Files
doc_processer/app/api/v1/endpoints/convert.py
2026-02-04 12:35:14 +08:00

67 lines
2.4 KiB
Python

"""Format conversion endpoints."""
from fastapi import APIRouter, Depends, HTTPException
from fastapi.responses import Response
from app.core.dependencies import get_converter
from app.schemas.convert import MarkdownToDocxRequest, LatexToOmmlRequest, LatexToOmmlResponse
from app.services.converter import Converter
router = APIRouter()
@router.post("/file")
async def convert_markdown_to_docx(
request: MarkdownToDocxRequest,
converter: Converter = Depends(get_converter),
) -> Response:
"""Convert markdown content to DOCX file.
Returns the generated DOCX file as a binary response.
"""
try:
docx_bytes = converter.export_to_file(request.markdown, export_type="docx")
return Response(
content=docx_bytes,
media_type="application/vnd.openxmlformats-officedocument.wordprocessingml.document",
headers={"Content-Disposition": f'attachment; filename="{request.filename}.docx"'},
)
except Exception as e:
raise HTTPException(status_code=500, detail=f"Conversion failed: {e}")
@router.post("/latex-to-omml", response_model=LatexToOmmlResponse)
async def convert_latex_to_omml(
request: LatexToOmmlRequest,
converter: Converter = Depends(get_converter),
) -> LatexToOmmlResponse:
"""Convert LaTeX formula to OMML (Office Math Markup Language).
OMML is the math format used by Microsoft Word and other Office applications.
This endpoint is separate from the main OCR endpoint due to the performance
overhead of OMML conversion (requires creating a temporary DOCX file).
Args:
request: Contains the LaTeX formula to convert (without $ or $$ delimiters).
Returns:
OMML representation of the formula.
Example:
```bash
curl -X POST "http://localhost:8000/api/v1/convert/latex-to-omml" \\
-H "Content-Type: application/json" \\
-d '{"latex": "\\\\frac{a}{b} + \\\\sqrt{c}"}'
```
"""
if not request.latex or not request.latex.strip():
raise HTTPException(status_code=400, detail="LaTeX formula cannot be empty")
try:
omml = converter.convert_to_omml(request.latex)
return LatexToOmmlResponse(omml=omml)
except ValueError as e:
raise HTTPException(status_code=400, detail=str(e))
except RuntimeError as e:
raise HTTPException(status_code=503, detail=str(e))