antfly-sdk 0.0.2.dev2__py3-none-any.whl

antfly/client_generated/models/extraction_schema_structures.py

@@ -0,0 +1,60 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+from typing import TYPE_CHECKING, Any, TypeVar
+
+from attrs import define as _attrs_define
+from attrs import field as _attrs_field
+
+if TYPE_CHECKING:
+    from ..models.extraction_structure_schema import ExtractionStructureSchema
+
+
+T = TypeVar("T", bound="ExtractionSchemaStructures")
+
+
+@_attrs_define
+class ExtractionSchemaStructures:
+    """ """
+
+    additional_properties: dict[str, ExtractionStructureSchema] = _attrs_field(init=False, factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+
+        field_dict: dict[str, Any] = {}
+        for prop_name, prop in self.additional_properties.items():
+            field_dict[prop_name] = prop.to_dict()
+
+        return field_dict
+
+    @classmethod
+    def from_dict(cls: type[T], src_dict: Mapping[str, Any]) -> T:
+        from ..models.extraction_structure_schema import ExtractionStructureSchema
+
+        d = dict(src_dict)
+        extraction_schema_structures = cls()
+
+        additional_properties = {}
+        for prop_name, prop_dict in d.items():
+            additional_property = ExtractionStructureSchema.from_dict(prop_dict)
+
+            additional_properties[prop_name] = additional_property
+
+        extraction_schema_structures.additional_properties = additional_properties
+        return extraction_schema_structures
+
+    @property
+    def additional_keys(self) -> list[str]:
+        return list(self.additional_properties.keys())
+
+    def __getitem__(self, key: str) -> ExtractionStructureSchema:
+        return self.additional_properties[key]
+
+    def __setitem__(self, key: str, value: ExtractionStructureSchema) -> None:
+        self.additional_properties[key] = value
+
+    def __delitem__(self, key: str) -> None:
+        del self.additional_properties[key]
+
+    def __contains__(self, key: str) -> bool:
+        return key in self.additional_properties

antfly/client_generated/models/extraction_structure_field_type_1.py

@@ -0,0 +1,47 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+from typing import Any, TypeVar
+
+from attrs import define as _attrs_define
+from attrs import field as _attrs_field
+
+T = TypeVar("T", bound="ExtractionStructureFieldType1")
+
+
+@_attrs_define
+class ExtractionStructureFieldType1:
+    """ """
+
+    additional_properties: dict[str, Any] = _attrs_field(init=False, factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+
+        field_dict: dict[str, Any] = {}
+        field_dict.update(self.additional_properties)
+
+        return field_dict
+
+    @classmethod
+    def from_dict(cls: type[T], src_dict: Mapping[str, Any]) -> T:
+        d = dict(src_dict)
+        extraction_structure_field_type_1 = cls()
+
+        extraction_structure_field_type_1.additional_properties = d
+        return extraction_structure_field_type_1
+
+    @property
+    def additional_keys(self) -> list[str]:
+        return list(self.additional_properties.keys())
+
+    def __getitem__(self, key: str) -> Any:
+        return self.additional_properties[key]
+
+    def __setitem__(self, key: str, value: Any) -> None:
+        self.additional_properties[key] = value
+
+    def __delitem__(self, key: str) -> None:
+        del self.additional_properties[key]
+
+    def __contains__(self, key: str) -> bool:
+        return key in self.additional_properties

