python/bin/data.py

from enum import Enum
from dataclasses import dataclass, asdict, fields
import dataclasses
from itertools import chain
import json
from pathlib import Path
import toml
from typing import List, Any, Dict, Type


def _custom_dataclass_init(self, *args, **kwargs):
    # print(self.__class__.__name__, "__init__")
    names = [field.name for field in fields(self)]
    used_names = set()

    # Handle positional arguments
    for value in args:
        try:
            name = names.pop(0)
        except IndexError:
            raise TypeError(f"__init__() given too many positional arguments")
        # print(f'setting {k}={v}')
        setattr(self, name, value)
        used_names.add(name)

    # Handle keyword arguments
    for name, value in kwargs.items():
        if name in names:
            # print(f'setting {k}={v}')
            setattr(self, name, value)
            used_names.add(name)
        elif name in used_names:
            raise TypeError(f"__init__() got multiple values for argument '{name}'")
        else:
            raise TypeError(
                f"Unrecognized field '{name}' for dataclass {self.__class__.__name__}."
                "\nIf this field is valid, please add it to the dataclass in data.py."
                "\nIf adding an object-type field, please create a new dataclass for it."
            )

    # Check for missing positional arguments
    missing = [
        f"'{field.name}'" for field in fields(self)
        if isinstance(field.default, dataclasses._MISSING_TYPE) and field.name not in used_names
    ]
    if len(missing) == 1:
        raise TypeError(f"__init__() missing 1 required positional argument: {missing[0]}")
    elif len(missing) == 2:
        raise TypeError(f"__init__() missing 2 required positional arguments: {' and '.join(missing)}")
    elif len(missing) != 0:
        missing[-1] = f"and {missing[-1]}"
        raise TypeError(f"__init__() missing {len(missing)} required positional arguments: {', '.join(missing)}")

    # Run post init if available
    if hasattr(self, "__post_init__"):
        self.__post_init__()


@dataclass
class TrackStatus:
    __init__ = _custom_dataclass_init

    concept_exercises: bool = False
    test_runner: bool = False
    representer: bool = False
    analyzer: bool = False


class IndentStyle(str, Enum):
    Space = "space"
    Tab = "tab"


@dataclass
class TestRunnerSettings:
    average_run_time: float = -1


@dataclass
class EditorSettings:
    __init__ = _custom_dataclass_init

    indent_style: IndentStyle = IndentStyle.Space
    indent_size: int = 4
    ace_editor_language: str = "python"
    highlightjs_language: str = "python"

    def __post_init__(self):
        if isinstance(self.indent_style, str):
            self.indent_style = IndentStyle(self.indent_style)


class ExerciseStatus(str, Enum):
    Active = "active"
    WIP = "wip"
    Beta = "beta"
    Deprecated = "deprecated"


@dataclass
class ExerciseFiles:
    __init__ = _custom_dataclass_init

    solution: List[str]
    test: List[str]
    editor: List[str] = None
    exemplar: List[str] = None


    # practice exercises are different
    example: List[str] = None

    def __post_init__(self):
        if self.exemplar is None:
            if self.example is None:
                raise ValueError(
                    "exercise config must have either files.exemplar or files.example"
                )
            else:
                self.exemplar = self.example
                delattr(self, "example")
        elif self.example is not None:
            raise ValueError(
                "exercise config must have either files.exemplar or files.example, but not both"
            )


@dataclass
class ExerciseConfig:
    __init__ = _custom_dataclass_init

    files: ExerciseFiles
    authors: List[str] = None
    forked_from: str = None
    contributors: List[str] = None
    language_versions: List[str] = None
    test_runner: bool = True
    source: str = None
    source_url: str = None
    blurb: str = None
    icon: str = None

    def __post_init__(self):
        if isinstance(self.files, dict):
            self.files = ExerciseFiles(**self.files)
        for attr in ["authors", "contributors", "language_versions"]:
            if getattr(self, attr) is None:
                setattr(self, attr, [])

    @classmethod
    def load(cls, config_file: Path) -> "ExerciseConfig":
        with config_file.open() as f:
            return cls(**json.load(f))


@dataclass
class ExerciseInfo:
    __init__ = _custom_dataclass_init

    path: Path
    slug: str
    name: str
    uuid: str
    prerequisites: List[str]
    type: str = "practice"
    status: ExerciseStatus = ExerciseStatus.Active

