Source code for parseur.schemas.mailbox

from marshmallow import fields, validate

from parseur.schemas import BaseSchema
from parseur.schemas.document import DocumentStatus
from parseur.schemas.paserfield import ParserFieldSchema, TableFieldSchema
from parseur.schemas.webhook import WebhookSchema



[docs]class PageRangeSchema(BaseSchema):

[docs]    start_index = fields.Int(required=True)


[docs]    end_index = fields.Int(allow_none=True)




[docs]class SplitKeyWordsSchema(BaseSchema):

[docs]    is_before = fields.Boolean(required=True)


[docs]    keyword = fields.String(required=True)




[docs]class MailboxSchema(BaseSchema):

[docs]    id = fields.Int(required=True)


[docs]    name = fields.String(required=True)


[docs]    email_prefix = fields.String(required=True)


[docs]    account_uuid = fields.String(required=True)



[docs]    ai_engine = fields.String(required=True)


[docs]    ai_instructions = fields.String(allow_none=True)


    # ################
    # Basic settings #
    # ################

[docs]    decimal_separator = fields.String(
        allow_none=True,
        validate=validate.OneOf([".", ",", ""], error="Must be '.', ',' or null."),
    )


[docs]    default_timezone = fields.String(allow_none=True)



[docs]    default_language = fields.String(allow_none=True)


    # Input date format for parsing dates. Accepts "MONTH_FIRST", "DAY_FIRST", or None.
    #   MONTH_FIRST: mm/dd/yyyy, mm-dd-yyyy
    #   DAY_FIRST: dd/mm/yyyy, dd-mm-yyyy

[docs]    input_date_format = fields.String(
        allow_none=True,
        validate=validate.OneOf(
            ["MONTH_FIRST", "DAY_FIRST"],
            error="Must be 'MONTH_FIRST', 'DAY_FIRST', or null.",
        ),
    )

    # Disable auto-detection of input_date_format

[docs]    input_date_format_autodetection = fields.Boolean()

    # Parseur will automatically delete documents once they get older than the selected threshold.

[docs]    retention_policy = fields.Int(allow_none=True)


    # ###################
    # Advanced settings #
    # ###################

    # List of allowed file extensions for document processing.
    #   Example: ["pdf", "docx", "png"]

[docs]    allowed_extensions = fields.List(fields.String(), allow_none=True)


    # Force use of OCR on PDFs. Enable if data is garbled or text is in images.
    # Reprocess documents after enabling. May slow down processing.

[docs]    force_ocr = fields.Boolean(allow_none=True)


    # Expand field names in JSON Result.
    # Example: "user.name": "John" -> {"user": {"name": "John"}}.

[docs]    expand_result = fields.Boolean(allow_none=True)


    # Disable links on documents. Useful for manual data entry.


    # Disable the deskew algorithm if it creates a staircase effect when straightening.

[docs]    disable_deskew = fields.Boolean(allow_none=True)


    # Extract XML from HTML comments into separate documents.

[docs]    extract_xml_from_comment = fields.Boolean(allow_none=True)


    # Email sender block/allow list.
    #   True = allowlist mode (only allow listed senders).
    #   False = blocklist mode (block listed senders).

[docs]    use_whitelist_instead_of_blacklist = fields.Boolean(allow_none=True)


[docs]    emails_or_domains = fields.List(fields.String(), allow_none=True)


    # Email processing: process emails and attachments.

[docs]    process_attachments = fields.Boolean(required=True)


    # Email processing: process attachments only. Skip emails.

[docs]    attachments_only = fields.Boolean(required=True)


    # Page processing: only even pages (2, 4, 6, ...)

[docs]    even_pages = fields.Boolean(required=True)

    # Page processing: only odd pages (1, 3, 5, ...)

[docs]    odd_pages = fields.Boolean(required=True)

    # Page processing: only this page ranges. (same as split_page_range_set)

[docs]    page_range_set = fields.Nested(PageRangeSchema, allow_none=True, many=True)


    # Split documents every N pages.

[docs]    split_page = fields.Int(allow_none=True)

    # Split documents by page ranges.
    #   Example input: 1-5, 8, 11-13
    #   Enter ranges separated by commas. Use brackets to count from the end.
    #   E.g., (1) is last page. Example: 1, 2-(1) splits into two docs:
    #   - first page only
    #   - from page 2 to the end.

