diff --git a/dsRag/Start.py b/dsRag/Start.py
index ba49b059..6e56680e 100644
--- a/dsRag/Start.py
+++ b/dsRag/Start.py
@@ -15,6 +15,10 @@ from Config.Config import MS_MODEL_PATH, MS_MODEL_LIMIT, MS_HOST, MS_PORT, MS_MA
from Milvus.Utils.MilvusCollectionManager import MilvusCollectionManager
from Milvus.Utils.MilvusConnectionPool import *
from Milvus.Utils.MilvusConnectionPool import MilvusConnectionPool
+from docx import Document
+from docx.shared import Inches
+from io import BytesIO
+import html2text
# 初始化日志
logger = logging.getLogger(__name__)
@@ -156,6 +160,42 @@ http://10.10.21.22:8000/static/ai.html
class QueryRequest(BaseModel):
query: str = Field(..., description="用户查询的问题")
+class SaveWordRequest(BaseModel):
+ html: str = Field(..., description="要保存为Word的HTML内容")
+
+@app.post("/api/save-word")
+async def save_to_word(request: Request):
+ try:
+ data = await request.json()
+ save_request = SaveWordRequest(**data)
+ except ValidationError as e:
+ logger.error(f"请求体验证失败: {e.errors()}")
+ raise HTTPException(status_code=422, detail=e.errors())
+ except Exception as e:
+ logger.error(f"请求解析失败: {str(e)}")
+ raise HTTPException(status_code=400, detail="无效的请求格式")
+
+ # 转换HTML为纯文本
+ h = html2text.HTML2Text()
+ h.ignore_links = True
+ plain_text = h.handle(save_request.html)
+
+ # 创建Word文档
+ doc = Document()
+ doc.add_heading('小学数学问答', 0)
+ doc.add_paragraph(plain_text)
+
+ # 保存到内存中的字节流
+ file_stream = BytesIO()
+ doc.save(file_stream)
+ file_stream.seek(0)
+
+ return StreamingResponse(
+ file_stream,
+ media_type="application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+ headers={"Content-Disposition": "attachment; filename=小学数学问答.docx"}
+ )
+
@app.post("/api/rag")
async def rag_stream(request: Request):
try:
diff --git a/dsRag/Test/TestWriteDoc.py b/dsRag/Test/TestWriteDoc.py
new file mode 100644
index 00000000..dc8469e1
--- /dev/null
+++ b/dsRag/Test/TestWriteDoc.py
@@ -0,0 +1,44 @@
+from bs4 import BeautifulSoup
+from docx import Document
+"""
+pip install python-docx html2text beautifulsoup4
+"""
+
+def html_to_word(html_content, word_path):
+ # 解析 HTML
+ soup = BeautifulSoup(html_content, 'html.parser')
+
+ # 创建 Word 文档
+ doc = Document()
+
+ # 遍历 HTML 的所有段落
+ for element in soup.find_all(['p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'div']):
+ # 获取文本内容
+ text = element.get_text(strip=True)
+ if text:
+ # 添加到 Word 文档
+ doc.add_paragraph(text)
+
+ # 保存 Word 文档
+ doc.save(word_path)
+ print(f"HTML content saved to {word_path}")
+
+
+# 示例 HTML 内容
+html_content = """
+
+
+
+ Sample HTML
+
+
+ Heading 1
+ This is a paragraph.
+ Heading 2
+ Content inside a div.
+
+
+"""
+
+# 调用函数
+html_to_word(html_content, "output.docx")
\ No newline at end of file
diff --git a/dsRag/Test/__init__.py b/dsRag/Test/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/dsRag/Test/output.docx b/dsRag/Test/output.docx
new file mode 100644
index 00000000..e389e623
Binary files /dev/null and b/dsRag/Test/output.docx differ
diff --git a/dsRag/static/ai.html b/dsRag/static/ai.html
index ead942c5..b724b497 100644
--- a/dsRag/static/ai.html
+++ b/dsRag/static/ai.html
@@ -90,7 +90,7 @@
background-color: #d32f2f;
}
- #downloadBtn {
+ #saveWordBtn {
background-color: #4CAF50;
color: white;
border: none;
@@ -99,12 +99,20 @@
border-radius: 5px;
cursor: pointer;
transition: background-color 0.3s;
+ display: flex;
+ align-items: center;
+ gap: 8px;
}
- #downloadBtn:hover {
+ #saveWordBtn:hover {
background-color: #45a049;
}
+ #saveWordBtn::before {
+ content: "\1F4C4"; /* Word文档图标 */
+ font-size: 18px;
+ }
+
.status {
text-align: center;
margin-bottom: 20px;
@@ -163,7 +171,38 @@
+
+
+