    # concept only
    concepts: List[str] = None

    # practice only
    difficulty: int = 1
    topics: List[str] = None
    practices: List[str] = None

    def __post_init__(self):
        if self.concepts is None:
            self.concepts = []
        if self.topics is None:
            self.topics = []
        if self.practices is None:
            self.practices = []
        if isinstance(self.status, str):
            self.status = ExerciseStatus(self.status)

    @property
    def solution_stub(self):
        return next(
            (
                p
                for p in self.path.glob("*.py")
                if not p.name.endswith("_test.py") and p.name != "example.py"
            ),
            None,
        )

    @property
    def helper_file(self):
        return next(self.path.glob("*_data.py"), None)

    @property
    def test_file(self):
        return next(self.path.glob("*_test.py"), None)

    @property
    def meta_dir(self):
        return self.path / ".meta"

    @property
    def exemplar_file(self):
        if self.type == "concept":
            return self.meta_dir / "exemplar.py"
        return self.meta_dir / "example.py"

    @property
    def template_path(self):
        return self.meta_dir / "template.j2"

    @property
    def config_file(self):
        return self.meta_dir / "config.json"

    def load_config(self) -> ExerciseConfig:
        return ExerciseConfig.load(self.config_file)


@dataclass
class Exercises:
    __init__ = _custom_dataclass_init

    concept: List[ExerciseInfo]
    practice: List[ExerciseInfo]
    foregone: List[str] = None

    def __post_init__(self):
        if self.foregone is None:
            self.foregone = []
        for attr_name in ["concept", "practice"]:
            base_path = Path("exercises") / attr_name
            setattr(
                self,
                attr_name,
                [
                    (
                        ExerciseInfo(path=(base_path / e["slug"]), type=attr_name, **e)
                        if isinstance(e, dict)
                        else e
                    )
                    for e in getattr(self, attr_name)
                ],
            )

    def all(self, status_filter={ExerciseStatus.Active, ExerciseStatus.Beta}):
        return [
            e for e in chain(self.concept, self.practice) if e.status in status_filter
        ]


@dataclass
class Concept:
    __init__ = _custom_dataclass_init

    uuid: str
    slug: str
    name: str


@dataclass
class Feature:
    __init__ = _custom_dataclass_init

    title: str
    content: str
    icon: str


@dataclass
class FilePatterns:
    __init__ = _custom_dataclass_init

    solution: List[str]
    test: List[str]
    example: List[str]
    exemplar: List[str]
    editor: List[str] = None



@dataclass
class Config:
    __init__ = _custom_dataclass_init

    language: str
    slug: str
    active: bool
    status: TrackStatus
    blurb: str
    version: int
    online_editor: EditorSettings
    exercises: Exercises
    concepts: List[Concept]
    key_features: List[Feature] = None
    tags: List[Any] = None
    test_runner: TestRunnerSettings = None
    files: FilePatterns = None

    def __post_init__(self):
        if isinstance(self.status, dict):
            self.status = TrackStatus(**self.status)
        if isinstance(self.online_editor, dict):
            self.online_editor = EditorSettings(**self.online_editor)
        if isinstance(self.test_runner, dict):
            self.test_runner = TestRunnerSettings(**self.test_runner)
        if isinstance(self.exercises, dict):
            self.exercises = Exercises(**self.exercises)
        if isinstance(self.files, dict):
            self.files = FilePatterns(**self.files)
        self.concepts = [
            (Concept(**c) if isinstance(c, dict) else c) for c in self.concepts
        ]
        if self.key_features is None:
            self.key_features = []
        if self.tags is None:
            self.tags = []

    @classmethod
    def load(cls, path="config.json"):
        try:
            with Path(path).open() as f:
                return cls(**json.load(f))
        except IOError:
            print(f"FAIL: {path} file not found")
            raise SystemExit(1)
        except TypeError as ex:
            print(f"FAIL: {ex}")
            raise SystemExit(1)


@dataclass
class TestCaseTOML:
    __init__ = _custom_dataclass_init

    uuid: str
    description: str
    include: bool = True
    comment: str = ''


@dataclass
class TestsTOML:
    __init__ = _custom_dataclass_init

    cases: Dict[str, TestCaseTOML]

    @classmethod
    def load(cls, toml_path: Path):
        with toml_path.open() as f:
            data = toml.load(f)
        return cls({uuid: TestCaseTOML(uuid, *opts) for uuid, opts in data.items()})


if __name__ == "__main__":

    class CustomEncoder(json.JSONEncoder):
        def default(self, obj):
            if isinstance(obj, Path):
                return str(obj)
            return json.JSONEncoder.default(self, obj)

    config = Config.load()
    print(json.dumps(asdict(config), cls=CustomEncoder, indent=2))