from __future__ import annotations

import importlib.util
from importlib import import_module
from typing import TYPE_CHECKING, Any

if TYPE_CHECKING:
    from . import v2 as v2
    from .auto_client import from_provider as from_provider
    from .batch import (
        BatchJob as BatchJob,
        BatchProcessor as BatchProcessor,
        BatchRequest as BatchRequest,
    )
    from .core import hooks as hooks
    from .core.client import (
        AsyncInstructor as AsyncInstructor,
        Instructor as Instructor,
    )
    from .core.patch import apatch as apatch, patch as patch
    from .distil import FinetuneFormat as FinetuneFormat, Instructions as Instructions
    from .mode import Mode as Mode
    from .utils.providers import Provider as Provider
    from .v2.core.function_calls import (
        OpenAISchema as OpenAISchema,
        ResponseSchema as ResponseSchema,
        openai_schema as openai_schema,
        response_schema as response_schema,
    )
    from .v2.core.multimodal import Audio as Audio, Image as Image
    from .v2.core.schema import (
        generate_anthropic_schema as generate_anthropic_schema,
        generate_gemini_schema as generate_gemini_schema,
        generate_openai_schema as generate_openai_schema,
    )
    from .v2.dsl import (
        CitationMixin as CitationMixin,
        IterableModel as IterableModel,
        Maybe as Maybe,
        Partial as Partial,
    )
    from .v2.providers.anthropic.client import from_anthropic as from_anthropic
    from .v2.providers.bedrock.client import from_bedrock as from_bedrock
    from .v2.providers.cerebras.client import from_cerebras as from_cerebras
    from .v2.providers.cohere.client import from_cohere as from_cohere
    from .v2.providers.fireworks.client import from_fireworks as from_fireworks
    from .v2.providers.gemini.client import from_gemini as from_gemini
    from .v2.providers.genai.client import from_genai as from_genai
    from .v2.providers.groq.client import from_groq as from_groq
    from .v2.providers.litellm.client import from_litellm as from_litellm
    from .v2.providers.mistral.client import from_mistral as from_mistral
    from .v2.providers.openai.client import (
        from_anyscale as from_anyscale,
        from_databricks as from_databricks,
        from_deepseek as from_deepseek,
        from_openai as from_openai,
        from_together as from_together,
    )
    from .v2.providers.openrouter.client import from_openrouter as from_openrouter
    from .v2.providers.perplexity.client import from_perplexity as from_perplexity
    from .v2.providers.vertexai.client import from_vertexai as from_vertexai
    from .v2.providers.writer.client import from_writer as from_writer
    from .v2.providers.xai.client import from_xai as from_xai
    from .v2.validation import (
        llm_validator as llm_validator,
        openai_moderation as openai_moderation,
    )

__version__ = "1.15.5"

__all__ = [
    "Instructor",
    "Image",
    "Audio",
    "from_openai",
    "from_anyscale",
    "from_together",
    "from_databricks",
    "from_deepseek",
    "from_openrouter",
    "from_litellm",
    "from_vertexai",
    "from_provider",
    "AsyncInstructor",
    "Provider",
    "ResponseSchema",
    "response_schema",
    "OpenAISchema",
    "CitationMixin",
    "IterableModel",
    "Maybe",
    "Partial",
    "openai_schema",
    "generate_openai_schema",
    "generate_anthropic_schema",
    "generate_gemini_schema",
    "Mode",
    "patch",
    "apatch",
    "FinetuneFormat",
    "Instructions",
    "BatchProcessor",
    "BatchRequest",
    "BatchJob",
    "llm_validator",
    "openai_moderation",
    "hooks",
    "v2",
]

