feat: add remote upload ability and csv upload ability (#289)

* feat: add remote upload ability and csv upload ability

* feat: add remote upload ability and csv upload ability

* fix: TS error

* remove unnecessary files

pyproject.toml

@@ -40,10 +40,11 @@ lzstring = "^1.0.4"
 azure-ai-documentintelligence = { version = "^1.0.0b4", optional = true }
 litellm = "^1.51.0"
 pydantic = "^2.9.2"
+httpx = { version = "^0.28.1", optional = true }
 
 [tool.poetry.extras]
 parsing = ["python-docx", "openpyxl", "pydub", "python-pptx", "azure-ai-documentintelligence", "paddlepaddle", "pymupdf"]
-server = ["fastapi", "uvicorn", "docling", "azure-ai-formrecognizer", "azure-ai-documentintelligence"]
+server = ["fastapi", "uvicorn", "docling", "azure-ai-formrecognizer", "azure-ai-documentintelligence", "httpx"]
 
 [tool.poetry.group.dev.dependencies]
 pytest = "^8.3.2"

server/app/routes/filesystem.py

@@ -1,9 +1,13 @@
 from fastapi import APIRouter, UploadFile, File, Form, HTTPException
 from fastapi.responses import FileResponse, JSONResponse
-from typing import List, Optional
+from typing import List, Optional, Union
 import os
 import yaml
 import shutil
+import httpx
+import json
+import csv
+from io import StringIO
 from pathlib import Path
 from server.app.models import PipelineConfigRequest
 
@@ -33,19 +37,133 @@ async def check_namespace(namespace: str):
     except Exception as e:
         raise HTTPException(status_code=500, detail=f"Failed to check/create namespace: {str(e)}")
 
+def validate_json_content(content: bytes) -> None:
+    """Validate that content can be parsed as JSON"""
+    try:
+        json.loads(content)
+    except json.JSONDecodeError as e:
+        raise HTTPException(status_code=400, detail=f"Invalid JSON format: {str(e)}")
+
+def convert_csv_to_json(csv_content: bytes) -> bytes:
+    """Convert CSV content to JSON format"""
+    try:
+        # Decode bytes to string and create a StringIO object
+        csv_string = csv_content.decode('utf-8')
+        csv_file = StringIO(csv_string)
+
+        # Read CSV and convert to list of dictionaries
+        reader = csv.DictReader(csv_file)
+        data = list(reader)
+
+        if not data:
+            raise HTTPException(status_code=400, detail="CSV file is empty")
+
+        # Convert back to JSON bytes
+        return json.dumps(data).encode('utf-8')
+    except UnicodeDecodeError:
+        raise HTTPException(status_code=400, detail="Invalid CSV encoding")
+    except csv.Error as e:
+        raise HTTPException(status_code=400, detail=f"Invalid CSV format: {str(e)}")
+
+def is_likely_csv(content: bytes, filename: str) -> bool:
+    """Check if content is likely to be CSV based on content and filename"""
+    # Check filename extension
+    if filename.lower().endswith('.csv'):
+        return True
+
+    # If no clear extension, try to detect CSV content
+    try:
+        # Take first line and check if it looks like CSV
+        first_line = content.split(b'\n')[0].decode('utf-8')
+        # Check if line contains commas and no obvious JSON characters
+        return ',' in first_line and not any(c in first_line for c in '{}[]')
+    except:
+        return False
+
 @router.post("/upload-file")
-async def upload_file(file: UploadFile = File(...), namespace: str = Form(...)):
-    """Upload a single file to the namespace files directory"""
+async def upload_file(
+    file: Optional[UploadFile] = File(None),
+    url: Optional[str] = Form(None),
+    namespace: str = Form(...)
+):
+    """Upload a file to the namespace files directory, either from a direct upload or a URL"""
     try:
+        if not file and not url:
+            raise HTTPException(status_code=400, detail="Either file or url must be provided")
+
         upload_dir = get_namespace_dir(namespace) / "files"
         upload_dir.mkdir(parents=True, exist_ok=True)
 
-        file_path = upload_dir / file.filename
-        with file_path.open("wb") as f:
-            shutil.copyfileobj(file.file, f)
+        if url:
+            # Get filename from URL or default to dataset.json
+            filename = url.split("/")[-1] or "dataset.json"
+
+            file_path = upload_dir / filename.replace('.csv', '.json')
+
+            # Handle URL download
+            async with httpx.AsyncClient() as client:
+                async with client.stream(
+                    'GET',
+                    url,
+                    follow_redirects=True,
+                ) as response:
+                    if response.status_code != 200:
+                        raise HTTPException(
+                            status_code=400,
+                            detail=f"Failed to download from URL: {response.status_code}"
+                        )
+
+                    # Save the file in chunks
+                    content_chunks = []
+                    async for chunk in response.aiter_bytes(chunk_size=8192):
+                        if chunk:  # filter out keep-alive new chunks
+                            content_chunks.append(chunk)
+
+                    # Combine chunks
+                    content = b''.join(content_chunks)
+
+                    # Check if content is CSV and convert if needed
+                    if is_likely_csv(content, filename):
+                        try:
+                            content = convert_csv_to_json(content)
+                        except HTTPException as e:
+                            raise HTTPException(
+                                status_code=400,
+                                detail=f"Failed to convert CSV to JSON: {str(e.detail)}"
+                            )
+
+                    # Validate JSON content
+                    validate_json_content(content)
+
+                    # Write to file
+                    with file_path.open("wb") as f:
+                        f.write(content)
+        else:
+            # Handle direct file upload
+            file_content = await file.read()
+
+            # Check if content is CSV and convert if needed
+            if file.filename.lower().endswith('.csv'):
+                try:
+                    file_content = convert_csv_to_json(file_content)
+                except HTTPException as e:
+                    raise HTTPException(
+                        status_code=400,
+                        detail=f"Failed to convert CSV to JSON: {str(e.detail)}"
+                    )
+
+            # Validate JSON content
+            validate_json_content(file_content)
+
+            # Always save as .json
+            file_path = upload_dir / file.filename.replace('.csv', '.json')
+            with file_path.open("wb") as f:
+                f.write(file_content)
 
         return {"path": str(file_path)}
     except Exception as e:
+        if isinstance(e, HTTPException):
+            raise e
         raise HTTPException(status_code=500, detail=f"Failed to upload file: {str(e)}")
 
 @router.post("/save-documents")