retab 0.0.35__py3-none-any.whl → 0.0.37__py3-none-any.whl

uiform/types/deployments/mailboxes.py

@@ -1,67 +0,0 @@
-import copy
-import datetime
-import json
-import os
-import re
-from typing import Any, ClassVar, Dict, List, Literal, Optional
-
-import nanoid  # type: ignore
-from pydantic import BaseModel, EmailStr, Field, HttpUrl, computed_field, field_serializer, field_validator
-from pydantic_core import Url
-
-from ..._utils.json_schema import clean_schema
-from ..._utils.mime import generate_blake2b_hash_from_string
-from ..modalities import Modality
-from ..pagination import ListMetadata
-
-domain_pattern = re.compile(r"^(?:[a-zA-Z0-9-]+\.)+[a-zA-Z]{2,}$")
-
-from openai.types.chat.chat_completion_reasoning_effort import ChatCompletionReasoningEffort
-
-from ..logs import AutomationConfig, UpdateAutomationRequest
-
-
-class Mailbox(AutomationConfig):
-    EMAIL_PATTERN: ClassVar[str] = f".*@{os.getenv('EMAIL_DOMAIN', 'mailbox.uiform.com')}$"
-    object: Literal['deployment.mailbox'] = "deployment.mailbox"
-    id: str = Field(default_factory=lambda: "mb_" + nanoid.generate(), description="Unique identifier for the mailbox")
-
-    # Email Specific config
-    email: str = Field(..., pattern=EMAIL_PATTERN)
-    authorized_domains: list[str] = Field(default_factory=list, description="List of authorized domains to receive the emails from")
-    authorized_emails: List[EmailStr] = Field(default_factory=list, description="List of emails to access the link")
-
-    # Normalize email fields (case-insensitive)
-    @field_validator("email", mode="before")
-    def normalize_email(cls, value: str) -> str:
-        return value.strip().lower()
-
-    @field_validator("authorized_emails", mode="before")
-    def normalize_authorized_emails(cls, emails: List[str]) -> List[str]:
-        return [email.strip().lower() for email in emails]
-
-    @field_validator('authorized_domains', mode='before')
-    def validate_domain(cls, list_domains: list[str]) -> list[str]:
-        for domain in list_domains:
-            if not domain_pattern.match(domain):
-                raise ValueError(f"Invalid domain: {domain}")
-        return list_domains
-
-
-class ListMailboxes(BaseModel):
-    data: list[Mailbox]
-    list_metadata: ListMetadata
-
-
-# Inherits from the methods of UpdateAutomationRequest
-class UpdateMailboxRequest(UpdateAutomationRequest):
-
-    # ------------------------------
-    # Email Specific config
-    # ------------------------------
-    authorized_domains: Optional[list[str]] = None
-    authorized_emails: Optional[List[EmailStr]] = None
-
-    @field_validator("authorized_emails", mode="before")
-    def normalize_authorized_emails(cls, emails: Optional[List[str]]) -> Optional[List[str]]:
-        return [email.strip().lower() for email in emails] if emails else None

