from pathlib import Path, PurePath
from typing import Annotated, Literal, Optional, Union

from pydantic import AnyUrl, BaseModel, Field, SecretStr


class BaseBackendOptions(BaseModel):
    """Common options for all declarative document backends."""

    enable_remote_fetch: bool = Field(
        False, description="Enable remote resource fetching."
    )
    enable_local_fetch: bool = Field(
        False, description="Enable local resource fetching."
    )


class DeclarativeBackendOptions(BaseBackendOptions):
    """Default backend options for a declarative document backend."""

    kind: Literal["declarative"] = Field("declarative", exclude=True, repr=False)


class HTMLBackendOptions(BaseBackendOptions):
    """Options specific to the HTML backend.

    This class can be extended to include options specific to HTML processing.
    """

    kind: Literal["html"] = Field("html", exclude=True, repr=False)
    fetch_images: bool = Field(
        False,
        description=(
            "Whether the backend should access remote or local resources to parse "
            "images in an HTML document."
        ),
    )
    source_uri: Optional[Union[AnyUrl, PurePath]] = Field(
        None,
        description=(
            "The URI that originates the HTML document. If provided, the backend "
            "will use it to resolve relative paths in the HTML document."
        ),
    )
    add_title: bool = Field(
        True, description="Add the HTML title tag as furniture in the DoclingDocument."
    )
    infer_furniture: bool = Field(
        True, description="Infer all the content before the first header as furniture."
    )


class MarkdownBackendOptions(BaseBackendOptions):
    """Options specific to the Markdown backend."""

    kind: Literal["md"] = Field("md", exclude=True, repr=False)
    fetch_images: bool = Field(
        False,
        description=(
            "Whether the backend should access remote or local resources to parse "
            "images in the markdown document."
        ),
    )
    source_uri: Optional[Union[AnyUrl, PurePath]] = Field(
        None,
        description=(
            "The URI that originates the markdown document. If provided, the backend "
            "will use it to resolve relative paths in the markdown document."
        ),
    )


class PdfBackendOptions(BaseBackendOptions):
    """Backend options for pdf document backends."""

    kind: Literal["pdf"] = Field("pdf", exclude=True, repr=False)
    password: Optional[SecretStr] = None


class MsExcelBackendOptions(BaseBackendOptions):
    """Options specific to the MS Excel backend."""

    kind: Literal["xlsx"] = Field("xlsx", exclude=True, repr=False)
    treat_singleton_as_text: bool = Field(
        False,
        description=(
            "Whether to treat singleton cells (1x1 tables with empty neighboring "
            "cells) as TextItem instead of TableItem."
        ),
    )
    gap_tolerance: int = Field(
        0,
        description=(
            "The tolerance (in number of empty rows/columns) for merging nearby "
            "data clusters into a single table. Default is 0 (strict)."
        ),
    )


class LatexBackendOptions(BaseBackendOptions):
    """Options specific to the LaTeX backend."""

    kind: Literal["latex"] = Field("latex", exclude=True, repr=False)
    parse_timeout: Optional[float] = Field(
        30.0,
        description=(
            "Maximum time allowed for parsing a LaTeX document. "
            "Set to None to disable the timeout. Defaults to 30 s."
        ),
    )


class XBRLBackendOptions(BaseBackendOptions):
    """Options specific to the XBRL backend."""

    kind: Annotated[Literal["xbrl"], Field("xbrl", exclude=True, repr=False)] = "xbrl"
    taxonomy: Annotated[
        Path | None,
        Field(
            description=(
                "Path to a folder with the taxonomy required by the XBRL instance"
                " reports. It should include schemas (`.xsd`) and linkbases (`.xml`)"
                " referenced by the XBRL reports in their relative locations."
                " Optionally, it can also include taxonomy packages (`.zip`)"
                " referenced by the reports with absolute URLs and mapped to files"
                " with a taxonomy catalog (`catalog.xml`) for offline parsing."
            )
        ),
    ] = None


BackendOptions = Annotated[
    Union[
        DeclarativeBackendOptions,
        HTMLBackendOptions,
        MarkdownBackendOptions,
        PdfBackendOptions,
        MsExcelBackendOptions,
        LatexBackendOptions,
        XBRLBackendOptions,
    ],
    Field(discriminator="kind"),
]