from fastapi import FastAPI, UploadFile, File, HTTPException
from fastapi.responses import HTMLResponse, JSONResponse
from fastapi.middleware.cors import CORSMiddleware
from pydantic import BaseModel
import pandas as pd
import openpyxl
import numpy as np
from sentence_transformers import SentenceTransformer
from typing import List
import io
app = FastAPI()
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
allow_methods=["*"],
allow_headers=["*"]
)
model = SentenceTransformer('all-MiniLM-L6-v2')
HTML_CONTENT = """
Analytics Dashboard
"""
class QueryRequest(BaseModel):
query: str
embeddings: List[List[float]]
@app.get("/")
async def root():
return HTMLResponse(HTML_CONTENT)
@app.post("/api/process-file")
async def process_file(file: UploadFile = File(...)):
if not file.filename:
raise HTTPException(status_code=400, detail="No file provided")
try:
contents = await file.read()
if file.filename.endswith('.xlsx'):
wb = openpyxl.load_workbook(io.BytesIO(contents), read_only=True)
sheet = wb.active
data = []
headers = next(sheet.rows)
header_values = [cell.value for cell in headers]
for row in sheet.rows:
if row != headers:
data.append({h: cell.value for h, cell in zip(header_values, row)})
df = pd.DataFrame(data)
elif file.filename.endswith('.csv'):
df = pd.read_csv(io.BytesIO(contents))
else:
raise HTTPException(status_code=400, detail="Unsupported file type")
# Convert timestamps to strings
for col in df.select_dtypes(include=['datetime64[ns]']).columns:
df[col] = df[col].astype(str)
df = df.replace({np.nan: None})
text_reps = [" ".join(f"{col}: {str(val)}" for col, val in row.items() if val is not None)
for _, row in df.iterrows()]
embeddings = model.encode(text_reps)
# Convert DataFrame to JSON-serializable format
raw_data = df.to_dict('records')
# Convert any non-serializable types to strings
for record in raw_data:
for key, value in record.items():
if not isinstance(value, (str, int, float, bool, type(None))):
record[key] = str(value)
return JSONResponse({
'embeddings': embeddings.tolist(),
'metadata': {
'columns': list(df.columns),
'row_count': len(df)
},
'raw_data': raw_data
})
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
@app.post("/api/query")
async def query_data(request: QueryRequest):
try:
query_embedding = model.encode([request.query])[0]
similarities = np.dot(request.embeddings, query_embedding)
return JSONResponse({"similar_indices": np.argsort(similarities)[-5:][::-1].tolist()})
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))