|
|
|
@dataclass
|
|
class TextChunk:
|
|
text: str
|
|
index: int
|
|
is_last: bool
|
|
|
|
|
|
def convert_table_to_bullets(text: str) -> str:
|
|
lines = text.strip().split("\n")
|
|
if len(lines) < 2:
|
|
return text
|
|
if "|" not in lines[0] or "|" not in lines[1]:
|
|
return text
|
|
|
|
header_line = lines[0]
|
|
if "---" not in lines[1]:
|
|
return text
|
|
|
|
result_lines: list[str] = []
|
|
headers = [h.strip() for h in header_line.split("|") if h.strip()]
|
|
|
|
for line in lines[2:]:
|
|
if "|" not in line:
|
|
continue
|
|
cells = [c.strip() for c in line.split("|") if c.strip()]
|
|
parts = [f"*{headers[i]}:* {cells[i] if i < len(cells) else 'N/A'}" for i in range(len(headers))]
|
|
result_lines.append(" • " + " | ".join(parts))
|
|
|
|
return "\n".join(result_lines) if result_lines else text
|
|
|
|
|
|
def resolve_text_chunks(
|
|
text: str,
|
|
chunk_limit: int,
|
|
*,
|
|
mode: ChunkMode = ChunkMode.LENGTH,
|
|
convert_table: bool = True,
|
|
) -> list[TextChunk]:
|
|
if not text:
|
|
return []
|
|
|
|
processed = text
|
|
if convert_table:
|
|
processed = convert_table_to_bullets(processed)
|
|
|
|
if len(processed) <= chunk_limit:
|
|
return [TextChunk(text=processed, index=0, is_last=True)]
|
|
|
|
if mode == ChunkMode.NEWLINE:
|
|
return _chunk_by_newline(processed, chunk_limit)
|
|
return _chunk_by_length(processed, chunk_limit)
|
|
|
|
|
|
def _chunk_by_length(text: str, limit: int) -> list[TextChunk]:
|
|
chunks: list[TextChunk] = []
|
|
remaining = text
|
|
idx = 0
|
|
|
|
while remaining:
|
|
if len(remaining) <= limit:
|
|
chunks.append(TextChunk(text=remaining, index=idx, is_last=True))
|
|
break
|
|
split_point = remaining.rfind("\n", 0, limit)
|
|
if split_point < limit // 2:
|
|
split_point = remaining.rfind(" ", 0, limit)
|
|
if split_point < limit // 2:
|
|
split_point = limit
|
|
chunk_text = remaining[:split_point]
|
|
remaining = remaining[split_point:].lstrip("\n ")
|
|
is_last = not remaining
|
|
chunks.append(TextChunk(text=chunk_text, index=idx, is_last=is_last))
|
|
idx += 1
|
|
|
|
if chunks and len(chunks) > 1:
|
|
chunks[-1].is_last = True
|
|
|
|
return chunks
|
|
|
|
|
|
def _chunk_by_newline(text: str, limit: int) -> list[TextChunk]:
|
|
lines = text.split("\n")
|
|
chunks: list[TextChunk] = []
|
|
current_lines: list[str] = []
|
|
current_len = 0
|
|
idx = 0
|
|
|
|
for line in lines:
|
|
line_len = len(line) + 1
|
|
if current_lines and current_len + line_len > limit:
|
|
chunks.append(TextChunk(text="\n".join(current_lines), index=idx, is_last=False))
|
|
idx += 1
|
|
current_lines = []
|
|
current_len = 0
|
|
|
|
if line_len > limit:
|
|
if current_lines:
|
|
chunks.append(TextChunk(text="\n".join(current_lines), index=idx, is_last=False))
|
|
idx += 1
|
|
current_lines = []
|
|
current_len = 0
|
|
sub_chunks = _chunk_by_length(line, limit)
|
|
for sc in sub_chunks:
|
|
chunks.append(TextChunk(text=sc.text, index=idx, is_last=False))
|
|
idx += 1
|
|
continue
|
|
|
|
current_lines.append(line)
|
|
current_len += line_len
|
|
|
|
if current_lines:
|
|
chunks.append(TextChunk(text="\n".join(current_lines), index=idx, is_last=True))
|
|
elif chunks:
|
|
chunks[-1].is_last = True
|
|
|
|
return chunks
|
|
|
|
|
|
DEFAULT_CHUNK_LIMIT = 8000
|
|
DEFAULT_CHUNK_MODE = ChunkMode.LENGTH
|