_LAZY_IMPORTS: dict[str, tuple[str, str | None]] = {
    "Instructor": (".core.client", "Instructor"),
    "AsyncInstructor": (".core.client", "AsyncInstructor"),
    "from_openai": (".v2.providers.openai.client", "from_openai"),
    "from_anyscale": (".v2.providers.openai.client", "from_anyscale"),
    "from_together": (".v2.providers.openai.client", "from_together"),
    "from_databricks": (".v2.providers.openai.client", "from_databricks"),
    "from_deepseek": (".v2.providers.openai.client", "from_deepseek"),
    "from_openrouter": (".v2.providers.openrouter.client", "from_openrouter"),
    "from_litellm": (".v2.providers.litellm.client", "from_litellm"),
    "Mode": (".mode", "Mode"),
    "patch": (".core.patch", "patch"),
    "apatch": (".core.patch", "apatch"),
    "hooks": (".core.hooks", None),
    "v2": (".v2", None),
    "Image": (".v2.core.multimodal", "Image"),
    "Audio": (".v2.core.multimodal", "Audio"),
    "CitationMixin": (".v2.dsl", "CitationMixin"),
    "IterableModel": (".v2.dsl", "IterableModel"),
    "Maybe": (".v2.dsl", "Maybe"),
    "Partial": (".v2.dsl", "Partial"),
    "ResponseSchema": (".v2.core.function_calls", "ResponseSchema"),
    "response_schema": (".v2.core.function_calls", "response_schema"),
    "OpenAISchema": (".v2.core.function_calls", "OpenAISchema"),
    "openai_schema": (".v2.core.function_calls", "openai_schema"),
    "generate_openai_schema": (".v2.core.schema", "generate_openai_schema"),
    "generate_anthropic_schema": (".v2.core.schema", "generate_anthropic_schema"),
    "generate_gemini_schema": (".v2.core.schema", "generate_gemini_schema"),
    "llm_validator": (".v2.validation", "llm_validator"),
    "openai_moderation": (".v2.validation", "openai_moderation"),
    "Provider": (".utils.providers", "Provider"),
    "from_provider": (".auto_client", "from_provider"),
    "BatchProcessor": (".batch", "BatchProcessor"),
    "BatchRequest": (".batch", "BatchRequest"),
    "BatchJob": (".batch", "BatchJob"),
    "FinetuneFormat": (".distil", "FinetuneFormat"),
    "Instructions": (".distil", "Instructions"),
    "from_anthropic": (".v2.providers.anthropic.client", "from_anthropic"),
    "from_gemini": (".v2.providers.gemini.client", "from_gemini"),
    "from_fireworks": (".v2.providers.fireworks.client", "from_fireworks"),
    "from_cerebras": (".v2.providers.cerebras.client", "from_cerebras"),
    "from_groq": (".v2.providers.groq.client", "from_groq"),
    "from_mistral": (".v2.providers.mistral.client", "from_mistral"),
    "from_cohere": (".v2.providers.cohere.client", "from_cohere"),
    "from_vertexai": (".v2.providers.vertexai.client", "from_vertexai"),
    "from_bedrock": (".v2.providers.bedrock.client", "from_bedrock"),
    "from_writer": (".v2.providers.writer.client", "from_writer"),
    "from_xai": (".v2.providers.xai.client", "from_xai"),
    "from_perplexity": (".v2.providers.perplexity.client", "from_perplexity"),
    "from_genai": (".v2.providers.genai.client", "from_genai"),
}


def __getattr__(name: str) -> Any:
    if name not in __all__:
        raise AttributeError(f"module {__name__!r} has no attribute {name!r}")

    module_path, attr_name = _LAZY_IMPORTS[name]
    module = import_module(module_path, package=__name__)
    value = module if attr_name is None else getattr(module, attr_name)
    globals()[name] = value
    return value


def _add_optional_export(name: str, *packages: str) -> None:
    if all(importlib.util.find_spec(package) is not None for package in packages):
        __all__.append(name)


_add_optional_export("from_anthropic", "anthropic")
_add_optional_export("from_gemini", "google", "google.generativeai")
_add_optional_export("from_fireworks", "fireworks")
_add_optional_export("from_cerebras", "cerebras")
_add_optional_export("from_groq", "groq")
_add_optional_export("from_mistral", "mistralai")
_add_optional_export("from_cohere", "cohere")
_add_optional_export("from_bedrock", "boto3")
_add_optional_export("from_writer", "writerai")
_add_optional_export("from_xai", "xai_sdk")
_add_optional_export("from_perplexity", "openai")
_add_optional_export("from_genai", "google", "google.genai")