uiform/types/deployments/outlook.py

@@ -1,76 +0,0 @@
-import copy
-import datetime
-import json
-import os
-import re
-from typing import Any, ClassVar, Dict, List, Literal, Optional
-
-import nanoid  # type: ignore
-from openai.types.chat.chat_completion_reasoning_effort import ChatCompletionReasoningEffort
-from pydantic import BaseModel, EmailStr, Field, HttpUrl, computed_field, field_serializer, field_validator, model_validator
-from pydantic_core import Url
-
-from ..._utils.json_schema import clean_schema, convert_schema_to_layout
-from ..._utils.mime import generate_blake2b_hash_from_string
-from ..logs import AutomationConfig, UpdateAutomationRequest
-from ..modalities import Modality
-from ..pagination import ListMetadata
-
-domain_pattern = re.compile(r"^(?:[a-zA-Z0-9-]+\.)+[a-zA-Z]{2,}$")
-
-
-class AutomationLevel(BaseModel):
-    distance_threshold: float = Field(default=0.9, description="Distance threshold for the automation")
-    score_threshold: float = Field(default=0.9, description="Score threshold for the automation")
-
-
-class MatchParams(BaseModel):
-    endpoint: str = Field(..., description="Endpoint for matching parameters")
-    headers: Dict[str, str] = Field(..., description="Headers for the request")
-    path: str = Field(..., description="Path for matching parameters")
-
-
-class FetchParams(BaseModel):
-    endpoint: str = Field(..., description="Endpoint for fetching parameters")
-    headers: Dict[str, str] = Field(..., description="Headers for the request")
-    name: str = Field(..., description="Name of the fetch parameter")
-
-
-class Outlook(AutomationConfig):
-    object: Literal['deployment.outlook'] = "deployment.outlook"
-    id: str = Field(default_factory=lambda: "outlook_" + nanoid.generate(), description="Unique identifier for the outlook")
-
-    authorized_domains: list[str] = Field(default_factory=list, description="List of authorized domains to receive the emails from")
-    authorized_emails: List[EmailStr] = Field(default_factory=list, description="List of emails to access the link")
-
-    layout_schema: Optional[dict[str, Any]] = Field(default=None, description="Layout schema format used to display the data")
-
-    # Optional Fields for data integration
-    match_params: List[MatchParams] = Field(default_factory=list, description="List of match parameters for the outlook automation")
-    fetch_params: List[FetchParams] = Field(default_factory=list, description="List of fetch parameters for the outlook automation")
-
-    @model_validator(mode='before')
-    @classmethod
-    def compute_layout_schema(cls, values: dict[str, Any]) -> dict[str, Any]:
-        if values.get('layout_schema') is None:
-            values['layout_schema'] = convert_schema_to_layout(values['json_schema'])
-        return values
-
-class ListOutlooks(BaseModel):
-    data: list[Outlook]
-    list_metadata: ListMetadata
-
-
-# Inherits from the methods of UpdateAutomationRequest
-class UpdateOutlookRequest(UpdateAutomationRequest):
-    authorized_domains: Optional[list[str]] = None
-    authorized_emails: Optional[List[EmailStr]] = None
-
-    match_params: Optional[List[MatchParams]] = None
-    fetch_params: Optional[List[FetchParams]] = None
-
-    layout_schema: Optional[dict[str, Any]] = None
-
-    @field_validator("authorized_emails", mode="before")
-    def normalize_authorized_emails(cls, emails: Optional[List[str]]) -> Optional[List[str]]:
-        return [email.strip().lower() for email in emails] if emails else None

uiform/types/deployments/webhooks.py

@@ -1,21 +0,0 @@
-from typing import Any, Optional
-
-from pydantic import BaseModel, EmailStr
-
-from uiform.types.documents.extractions import UiParsedChatCompletion
-
-from ..mime import BaseMIMEData, MIMEData
-
-
-class WebhookRequest(BaseModel):
-    completion: UiParsedChatCompletion
-    user: Optional[EmailStr] = None
-    file_payload: MIMEData
-    metadata: Optional[dict[str, Any]] = None
-
-
-class BaseWebhookRequest(BaseModel):
-    completion: UiParsedChatCompletion
-    user: Optional[EmailStr] = None
-    file_payload: BaseMIMEData
-    metadata: Optional[dict[str, Any]] = None