antfly/client_generated/models/extraction_structure_schema.py

@@ -0,0 +1,74 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+from typing import TYPE_CHECKING, Any, TypeVar
+
+from attrs import define as _attrs_define
+from attrs import field as _attrs_field
+
+from ..types import UNSET, Unset
+
+if TYPE_CHECKING:
+    from ..models.extraction_structure_schema_fields import ExtractionStructureSchemaFields
+
+
+T = TypeVar("T", bound="ExtractionStructureSchema")
+
+
+@_attrs_define
+class ExtractionStructureSchema:
+    """
+    Attributes:
+        fields (ExtractionStructureSchemaFields | Unset):
+    """
+
+    fields: ExtractionStructureSchemaFields | Unset = UNSET
+    additional_properties: dict[str, Any] = _attrs_field(init=False, factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+        fields: dict[str, Any] | Unset = UNSET
+        if not isinstance(self.fields, Unset):
+            fields = self.fields.to_dict()
+
+        field_dict: dict[str, Any] = {}
+        field_dict.update(self.additional_properties)
+        field_dict.update({})
+        if fields is not UNSET:
+            field_dict["fields"] = fields
+
+        return field_dict
+
+    @classmethod
+    def from_dict(cls: type[T], src_dict: Mapping[str, Any]) -> T:
+        from ..models.extraction_structure_schema_fields import ExtractionStructureSchemaFields
+
+        d = dict(src_dict)
+        _fields = d.pop("fields", UNSET)
+        fields: ExtractionStructureSchemaFields | Unset
+        if isinstance(_fields, Unset):
+            fields = UNSET
+        else:
+            fields = ExtractionStructureSchemaFields.from_dict(_fields)
+
+        extraction_structure_schema = cls(
+            fields=fields,
+        )
+
+        extraction_structure_schema.additional_properties = d
+        return extraction_structure_schema
+
+    @property
+    def additional_keys(self) -> list[str]:
+        return list(self.additional_properties.keys())
+
+    def __getitem__(self, key: str) -> Any:
+        return self.additional_properties[key]
+
+    def __setitem__(self, key: str, value: Any) -> None:
+        self.additional_properties[key] = value
+
+    def __delitem__(self, key: str) -> None:
+        del self.additional_properties[key]
+
+    def __contains__(self, key: str) -> bool:
+        return key in self.additional_properties

antfly/client_generated/models/extraction_structure_schema_fields.py

@@ -0,0 +1,76 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+from typing import TYPE_CHECKING, Any, TypeVar, cast
+
+from attrs import define as _attrs_define
+from attrs import field as _attrs_field
+
+if TYPE_CHECKING:
+    from ..models.extraction_structure_field_type_1 import ExtractionStructureFieldType1
+
+
+T = TypeVar("T", bound="ExtractionStructureSchemaFields")
+
+
+@_attrs_define
+class ExtractionStructureSchemaFields:
+    """ """
+
+    additional_properties: dict[str, ExtractionStructureFieldType1 | str] = _attrs_field(init=False, factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+        from ..models.extraction_structure_field_type_1 import ExtractionStructureFieldType1
+
+        field_dict: dict[str, Any] = {}
+        for prop_name, prop in self.additional_properties.items():
+            if isinstance(prop, ExtractionStructureFieldType1):
+                field_dict[prop_name] = prop.to_dict()
+            else:
+                field_dict[prop_name] = prop
+
+        return field_dict
+
+    @classmethod
+    def from_dict(cls: type[T], src_dict: Mapping[str, Any]) -> T:
+        from ..models.extraction_structure_field_type_1 import ExtractionStructureFieldType1
+
+        d = dict(src_dict)
+        extraction_structure_schema_fields = cls()
+
+        additional_properties = {}
+        for prop_name, prop_dict in d.items():
+
+            def _parse_additional_property(data: object) -> ExtractionStructureFieldType1 | str:
+                try:
+                    if not isinstance(data, dict):
+                        raise TypeError()
+                    componentsschemas_extraction_structure_field_type_1 = ExtractionStructureFieldType1.from_dict(data)
+
+                    return componentsschemas_extraction_structure_field_type_1
+                except (TypeError, ValueError, AttributeError, KeyError):
+                    pass
+                return cast(ExtractionStructureFieldType1 | str, data)
+
+            additional_property = _parse_additional_property(prop_dict)
+
+            additional_properties[prop_name] = additional_property
+
+        extraction_structure_schema_fields.additional_properties = additional_properties
+        return extraction_structure_schema_fields
+
+    @property
+    def additional_keys(self) -> list[str]:
+        return list(self.additional_properties.keys())
+
+    def __getitem__(self, key: str) -> ExtractionStructureFieldType1 | str:
+        return self.additional_properties[key]
+
+    def __setitem__(self, key: str, value: ExtractionStructureFieldType1 | str) -> None:
+        self.additional_properties[key] = value
+
+    def __delitem__(self, key: str) -> None:
+        del self.additional_properties[key]
+
+    def __contains__(self, key: str) -> bool:
+        return key in self.additional_properties

