import itertools
import unittest
from pprint import pprint
from typing import List
from dataclasses import asdict

from gec_worker.utils import generate_spans
from gec_worker.dataclasses import Correction


def apply_corrections(original: str, corrections: List[Correction]):
    idx = 0
    text = ''
    for correction in corrections:
        if correction.replacements[0].value is None:
            value = ''
        else:
            value = correction.replacements[0].value
        text += original[idx:correction.span.start] + value
        idx = correction.span.end
    text += original[idx:]

    return text


class Test(unittest.TestCase):
    def test_model(self):
        """
        Test that the model can be loaded and used.
        :return:
        """
        from gec_worker import GEC, read_model_config
        from gec_worker.dataclasses import Response, Request

        model_config = read_model_config('models/config.yaml')
        gec = GEC(model_config)

        request = Request(text="Juku joksis koolis. Aitüma.", language='et')
        response = gec.process_request(request)
        pprint(asdict(response))
        self.assertIsInstance(response, Response)


    def test_spell_model(self):
        """
        Test that the model can be loaded and used.
        :return:
        """
        from gec_worker import spelling
        from gec_worker.dataclasses import Response, Request

#        spelling1 = spelling.Spelling("etnc19_web_2019.bin")
        spelling1 = spelling.Spelling("etnc19_reference_corpus_6000000_web_2019_600000.bin")
#        spelling1 = spelling.Spelling("etnc19_reference_corpus_model_6000000_lines.bin")
#        spelling1 = spelling.Spelling("etnc19_wikipedia2017.bin")

        request = Request(text="Juku joksis koolis. Aitüma.", language='et')
        response = spelling1.process_request(request)
        pprint(asdict(response))
        self.assertIsInstance(response, Response)


    def test_multiple_correction_model(self):
        """
        Test that the model can be loaded and used.
        :return:
        """
        from gec_worker import GEC, read_model_config
        from gec_worker import spelling
        from gec_worker.dataclasses import Response, Request
        from gec_worker import multiple_corrections
        model_config = read_model_config('models/config.yaml')
        gec = GEC(model_config)
        spelling1 = spelling.Spelling("etnc19_reference_corpus_6000000_web_2019_600000.bin")
        model_list=multiple_corrections.MultipleCorrections()
        model_list.add_corrector(spelling1)
        model_list.add_corrector(gec)
        request = Request(text="Juku joksis koolis. Aitüma.", language='et')
        response = model_list.process_request(request)
        pprint(asdict(response))
        self.assertIsInstance(response, Response)




    def test_generate_spans(self):
        """
        Test that correct differences are detected and the sentence can be restored to the target.
        :return:
        """
        a = "Lapsed peavad näitama paremaid tulemusi, rahuldada vanemate, õetajate ja ühiskonna mis tõttu " \
            "nõyrdfujudmisi, kuid kõik see võib tekitada lapses  stressi."
        b = "Lapsed jah peavad näitama häid ka tulemusi, et vanemate, õpetajae ja ühiskonna nõudmisi, " \
            "kuiyhfd kõik see võib tekitada lapses stressi."

        # combining sentences to test differences in both directions
        delimiters = ['', '\n\n', '']
        source = ''.join(itertools.chain.from_iterable(zip(delimiters, [a, b]))) + delimiters[-1]
        target = ''.join(itertools.chain.from_iterable(zip(delimiters, [b, a]))) + delimiters[-1]

        corrections = generate_spans([a, b], [b, a], ['', '\n\n', ''])
        pprint(corrections)
        restored = apply_corrections(source, corrections)
        self.assertEqual(restored, target)


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