import unittest
import os
import shutil
from scrape_script import append_to_parquet
import pandas as pd

class TestAppendToParquet(unittest.TestCase):
    def setUp(self):
        self.temp_dir = "temp_data"
        if not os.path.exists(self.temp_dir):
            os.makedirs(self.temp_dir)

    def tearDown(self):
        shutil.rmtree(self.temp_dir)

    def test_cleaning_rules(self):
        content = (
            "<span>Test text with HTML tags.</span> "
            "https://example.com\n"
            "Chapter 1\n"
            "Halakhah 1\n"
            "<b>bold text</b> <strong>strong text</strong> <small>small text</small>"
        )
        file_path = "sample_file.txt"
        lang = "en"
        top_level_directory = "test"

        append_to_parquet(content, file_path, lang, top_level_directory, data_dir=self.temp_dir)

        parquet_file = os.path.join(self.temp_dir, f"train_{top_level_directory}_english.parquet")
        self.assertTrue(os.path.exists(parquet_file))

        df = pd.read_parquet(parquet_file)
        self.assertEqual(len(df), 1)

        cleaned_content = df.loc[0, "text"]
        expected_cleaned_content = (
            "Test text with HTML tags. "
            "\n"
            "bold text strong text small text"
        )
        cleaned_content_lines = [line.strip() for line in cleaned_content.split("\n") if line.strip()]
        expected_cleaned_content_lines = [line.strip() for line in expected_cleaned_content.split("\n") if line.strip()]

        self.assertEqual(len(cleaned_content_lines), len(expected_cleaned_content_lines))

        for clean_line, expected_line in zip(cleaned_content_lines, expected_cleaned_content_lines):
            self.assertEqual(clean_line, expected_line)



if __name__ == "__main__":
    unittest.main()