main.py

from fastapi import FastAPI, UploadFile, File
from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import JSONResponse
from lxml import etree
from pathlib import Path
import time

app = FastAPI(title="NeTEx Validator CH", version="1.0.0")

app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],
    allow_methods=["POST", "GET"],
    allow_headers=["*"],
)

# Load XSD once at startup
XSD_PATH = Path(__file__).parent / "xsd" / "NeTEx_publication.xsd"
_schema = None

def get_schema() -> etree.XMLSchema:
    global _schema
    if _schema is None:
        _schema = etree.XMLSchema(etree.parse(str(XSD_PATH)))
    return _schema


@app.on_event("startup")
async def startup():
    """Pre-load schema at startup so first request is fast."""
    try:
        get_schema()
        print("XSD schema loaded OK")
    except Exception as e:
        print(f"XSD load error: {e}")


@app.get("/health")
def health():
    return {"status": "ok", "xsd": str(XSD_PATH.exists())}


@app.post("/validate")
async def validate(file: UploadFile = File(...)):
    t0 = time.time()
    try:
        content = await file.read()

        # Parse XML
        try:
            doc = etree.fromstring(content)
        except etree.XMLSyntaxError as e:
            return JSONResponse({
                "valid": False,
                "filename": file.filename,
                "errors": [{
                    "line": e.lineno,
                    "col":  e.offset,
                    "message": str(e.msg),
                    "level": "error",
                }],
                "duration_ms": int((time.time()-t0)*1000),
            })

        # Validate against XSD
        schema = get_schema()
        valid  = schema.validate(doc)
        errors = []
        for e in schema.error_log:
            errors.append({
                "line":    e.line,
                "col":     e.column,
                "message": e.message,
                "level":   e.level_name.lower(),  # "error" or "warning"
            })

        return JSONResponse({
            "valid":       valid,
            "filename":    file.filename,
            "errors":      errors,
            "error_count": len(errors),
            "duration_ms": int((time.time()-t0)*1000),
        })

    except Exception as e:
        return JSONResponse({"valid": False, "errors": [{"message": str(e), "level": "error"}]}, status_code=500)