import datasets

_CITATION = """\
@InProceedings{huggingface:dataset,
title = {Small audio-text set},
author={James Briggs},
year={2022}
}
"""

_DESCRIPTION = """\
Demo dataset for testing or showing audio-text capabilities.
"""
##_HOMEPAGE = "https://huggingface.co/datasets/jamescalam/audio-text-demo"
_HOMEPAGE = "https://huggingface.co/datasets/lucasjca/audio-files"

_LICENSE = ""

#_REPO = "https://huggingface.co/datasets/jamescalam/audio-text-demo"
_REPO = "https://huggingface.co/datasets/lucasjca/audio-files"


descriptions =['ACOLHIMENTO NOTURNO DE PACIENTE EM CENTRO DE ATENÇÃO PSICOSSOCIAL',
               'ADALIMUMABE 40 MG INJETÁVEL (POR SERINGA PREENCHIDA)',
               'ADALIMUMABE 40 MG INJETAVEL (POR SERINGA PREENCHIDA)',
               'ADALIMUMABE 40 MG INJETÁVEL (FRASCO AMPOLA)']


class audioSet(datasets.GeneratorBasedBuilder):
    """Small sample of audio-text pairs"""

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    'text': datasets.Value("string"),
                    'audio': datasets.audio(),
                }
            ),
            supervised_keys=None,
            homepage=_HOMEPAGE,
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        audios_archive = dl_manager.download(f"{_REPO}/resolve/main/audios.tgz")
        audio_iters = dl_manager.iter_archive(audios_archive)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={
                    "audios": audio_iters
                }
            ),
        ]

    def _generate_examples(self, audios):
        """ This function returns the examples in the raw (text) form."""
        
        for idx, (filepath, audio) in enumerate(audios):
            description = filepath.split('/')[-1][:-4]
            description = description.replace('_', ' ')
            yield idx, {
                "audio": {"path": filepath, "bytes": audio.read()},
                "text": description,
            }