antfly/client_generated/models/extraction_token.py

@@ -0,0 +1,74 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+from typing import Any, TypeVar, cast
+
+from attrs import define as _attrs_define
+from attrs import field as _attrs_field
+
+from ..types import UNSET, Unset
+
+T = TypeVar("T", bound="ExtractionToken")
+
+
+@_attrs_define
+class ExtractionToken:
+    """
+    Attributes:
+        text (str):
+        box (list[int] | Unset):
+    """
+
+    text: str
+    box: list[int] | Unset = UNSET
+    additional_properties: dict[str, Any] = _attrs_field(init=False, factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+        text = self.text
+
+        box: list[int] | Unset = UNSET
+        if not isinstance(self.box, Unset):
+            box = self.box
+
+        field_dict: dict[str, Any] = {}
+        field_dict.update(self.additional_properties)
+        field_dict.update(
+            {
+                "text": text,
+            }
+        )
+        if box is not UNSET:
+            field_dict["box"] = box
+
+        return field_dict
+
+    @classmethod
+    def from_dict(cls: type[T], src_dict: Mapping[str, Any]) -> T:
+        d = dict(src_dict)
+        text = d.pop("text")
+
+        box = cast(list[int], d.pop("box", UNSET))
+
+        extraction_token = cls(
+            text=text,
+            box=box,
+        )
+
+        extraction_token.additional_properties = d
+        return extraction_token
+
+    @property
+    def additional_keys(self) -> list[str]:
+        return list(self.additional_properties.keys())
+
+    def __getitem__(self, key: str) -> Any:
+        return self.additional_properties[key]
+
+    def __setitem__(self, key: str, value: Any) -> None:
+        self.additional_properties[key] = value
+
+    def __delitem__(self, key: str) -> None:
+        del self.additional_properties[key]
+
+    def __contains__(self, key: str) -> bool:
+        return key in self.additional_properties

antfly/client_generated/models/failed_operation.py

@@ -0,0 +1,87 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+from typing import Any, TypeVar
+
+from attrs import define as _attrs_define
+from attrs import field as _attrs_field
+
+from ..models.failed_operation_operation import FailedOperationOperation
+from ..types import UNSET, Unset
+
+T = TypeVar("T", bound="FailedOperation")
+
+
+@_attrs_define
+class FailedOperation:
+    """
+    Attributes:
+        id (str | Unset):
+        operation (FailedOperationOperation | Unset):
+        error (str | Unset):
+    """
+
+    id: str | Unset = UNSET
+    operation: FailedOperationOperation | Unset = UNSET
+    error: str | Unset = UNSET
+    additional_properties: dict[str, Any] = _attrs_field(init=False, factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+        id = self.id
+
+        operation: str | Unset = UNSET
+        if not isinstance(self.operation, Unset):
+            operation = self.operation.value
+
+        error = self.error
+
+        field_dict: dict[str, Any] = {}
+        field_dict.update(self.additional_properties)
+        field_dict.update({})
+        if id is not UNSET:
+            field_dict["id"] = id
+        if operation is not UNSET:
+            field_dict["operation"] = operation
+        if error is not UNSET:
+            field_dict["error"] = error
+
+        return field_dict
+
+    @classmethod
+    def from_dict(cls: type[T], src_dict: Mapping[str, Any]) -> T:
+        d = dict(src_dict)
+        id = d.pop("id", UNSET)
+
+        _operation = d.pop("operation", UNSET)
+        operation: FailedOperationOperation | Unset
+        if isinstance(_operation, Unset):
+            operation = UNSET
+        else:
+            operation = FailedOperationOperation(_operation)
+
+        error = d.pop("error", UNSET)
+
+        failed_operation = cls(
+            id=id,
+            operation=operation,
+            error=error,
+        )
+
+        failed_operation.additional_properties = d
+        return failed_operation
+
+    @property
+    def additional_keys(self) -> list[str]:
+        return list(self.additional_properties.keys())
+
+    def __getitem__(self, key: str) -> Any:
+        return self.additional_properties[key]
+
+    def __setitem__(self, key: str, value: Any) -> None:
+        self.additional_properties[key] = value
+
+    def __delitem__(self, key: str) -> None:
+        del self.additional_properties[key]
+
+    def __contains__(self, key: str) -> bool:
+        return key in self.additional_properties

antfly/client_generated/models/failed_operation_operation.py

@@ -0,0 +1,9 @@
+from enum import Enum
+
+
+class FailedOperationOperation(str, Enum):
+    DELETE = "delete"
+    UPSERT = "upsert"
+
+    def __str__(self) -> str:
+        return str(self.value)

