malayalam_wiki.py

import os
import re

import datasets


logger = datasets.logging.get_logger(__name__)

_DESCRIPTION = """\
Common Crawl - Malayalam.
"""

_CITATION = """\
@article{qburst,
    title={Common Crawl - Malayalam},
    author={n.d},
    year={2020},
    journal={n.d},
}
"""

_URLs = {
 "malayalam_wiki_2020": "https://huggingface.co/datasets/rajeshradhakrishnan/malayalam_2020_wiki/resolve/main/",
 "checksum_url": "https://huggingface.co/datasets/rajeshradhakrishnan/malayalam_2020_wiki/resolve/main/ml_sha256.txt"

}


class MalayalamWikiConfig(datasets.BuilderConfig):
    """BuilderConfig for MalayalamWiki."""

    def __init__(self, **kwargs):
        """BuilderConfig for MalayalamWiki.

        Args:
            **kwargs: keyword arguments forwarded to super.
        """
        super(MalayalamWikiConfig, self).__init__(**kwargs)

class MalayalamWiki(datasets.GeneratorBasedBuilder):
    """Malayalam News topic classification dataset."""

    VERSION = datasets.Version("1.0.0")

    BUILDER_CONFIGS = [
        MalayalamWikiConfig(
            name="malayalam_wiki_2020", version=VERSION, description="Common Crawl - Malayalam."
        ),
    ]


    def remove_special_characters(self, txt):
        chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\%\‘\”\�Utrnle\_]'
        unicode_ignore_regex = r'[\u200e\u200c\u200d]'
        english_ignore_regex = r'[a-zA-Z]'
        
        txt = txt.strip()
        txt = re.sub(chars_to_ignore_regex, '',txt)
        txt = re.sub(unicode_ignore_regex, '',txt) + " "
        txt = re.sub(english_ignore_regex, '',txt) + " "
        return txt

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "text": datasets.Value("string")
                }
            ),
            supervised_keys=None,
            homepage="https://github.com/qburst/common-crawl-malayalam",
            citation=_CITATION
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        checksum_url = _URLs["checksum_url"]
        checksum_file = dl_manager.download(checksum_url)
        with open(checksum_file, encoding="utf-8") as f:
            data_filenames = [line.strip() for line in f if line]
            data_urls = [_URLs["malayalam_wiki_2020"] + data_filename for data_filename in data_filenames[1:2]]
        downloaded_files = dl_manager.download(data_urls)
        return [
            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepaths": downloaded_files}),
        ]

    def _generate_examples(self, filepaths):
        """This function returns the examples in the raw (text) form by iterating on all the files."""
        for file_id,filepath in enumerate(filepaths):
            logger.info("generating examples from = %s", filepath)
            with open(filepath, encoding="utf-8") as f:
                  for row_id, row in enumerate(f):
                      yield  f"{file_id}_{row_id}", {"text": self.remove_special_characters(row).strip()}