Multilingual-SBD.py

import datasets
from huggingface_hub.file_download import hf_hub_url
import glob
import json
import pandas as pd

try:
    import lzma as xz
except ImportError:
    import pylzma as xz
datasets.logging.set_verbosity_info()
logger = datasets.logging.get_logger(__name__)

_DESCRIPTION ="""\

    """

_HOMEPAGE =  ""

_LICENSE = ""

_CITATION = ""

_URL = {
    'data/'
}
_LANGUAGES = [
    "fr","it","es","en","de","pt"
]

_TYPES = [
    "laws", "judgements"
]

_SOURCES = [
    "MultiLegalPile", "Wipolex", "Jug", "BVA", "CC", "IP", "SCOTUS", "SwissJudgementPrediction"
    "Gesetz", "Constitution", "CivilCode", "CriminalCode",
]


"""
see https://huggingface.co/datasets/joelito/MultiLegalPile_Wikipedia_Filtered/blob/main/MultiLegalPile_Wikipedia_Filtered.py
"""
_HIGHEST_NUMBER_OF_SHARDS = 4
class MultilingualSBDConfig(datasets.BuilderConfig):
    
    def __init__(self, name:str, **kwargs):
        super( MultilingualSBDConfig, self).__init__(**kwargs)
        self.name = name
        self.language = name.split("_")[0]
        self.type = name.split("_")[1]
        #self.source = name.split("_")[2]

class MultilingualSBD(datasets.GeneratorBasedBuilder):

    
    BUILDER_CONFIG_CLASS = MultilingualSBDConfig
    
    BUILDER_CONFIGS = [
        MultilingualSBDConfig(f"{language}_{type}")
        for language in _LANGUAGES + ['all']
        for type in _TYPES  + ["all"]
    ]
    DEFAULT_CONFIG_NAME = 'all_all'
    def _info(self):
        features = datasets.Features(
            {   
                #"language": datasets.Value("string"), #in _LANGUAGES
                #"type": datasets.Value("string"), # in _TYPES
                #"subtype": datasets.Value("string"), # in _SUBTYPES
                "text": datasets.Value("string"),
                "source": datasets.Value("string"),
                "spans": [
                    {
                        "start": datasets.Value("int64"),
                        "end": datasets.Value("int64"),
                        "token_start": datasets.Value("int64"),
                        "token_end": datasets.Value("int64"),
                        "label": datasets.Value("string")
                    }
                ],
                "tokens": [
                    {
                        "text": datasets.Value("string"),
                        "start": datasets.Value("int64"),
                        "end": datasets.Value("int64"),
                        "id": datasets.Value("int64"),
                        "ws": datasets.Value("bool")

                    }
                ]
            }
        )
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features = features,
            homepage = _HOMEPAGE,
            citation=_CITATION
        )
    
    def _split_generators(self, dl_manager):
        def download_url(filename):
            url = hf_hub_url(
                repo_id="tbrugger/Multilingual-SBD",
                filename = f'data/{filename}.jsonl.xz',
                repo_type='dataset'
                )
            return dl_manager.download(url)
        languages = _LANGUAGES if self.config.language == "all" else [self.config.language]
        types = _TYPES if self.config.type == 'all' else [self.config.type]
        #sources = _SOURCES if self.config.source == 'all' else [self.config.source]

        split_generators = []
        for split in [datasets.Split.TRAIN]:
            filepaths = []
            for language in languages:
                for type in types:
                    for shard in range(_HIGHEST_NUMBER_OF_SHARDS):
                        try:
                            filepaths.append(download_url(f'{language}_{type}_{shard}'))
                        except:
                            break

            split_generators.append(
                datasets.SplitGenerator(name=split, gen_kwargs={'filepaths': filepaths})
            )
            
        return split_generators     

    def _generate_examples(self,filepaths):
        id_ = 0
        for filepath in filepaths:
            if filepath:
                logger.info("Generating examples from = %s", filepath)
                try:
                    with xz.open(open(filepath,'rb'), 'rt', encoding='utf-8') as f:
                        json_list = list(f)
                    
                    for json_str in json_list:
                        example = json.loads(json_str)
                        if example is not None and isinstance(example, dict):
                            yield id_, example
                            id_ +=1

                except Exception:
                    logger.exception("Error while processing file %s", filepath)