antfly/client_generated/models/fetch_config.py

@@ -0,0 +1,140 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+from typing import TYPE_CHECKING, Any, TypeVar, cast
+
+from attrs import define as _attrs_define
+from attrs import field as _attrs_field
+
+from ..types import UNSET, Unset
+
+if TYPE_CHECKING:
+    from ..models.credentials import Credentials
+
+
+T = TypeVar("T", bound="FetchConfig")
+
+
+@_attrs_define
+class FetchConfig:
+    """Configuration for URL content fetching.
+
+    Uses go/pkg/antfly/lib/scraping for downloading and processing. Supports:
+    - HTTP/HTTPS URLs with security validation
+    - HTML pages (extracts readable text via go-readability)
+    - PDF files (extracts text)
+    - Images (returns as data URIs)
+    - Plain text files
+    - S3 URLs (requires s3_credentials)
+
+    Security features (from go/pkg/antfly/lib/scraping.ContentSecurityConfig):
+    - Allowed host whitelist
+    - Private IP blocking (SSRF prevention)
+    - Download size limits
+    - Timeout controls
+
+        Attributes:
+            s3_credentials (Credentials | Unset):
+            max_content_length (int | Unset): Maximum content length in characters (truncated if exceeded) Default: 50000.
+            allowed_hosts (list[str] | Unset): Whitelist of allowed hostnames for fetching.
+                If empty, all hosts are allowed (except private IPs).
+                Example: ["docs.example.com", "api.example.com"]
+            block_private_ips (bool | Unset): Block requests to private IP ranges (SSRF prevention).
+                Blocked: 127.0.0.0/8, 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16
+                 Default: True.
+            max_download_size_bytes (int | Unset): Maximum download size in bytes (default: 100MB) Default: 104857600.
+            timeout_seconds (int | Unset): Download timeout in seconds Default: 30.
+    """
+
+    s3_credentials: Credentials | Unset = UNSET
+    max_content_length: int | Unset = 50000
+    allowed_hosts: list[str] | Unset = UNSET
+    block_private_ips: bool | Unset = True
+    max_download_size_bytes: int | Unset = 104857600
+    timeout_seconds: int | Unset = 30
+    additional_properties: dict[str, Any] = _attrs_field(init=False, factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+        s3_credentials: dict[str, Any] | Unset = UNSET
+        if not isinstance(self.s3_credentials, Unset):
+            s3_credentials = self.s3_credentials.to_dict()
+
+        max_content_length = self.max_content_length
+
+        allowed_hosts: list[str] | Unset = UNSET
+        if not isinstance(self.allowed_hosts, Unset):
+            allowed_hosts = self.allowed_hosts
+
+        block_private_ips = self.block_private_ips
+
+        max_download_size_bytes = self.max_download_size_bytes
+
+        timeout_seconds = self.timeout_seconds
+
+        field_dict: dict[str, Any] = {}
+        field_dict.update(self.additional_properties)
+        field_dict.update({})
+        if s3_credentials is not UNSET:
+            field_dict["s3_credentials"] = s3_credentials
+        if max_content_length is not UNSET:
+            field_dict["max_content_length"] = max_content_length
+        if allowed_hosts is not UNSET:
+            field_dict["allowed_hosts"] = allowed_hosts
+        if block_private_ips is not UNSET:
+            field_dict["block_private_ips"] = block_private_ips
+        if max_download_size_bytes is not UNSET:
+            field_dict["max_download_size_bytes"] = max_download_size_bytes
+        if timeout_seconds is not UNSET:
+            field_dict["timeout_seconds"] = timeout_seconds
+
+        return field_dict
+
+    @classmethod
+    def from_dict(cls: type[T], src_dict: Mapping[str, Any]) -> T:
+        from ..models.credentials import Credentials
+
+        d = dict(src_dict)
+        _s3_credentials = d.pop("s3_credentials", UNSET)
+        s3_credentials: Credentials | Unset
+        if isinstance(_s3_credentials, Unset):
+            s3_credentials = UNSET
+        else:
+            s3_credentials = Credentials.from_dict(_s3_credentials)
+
+        max_content_length = d.pop("max_content_length", UNSET)
+
+        allowed_hosts = cast(list[str], d.pop("allowed_hosts", UNSET))
+
+        block_private_ips = d.pop("block_private_ips", UNSET)
+
+        max_download_size_bytes = d.pop("max_download_size_bytes", UNSET)
+
+        timeout_seconds = d.pop("timeout_seconds", UNSET)
+
+        fetch_config = cls(
+            s3_credentials=s3_credentials,
+            max_content_length=max_content_length,
+            allowed_hosts=allowed_hosts,
+            block_private_ips=block_private_ips,
+            max_download_size_bytes=max_download_size_bytes,
+            timeout_seconds=timeout_seconds,
+        )
+
+        fetch_config.additional_properties = d
+        return fetch_config
+
+    @property
+    def additional_keys(self) -> list[str]:
+        return list(self.additional_properties.keys())
+
+    def __getitem__(self, key: str) -> Any:
+        return self.additional_properties[key]
+
+    def __setitem__(self, key: str, value: Any) -> None:
+        self.additional_properties[key] = value
+
+    def __delitem__(self, key: str) -> None:
+        del self.additional_properties[key]
+
+    def __contains__(self, key: str) -> bool:
+        return key in self.additional_properties

