fix: strip markdown links, URLs, bold/italic from antaf response
Antaf returns markdown with hyperlinks and formatting that TTS reads out as raw text. Now cleaned before sending to TTS. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
parent
6707351540
commit
49ae06ae45
|
|
@ -40,13 +40,24 @@ class LLMProvider(LLMProviderBase):
|
||||||
|
|
||||||
@staticmethod
|
@staticmethod
|
||||||
def _clean_text(text):
|
def _clean_text(text):
|
||||||
"""清理阿福返回文本中的脏数据"""
|
"""清理阿福返回文本中的脏数据、链接、markdown"""
|
||||||
|
import re
|
||||||
# 去掉阿福内部状态文本
|
# 去掉阿福内部状态文本
|
||||||
junk = [
|
junk = [
|
||||||
"完成资料引用", "内容生成", "正在思考", "正在搜索",
|
"完成资料引用", "内容生成", "正在思考", "正在搜索",
|
||||||
]
|
]
|
||||||
for j in junk:
|
for j in junk:
|
||||||
text = text.replace(j, "")
|
text = text.replace(j, "")
|
||||||
|
# Markdown链接 [文字](url) → 只保留文字
|
||||||
|
text = re.sub(r'\[([^\]]+)\]\([^)]+\)', r'\1', text)
|
||||||
|
# 裸URL
|
||||||
|
text = re.sub(r'https?://\S+', '', text)
|
||||||
|
# Markdown加粗 **文字** → 文字
|
||||||
|
text = re.sub(r'\*\*([^*]+)\*\*', r'\1', text)
|
||||||
|
# Markdown斜体 *文字* → 文字
|
||||||
|
text = re.sub(r'\*([^*]+)\*', r'\1', text)
|
||||||
|
# 多余空格
|
||||||
|
text = re.sub(r' +', ' ', text)
|
||||||
return text.strip()
|
return text.strip()
|
||||||
|
|
||||||
@staticmethod
|
@staticmethod
|
||||||
|
|
|
||||||
Loading…
Reference in New Issue