uiform/types/events.py

@@ -1,76 +0,0 @@
-import datetime
-from typing import Any, Literal, Optional
-
-import nanoid  # type: ignore
-from pydantic import BaseModel, Field
-
-metadata_key = Literal[
-    'user',
-    'organization',
-    'link',
-    'mailbox',
-    'cron',
-    'outlook',
-    'extraction',
-    'webhook',
-    'reconciliation',
-    'preprocessing',
-    'schema',
-    'data_structure',
-    'file',
-    'preprocessing',
-    'dataset',
-    'dataset_membership',
-    'endpoint',
-    'automation',
-    'template',
-]
-
-event_type = Literal[
-    'extraction.created',
-    'messages.created',
-    'document.orientation_corrected',
-    'consensus.reconciled',
-    'automation.created',
-    'automation.updated',
-    'automation.deleted',
-    'automation.webhook',
-    'preprocessing.created',
-    'link.created',
-    'link.updated',
-    'link.deleted',
-    'link.webhook',
-    'mailbox.created',
-    'mailbox.updated',
-    'mailbox.deleted',
-    'mailbox.webhook',
-    'outlook.created',
-    'outlook.updated',
-    'outlook.deleted',
-    'outlook.webhook',
-    'schema.generated',
-    'schema.promptified',
-    'schema.system_promptfile.created',
-    'file.updated',
-    'file.deleted',
-    'template.created',
-    'template.deleted',
-    'template.sample_document_uploaded',
-    'template.sample_document_deleted',
-    'template.updated',
-]
-
-
-class Event(BaseModel):
-    object: Literal['event'] = "event"
-    id: str = Field(default_factory=lambda: "event_" + nanoid.generate(), description="Unique identifier for the event")
-    event: str = Field(..., description="A string that distinguishes the event type. Ex: user.created, user.updated, user.deleted, etc.")
-    created_at: datetime.datetime = Field(default_factory=lambda: datetime.datetime.now(datetime.timezone.utc))
-    data: dict[str, Any] = Field(..., description="Event payload. Payloads match the corresponding API objects.")
-    metadata: Optional[dict[metadata_key, str]] = Field(
-        default=None, description="Ids giving informations about the event. Ex: user.created.metadata = {'user': 'usr_8478973619047837'}"
-    )
-
-
-class StoredEvent(Event):
-    organization_id: str = Field(..., description="Organization ID")

