backend/tests/test_excel_parser.py

import os
import pandas as pd
import pytest
from rag_pipeline.excel_parser import parse_excel_financials

@pytest.fixture
def sample_csv(tmp_path):
    df = pd.DataFrame({
        "Miesiąc": ["Styczeń", "Luty", "Marzec"],
        "Przychody": [10000, 15000, 20000],
        "EBITDA": [2000, 3000, 4000]
    })
    file_path = tmp_path / "finanse.csv"
    df.to_csv(file_path, index=False)
    return str(file_path)

@pytest.fixture
def sample_excel(tmp_path):
    df1 = pd.DataFrame({
        "Koszty": ["Marketing", "R&D", "IT"],
        "Wartość": [5000, 20000, 10000]
    })
    df2 = pd.DataFrame({
        "Przychody ze sprzedaży": [100000],
        "Zysk Netto": [25000]
    })
    file_path = tmp_path / "finanse.xlsx"
    with pd.ExcelWriter(file_path) as writer:
        df1.to_excel(writer, sheet_name="Koszty", index=False)
        df2.to_excel(writer, sheet_name="Wyniki", index=False)
    return str(file_path)

def test_parse_csv(sample_csv):
    result = parse_excel_financials(sample_csv)
    assert result["parser"] == "pandas_excel"
    assert "Przychody" in result["text"]
    assert "EBITDA" in result["text"]
    assert result["metadata"].get("has_ebitda") is True
    assert result["metadata"].get("has_revenue") is True

def test_parse_excel(sample_excel):
    result = parse_excel_financials(sample_excel)
    assert result["parser"] == "pandas_excel"
    assert "Koszty" in result["text"]
    assert "Wyniki" in result["text"]
    assert "Przychody" in result["text"]
    assert result["metadata"].get("has_revenue") is True
    assert result["metadata"].get("has_ebitda") is None

def test_parse_invalid_file(tmp_path):
    invalid_file = tmp_path / "broken.xlsx"
    invalid_file.write_text("not an excel file")
    result = parse_excel_financials(str(invalid_file))
    assert result["parser"] == "error"
    assert result["text"] == ""