2026-05-22 10:42:31 +08:00
|
|
|
|
from __future__ import annotations
|
|
|
|
|
|
|
|
|
|
|
|
import re
|
|
|
|
|
|
|
|
|
|
|
|
KNOWLEDGE_DIRECT_ANSWER_HINTS = (
|
|
|
|
|
|
"是什么",
|
2026-05-24 21:44:17 +08:00
|
|
|
|
"介绍",
|
|
|
|
|
|
"说明",
|
|
|
|
|
|
"概述",
|
2026-05-22 10:42:31 +08:00
|
|
|
|
"标准",
|
|
|
|
|
|
"限额",
|
|
|
|
|
|
"流程",
|
|
|
|
|
|
"条件",
|
|
|
|
|
|
"规则",
|
|
|
|
|
|
"怎么",
|
|
|
|
|
|
"如何",
|
|
|
|
|
|
"哪些",
|
|
|
|
|
|
"需要",
|
|
|
|
|
|
"是否",
|
|
|
|
|
|
"区别",
|
|
|
|
|
|
"范围",
|
|
|
|
|
|
"额度",
|
|
|
|
|
|
"金额",
|
|
|
|
|
|
"多少",
|
|
|
|
|
|
"多少钱",
|
|
|
|
|
|
"上限",
|
|
|
|
|
|
)
|
|
|
|
|
|
KNOWLEDGE_QUERY_STOPWORDS = {
|
|
|
|
|
|
"什么",
|
|
|
|
|
|
"多少",
|
|
|
|
|
|
"哪些",
|
|
|
|
|
|
"怎么",
|
|
|
|
|
|
"如何",
|
|
|
|
|
|
"请问",
|
|
|
|
|
|
"一下",
|
|
|
|
|
|
"关于",
|
|
|
|
|
|
"规定",
|
|
|
|
|
|
"标准",
|
|
|
|
|
|
"可以",
|
|
|
|
|
|
"是否",
|
|
|
|
|
|
"一个",
|
|
|
|
|
|
"哪些人",
|
|
|
|
|
|
"目前",
|
|
|
|
|
|
"当前",
|
|
|
|
|
|
"一下子",
|
|
|
|
|
|
}
|
|
|
|
|
|
MAX_KNOWLEDGE_QUERY_TERMS = 12
|
|
|
|
|
|
MAX_KNOWLEDGE_DIRECT_EVIDENCE = 4
|
|
|
|
|
|
MAX_KNOWLEDGE_MODEL_HITS = 5
|
|
|
|
|
|
KNOWLEDGE_SECTION_HEADING_PATTERN = re.compile(
|
2026-05-24 21:44:17 +08:00
|
|
|
|
r"^(#\s*.+|##\s*.+|###\s*.+|第[一二三四五六七八九十百零0-9]+[部分章节条]\s*.*|[一二三四五六七八九十]+、.*|（[一二三四五六七八九十]+）.*|\([一二三四五六七八九十]+\).*)$"
|
2026-05-22 10:42:31 +08:00
|
|
|
|
)
|
|
|
|
|
|
KNOWLEDGE_LIST_ITEM_PATTERN = re.compile(r"^[-*•]\s+.+$")
|
|
|
|
|
|
KNOWLEDGE_NUMBERED_ITEM_PATTERN = re.compile(
|
|
|
|
|
|
r"^(?:(?:\d+[.)、])|(?:[（(][一二三四五六七八九十百零0-9]+[)）])|[①②③④⑤⑥⑦⑧⑨⑩])\s*.+$"
|
|
|
|
|
|
)
|
|
|
|
|
|
KNOWLEDGE_ARTICLE_PATTERN = re.compile(r"^(第[一二三四五六七八九十百零0-9]+条)\s*.*$")
|
