mqa.py

# coding=utf-8
# Copyright 2020 HuggingFace Datasets Authors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

# Lint as: python3
import os
import gzip
import json

import datasets


_DESCRIPTION = """MQA is a multilingual corpus of questions and answers parsed from the Common Crawl. Questions are divided between Frequently Asked Questions (FAQ) pages and Community Question Answering (CQA) pages."""
_HOMEPAGE_URL = "https://huggingface.co/datasets/clips/mqa"
_CITATION = """
@misc{debruyn2021mfaq,
      title={MFAQ: a Multilingual FAQ Dataset}, 
      author={Maxime {De Bruyn} and Ehsan Lotfi and Jeska Buhmann and Walter Daelemans},
      year={2021},
      booktitle={MRQA@EMNLP2021},
}
"""

_VERSION = "0.1"
_BASE_NAME = ""
_BASE_URL = "data/data.{}.{}.json.gz"


_LANGUAGES = [
    "ca", "en", "de", "es", "fr",
    "ru", "ja", "it", "zh", "pt",
    "nl", "tr", "pl", "vi", "ar",
    "id", "uk", "ro", "no", "th",
    "sv", "el", "fi", "he", "da",
    "cs", "ko", "fa", "hi", "hu",
    "sk", "lt", "et", "hr", "is",
    "lv", "ms", "bg", "sr", 
]
_SCOPES = ["faq", "cqa"]
_LEVELS = ["domain", "page", "question"]


class MQAConfig(datasets.BuilderConfig):
    def __init__(self, *args, language="en", scope="all", level="question", **kwargs):
        super().__init__(
            *args,
            name=f"{language}-{scope}-{level}",
            **kwargs,
        )
        self.language = language
        self.scope = scope
        self.level = level


class MQA(datasets.GeneratorBasedBuilder):
    BUILDER_CONFIGS = []
    for language in _LANGUAGES:
        for scope in _SCOPES:
            for level in _LEVELS:
                BUILDER_CONFIGS.append(MQAConfig(language=language, scope=scope, level=level))
    for language in _LANGUAGES:
        BUILDER_CONFIGS.append(MQAConfig(language=language, scope="all", level=level))
    for scope in _SCOPES:
        BUILDER_CONFIGS.append(MQAConfig(language="all", scope=scope, level=level))
    BUILDER_CONFIG_CLASS = MQAConfig

    def _info(self):
        question = {
            "id": datasets.Value("string"),
            "text": datasets.Value("string"),
            "name": datasets.Value("string"),
            "domain": datasets.Value("string"),
            "bucket": datasets.Value("string"),
            "answers": [{
                "text": datasets.Value("string"),
                "name": datasets.Value("string"),
                "is_accepted": datasets.Value("bool"),
            }]                        
        }
        page = {
            "id": datasets.Value("string"),
            "bucket": datasets.Value("string"),
            "domain": datasets.Value("string"),
            "questions": [question]                        
        }
        domain = {
            "domain": datasets.Value("string"),
            "pages": [page]
        }
        if self.config.level == "question":
            features = question
        elif self.config.level == "page":
            features = page
        elif self.config.level == "domain":
            features = domain
        else:
            raise NotImplementedError()
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(features),
            supervised_keys=None,
            homepage=_HOMEPAGE_URL,
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        filenames = []
        languages = _LANGUAGES if self.config.language == "all" else [self.config.language]
        scopes = _SCOPES if self.config.scope == "all" else [self.config.scope]
        for language in languages:
            for scope in scopes:
                path = dl_manager.download_and_extract(_BASE_URL.format(language, scope))
                filenames.append(path)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={"filenames": filenames},
            )
        ]

    def _generate_examples(self, filenames):

        def default(e, key, default_value=""):
            if e[key] is None:
                return default_value
            return e[key]

        for filename in filenames:
            with open(filename, "r") as f:
                domain = []
                previous_domain = ''
                for line in f:
                    page = json.loads(line)
                    questions = [{
                        "text": default(question, "text"),
                        "name": default(question, "name"),
                        "domain": page["domain"],
                        "bucket": page["bucket"],
                        "id": question["hash"],
                        "answers": [{
                            "text": default(answer, "text"),
                            "name": default(answer, "name"),
                            "is_accepted": answer["is_accepted"]
                        } for answer in question["answers"]]
                    } for question in page["questions"]]
                    page = {
                        "id": page["page_hash"],
                        "domain": page["domain"],
                        "bucket": page["bucket"],
                        "questions": questions
                    }
                    if self.config.level == "question":
                        for question in questions:
                            yield question["id"], question
                    if self.config.level == "page":
                        yield page["id"], page
                    if self.config.level == "domain":
                        if page["domain"] == previous_domain or previous_domain == "":
                            domain.append(page)
                        else:
                            yield previous_domain, {
                                "domain": previous_domain,
                                "pages": domain
                            }
                            domain = []
                        previous_domain = page["domain"]