antfly/client_generated/models/field_statistics.py

@@ -0,0 +1,98 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+from typing import Any, TypeVar
+
+from attrs import define as _attrs_define
+from attrs import field as _attrs_field
+
+from ..types import UNSET, Unset
+
+T = TypeVar("T", bound="FieldStatistics")
+
+
+@_attrs_define
+class FieldStatistics:
+    """Statistics about a specific field.
+
+    Attributes:
+        cardinality (int | Unset): Approximate number of unique values (via HyperLogLog).
+        null_count (int | Unset): Number of rows with null values for this field.
+        min_value (Any | Unset): Minimum value for numeric/date fields.
+        max_value (Any | Unset): Maximum value for numeric/date fields.
+        avg_size (int | Unset): Average size in bytes for variable-length fields.
+    """
+
+    cardinality: int | Unset = UNSET
+    null_count: int | Unset = UNSET
+    min_value: Any | Unset = UNSET
+    max_value: Any | Unset = UNSET
+    avg_size: int | Unset = UNSET
+    additional_properties: dict[str, Any] = _attrs_field(init=False, factory=dict)
+
+    def to_dict(self) -> dict[str, Any]:
+        cardinality = self.cardinality
+
+        null_count = self.null_count
+
+        min_value = self.min_value
+
+        max_value = self.max_value
+
+        avg_size = self.avg_size
+
+        field_dict: dict[str, Any] = {}
+        field_dict.update(self.additional_properties)
+        field_dict.update({})
+        if cardinality is not UNSET:
+            field_dict["cardinality"] = cardinality
+        if null_count is not UNSET:
+            field_dict["null_count"] = null_count
+        if min_value is not UNSET:
+            field_dict["min_value"] = min_value
+        if max_value is not UNSET:
+            field_dict["max_value"] = max_value
+        if avg_size is not UNSET:
+            field_dict["avg_size"] = avg_size
+
+        return field_dict
+
+    @classmethod
+    def from_dict(cls: type[T], src_dict: Mapping[str, Any]) -> T:
+        d = dict(src_dict)
+        cardinality = d.pop("cardinality", UNSET)
+
+        null_count = d.pop("null_count", UNSET)
+
+        min_value = d.pop("min_value", UNSET)
+
+        max_value = d.pop("max_value", UNSET)
+
+        avg_size = d.pop("avg_size", UNSET)
+
+        field_statistics = cls(
+            cardinality=cardinality,
+            null_count=null_count,
+            min_value=min_value,
+            max_value=max_value,
+            avg_size=avg_size,
+        )
+
+        field_statistics.additional_properties = d
+        return field_statistics
+
+    @property
+    def additional_keys(self) -> list[str]:
+        return list(self.additional_properties.keys())
+
+    def __getitem__(self, key: str) -> Any:
+        return self.additional_properties[key]
+
+    def __setitem__(self, key: str, value: Any) -> None:
+        self.additional_properties[key] = value
+
+    def __delitem__(self, key: str) -> None:
+        del self.additional_properties[key]
+
+    def __contains__(self, key: str) -> bool:
+        return key in self.additional_properties