uiform/types/jobs/base.py

@@ -1,150 +0,0 @@
-from typing import Literal, Optional, Self
-
-from openai.types.chat.chat_completion_reasoning_effort import ChatCompletionReasoningEffort
-from pydantic import BaseModel, Field, model_validator
-
-from ..modalities import Modality
-
-SelectionMode = Literal["all", "manual"]
-
-
-# This is the input data for the prepare_dataset job
-class PrepareDatasetInputData(BaseModel):
-    dataset_id: Optional[str] = None
-    schema_id: Optional[str] = None
-    schema_data_id: Optional[str] = None
-
-    selection_model: SelectionMode = "all"
-
-    @model_validator(mode="after")
-    def validate_input(self) -> Self:
-        # The preference is:
-        # 1. dataset_id
-        # 2. schema_id
-        # 3. schema_data_id
-        if self.dataset_id is None and self.schema_id is None and self.schema_data_id is None:
-            raise ValueError("At least one of dataset_id, schema_id, or schema_data_id must be provided")
-
-        return self
-
-
-# This is the input data for the split_dataset job
-class DatasetSplitInputData(BaseModel):
-    dataset_id: str
-    train_size: Optional[int | float] = None
-    eval_size: Optional[int | float] = None
-
-    @model_validator(mode="after")
-    def validate_input(self) -> Self:
-        if self.train_size is not None and self.eval_size is not None:
-            raise ValueError("train_size and eval_size cannot both be provided")
-        return self
-
-
-# This is the input data for the batch annotation job
-class InferenceSettings(BaseModel):
-    model: str = "gpt-4o-mini"
-    temperature: float = 0.0
-    modality: Modality = "native"
-    reasoning_effort: ChatCompletionReasoningEffort = "medium"
-    image_resolution_dpi: int = 96
-    browser_canvas: Literal['A3', 'A4', 'A5'] = 'A4'
-    n_consensus: int = Field(default=1, description="Number of consensus rounds to perform")
-
-
-class AnnotationInputData(BaseModel):
-    data_file: str
-    schema_id: str
-    inference_settings: InferenceSettings
-
-
-# This is the input data for the evaluation job
-class EvaluationInputData(BaseModel):
-    eval_data_file: str
-    schema_id: str
-    inference_settings_1: InferenceSettings | None = None
-    inference_settings_2: InferenceSettings
-
-
-# from pydantic import BaseModel, Field, model_validator
-# from typing import Literal, Optional, Any
-# import datetime
-
-
-# JobType = Literal["prompt-optimization", "annotate-files", "finetune-dataset", "webcrawl"]
-# JobStatus = Literal["pending", "running", "completed", "failed"]
-#### JOBS ####
-
-# class JobTemplateCreateRequest(BaseModel):
-#     job_type: JobType
-#     default_input_data: dict = Field(default_factory=dict)
-#     description: Optional[str] = None
-#     cron: Optional[str] = None
-
-
-# class JobTemplateDocument(BaseModel):
-#     object: Literal["job_template"] = "job_template"
-#     id: str
-#     type: JobType
-#     identity: Any | None = None
-#     description: Optional[str] = None
-#     default_input_data: dict = Field(default_factory=dict)
-#     # For scheduled jobs, include a valid CRON expression (None for on-demand only jobs)
-#     cron: Optional[str] = None
-#     next_run: Optional[datetime.datetime] = None
-#     created_at: Optional[datetime.datetime] = None
-#     updated_at: Optional[datetime.datetime] = None
-#     is_active: bool = True  # Change to status.
-
-# class JobTemplateUpdateRequest(BaseModel):
-#     cron: Optional[str] = None
-#     default_input_data: Optional[dict] = None
-#     description: Optional[str] = None
-#     is_active: Optional[bool] = None    # Change to status.
-
-
-#### EXECUTIONS ####
-
-# class JobExecutionCreateRequest(BaseModel):
-#     type: JobType
-#     template_id: Optional[str] = None
-#     input_data: dict = Field(default_factory=dict)
-
-#     @model_validator(mode='before')
-#     @classmethod
-#     def validate_job_identifiers(cls, data: Any) -> Any:
-#         if isinstance(data, dict):
-#             if bool(data.get('job_type')) == bool(data.get('job_template_id')):
-#                 raise ValueError("Either job_type or job_template_id must be provided")
-#         return data
-
-# class JobExecutionResponse(BaseModel):
-#     id: str
-#     template_id: Optional[str] = None
-#     type: JobType
-#     status: JobStatus
-#     result: Optional[dict] = None
-#     error: Optional[str] = None
-#     created_at: Optional[datetime.datetime] = None
-#     updated_at: Optional[datetime.datetime] = None
-
-# class JobExecutionDocument(BaseModel):
-#     object: Literal["job_execution"] = "job_execution"
-#     id: str
-#     template_id: Optional[str] = None
-#     type: JobType
-#     identity: Any | None = None
-#     status: JobStatus
-#     input_data_gcs_path: str
-#     result: Optional[dict] = None
-#     error: Optional[str] = None
-#     created_at: Optional[datetime.datetime] = None
-#     updated_at: Optional[datetime.datetime] = None
-#     checkpoint: Any = None  # Useful for jobs that need to be resumed
-#     checkpoint_data: Optional[dict] = None
-#     needs: list[str] = Field(default_factory=list, description="list of jobs execution id that must be completed before this job can run")
-
-
-# class Workflow(BaseModel):
-#     name: str
-#     jobs: list[JobExecutionDocument]

uiform/types/jobs/batch_annotation.py

@@ -1,22 +0,0 @@
-from typing import Literal, Optional
-
-from openai.types.chat.chat_completion_reasoning_effort import ChatCompletionReasoningEffort
-from pydantic import BaseModel, Field
-
-from ..modalities import Modality
-
-
-class InferenceSettings(BaseModel):
-    model: str = "gpt-4o-mini"
-    temperature: float = 0.0
-    modality: Modality
-    image_resolution_dpi: int = 96
-    browser_canvas: Literal['A3', 'A4', 'A5'] = 'A4'
-    reasoning_effort: ChatCompletionReasoningEffort = "medium"
-
-
-class AnnotationInputData(BaseModel):
-    dataset_id: str
-    files_ids: Optional[list[str]] = None
-    upsert: bool = False
-    inference_settings: InferenceSettings