codequeries.py

# coding=utf-8
# Copyright 2022 CodeQueries Authors and the 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.

"""The CodeQueries benchmark."""


import json
import os

import datasets

logger = datasets.logging.get_logger(__name__)


_CODEQUERIES_CITATION = """\
@article{codequeries2022,
  title={Learning to Answer Semantic Queries over Code},
  author={A, B, C, D, E, F},
  journal={arXiv preprint arXiv:<.>},
  year={2022}
}
"""

_IDEAL_DESCRIPTION = """\
CodeQueries Ideal setup.

"""

_PREFIX_DESCRIPTION = """\
CodeQueries Prefix setup."""

_FILE_IDEAL_DESCRIPTION = """\
CodeQueries File level Ideal setup."""

_TWOSTEP_DESCRIPTION = """\
CodeQueries Twostep setup."""


class CodequeriesConfig(datasets.BuilderConfig):
    """BuilderConfig for Codequeries."""

    def __init__(self, features, citation, data_url, url, **kwargs):
        """BuilderConfig for Codequeries.

        Args:
          features: `list[string]`, list of the features that will appear in the
            feature dict. Should not include "label".
          citation: `string`, citation for the data set.
          **kwargs: keyword arguments forwarded to super.
        """
        # Version history:
        # 1.0.0: Initial version.
        super(CodequeriesConfig, self).__init__(version=datasets.Version("1.0.0"), **kwargs)
        self.features = features
        self.citation = citation
        self.data_url = data_url
        self.url = url


class Codequeries(datasets.GeneratorBasedBuilder):
    """The Codequeries benchmark."""

    BUILDER_CONFIGS = [
        CodequeriesConfig(
            name="ideal",
            description=_IDEAL_DESCRIPTION,
            features=["query_name", "context_blocks", "answer_spans",
                      "supporting_fact_spans", "code_file_path", "example_type",
                      "subtokenized_input_sequence", "label_sequence"],
            citation=_CODEQUERIES_CITATION,
            data_url={
                "train": "ideal_train.json",
                "dev": "ideal_val.json",
                "test": "ideal_test.json"
            },
            url="",
        ),
        CodequeriesConfig(
            name="prefix",
            description=_PREFIX_DESCRIPTION,
            features=["query_name", "answer_spans",
                      "supporting_fact_spans", "code_file_path", "example_type",
                      "subtokenized_input_sequence", "label_sequence"],
            citation=_CODEQUERIES_CITATION,
            data_url={
                "test": "prefix_test.json"
            },
            url="",
        ),
        CodequeriesConfig(
            name="file_ideal",
            description=_FILE_IDEAL_DESCRIPTION,
            features=["query_name", "context_blocks", "answer_spans",
                      "supporting_fact_spans", "code_file_path", "example_type",
                      "subtokenized_input_sequence", "label_sequence"],
            citation=_CODEQUERIES_CITATION,
            data_url={
                "test": "file_ideal_test.json"
            },
            url="",
        ),
        CodequeriesConfig(
            name="twostep",
            description=_TWOSTEP_DESCRIPTION,
            features=["query_name", "context_blocks", "answer_spans",
                      "supporting_fact_spans", "code_file_path", "example_type",
                      "subtokenized_input_sequence", "label_sequence"],
            citation=_CODEQUERIES_CITATION,
            data_url={
                "test": "twostep_relevance/twostep_relevance_test_"
            },
            url="",
        ),
    ]

    DEFAULT_CONFIG_NAME = "ideal"

    def _info(self):
        features = {}
        features["query_name"] = datasets.Value("string")
        features["context_blocks"] = [
            {
                "content": datasets.Value("string"),
                "metadata": datasets.Value("string"),
                "header": datasets.Value("string")
            }
        ]
        features["answer_spans"] = [
            {
                'span': datasets.Value("string"),
                'start_line': datasets.Value("int32"),
                'start_column': datasets.Value("int32"),
                'end_line': datasets.Value("int32"),
                'end_column': datasets.Value("int32")
            }
        ]
        features["supporting_fact_spans"] = [
            {
                'span': datasets.Value("string"),
                'start_line': datasets.Value("int32"),
                'start_column': datasets.Value("int32"),
                'end_line': datasets.Value("int32"),
                'end_column': datasets.Value("int32")
            }
        ]
        features["code_file_path"] = datasets.Value("string")
        features["example_type"] = datasets.Value("int32")
        features["subtokenized_input_sequence"] = datasets.features.Sequence(datasets.Value("string"))
        features["label_sequence"] = datasets.features.Sequence(datasets.Value("int32"))

        return datasets.DatasetInfo(
            description=self.config.description,
            features=datasets.Features(features),
            homepage=self.config.url,
            citation=_CODEQUERIES_CITATION,
        )

    def _split_generators(self, dl_manager):
        dl_dir = dl_manager.download_and_extract(self.config.data_url)
        if self.config.name in ["prefix", "file_ideal", "twostep"]:
            return [
                datasets.SplitGenerator(
                    name=datasets.Split.TEST,
                    gen_kwargs={
                        "filepath": os.path.join(dl_dir["test"]),
                        "split": datasets.Split.TEST,
                    },
                ),
            ]
        else:
            return [
                datasets.SplitGenerator(
                    name=datasets.Split.TRAIN,
                    gen_kwargs={
                        "filepath": os.path.join(dl_dir["train"]),
                        "split": datasets.Split.TRAIN,
                    },
                ),
                datasets.SplitGenerator(
                    name=datasets.Split.VALIDATION,
                    gen_kwargs={
                        "filepath": os.path.join(dl_dir["dev"]),
                        "split": datasets.Split.VALIDATION,
                    },
                ),
                datasets.SplitGenerator(
                    name=datasets.Split.TEST,
                    gen_kwargs={
                        "filepath": os.path.join(dl_dir["test"]),
                        "split": datasets.Split.TEST,
                    },
                ),
            ]

    def _generate_examples(self, filepath, split):
        if self.config.name in ["prefix", "file_ideal", "twostep"]:
            assert split == datasets.Split.TEST
        logger.info("generating examples from = %s", filepath)

        if self.config.name == "twostep":
            key = 0
            for i in range(10):
                with open(filepath + str(i) + '.json', encoding="utf-8") as f:
                    for line in f:
                        row = json.loads(line)

                        instance_key = str(key) + "_" + row["query_name"] + "_" + row["code_file_path"]
                        yield instance_key, {
                            "query_name": row["query_name"],
                            "context_blocks": row["context_blocks"],
                            "answer_spans": row["answer_spans"],
                            "supporting_fact_spans": row["supporting_fact_spans"],
                            "code_file_path": row["code_file_path"],
                            "example_type": row["example_type"],
                            "subtokenized_input_sequence": row["subtokenized_input_sequence"],
                            "label_sequence": row["label_sequence"],
                        }
                        key += 1
        else:
            with open(filepath, encoding="utf-8") as f:
                key = 0
                for line in f:
                    row = json.loads(line)

                    instance_key = str(key) + "_" + row["query_name"] + "_" + row["code_file_path"]
                    yield instance_key, {
                        "query_name": row["query_name"],
                        "context_blocks": row["context_blocks"],
                        "answer_spans": row["answer_spans"],
                        "supporting_fact_spans": row["supporting_fact_spans"],
                        "code_file_path": row["code_file_path"],
                        "example_type": row["example_type"],
                        "subtokenized_input_sequence": row["subtokenized_input_sequence"],
                        "label_sequence": row["label_sequence"],
                    }
                    key += 1