[docs]    split_page_range_set = fields.Nested(PageRangeSchema, allow_none=True, many=True)

    # Split documents by keywords.
    #   Enter the list of keywords to split on.
    #   Supports splitting before or after keywords.
    #   Keywords are case-sensitive.

[docs]    split_keywords = fields.Nested(SplitKeyWordsSchema, allow_none=True, many=True)


    # Counters

[docs]    document_count = fields.Int(allow_none=True)


[docs]    webhook_count = fields.Int(allow_none=True)


[docs]    template_count = fields.Int(allow_none=True)


[docs]    parser_object_count = fields.Int(allow_none=True)

    # Document per status count

[docs]    document_per_status_count = fields.Dict(
        keys=fields.String(validate=validate.OneOf([e.value for e in DocumentStatus])),
        values=fields.Int(),
        required=True,
    )


    # Last activity and modification timestamps

[docs]    last_activity = fields.DateTime(allow_none=True)


[docs]    template_set_last_modified = fields.DateTime(allow_none=True)


[docs]    parser_object_set_last_modified = fields.DateTime(allow_none=True)


    # URLs

[docs]    csv_download = fields.String(allow_none=True)


[docs]    json_download = fields.String(allow_none=True)


[docs]    xls_download = fields.String(allow_none=True)


    # Metadata fields

[docs]    attachments_field = fields.Boolean(required=True)


[docs]    bcc_field = fields.Boolean(required=True)


[docs]    cc_field = fields.Boolean(required=True)


[docs]    content_field = fields.Boolean(required=True)


[docs]    credit_count_field = fields.Boolean(required=True)


[docs]    document_id_field = fields.Boolean(required=True)


[docs]    document_url_field = fields.Boolean(required=True)


[docs]    headers_field = fields.Boolean(required=True)


[docs]    html_document_field = fields.Boolean(required=True)


[docs]    last_reply_field = fields.Boolean(required=True)


[docs]    mailbox_id_field = fields.Boolean(required=True)


[docs]    original_document_field = fields.Boolean(required=True)


[docs]    original_recipient_field = fields.Boolean(required=True)


[docs]    page_count_field = fields.Boolean(required=True)


[docs]    parsing_engine_field = fields.Boolean(required=True)


[docs]    processed_date_field = fields.Boolean(required=True)


[docs]    processed_field = fields.Boolean(required=True)


[docs]    processed_time_field = fields.Boolean(required=True)


[docs]    public_document_url_field = fields.Boolean(required=True)


[docs]    received_date_field = fields.Boolean(required=True)


[docs]    received_field = fields.Boolean(required=True)


[docs]    received_time_field = fields.Boolean(required=True)


[docs]    recipient_field = fields.Boolean(required=True)


[docs]    recipient_suffix_field = fields.Boolean(required=True)


[docs]    reply_to_field = fields.Boolean(required=True)


[docs]    searchable_pdf_field = fields.Boolean(required=True)


[docs]    sender_field = fields.Boolean(required=True)


[docs]    sender_name_field = fields.Boolean(required=True)


[docs]    split_page_range_field = fields.Boolean(required=True)


[docs]    split_parent_id_field = fields.Boolean(required=True)


[docs]    subject_field = fields.Boolean(required=True)


[docs]    template_field = fields.Boolean(required=True)


[docs]    text_document_field = fields.Boolean(required=True)


[docs]    to_field = fields.Boolean(required=True)


    # Webhooks

[docs]    available_webhook_set = fields.List(fields.Nested(WebhookSchema), required=True)


[docs]    webhook_set = fields.List(fields.Nested(WebhookSchema), required=True)


    # Parser and tables fields

[docs]    table_set = fields.List(fields.Nested(TableFieldSchema))


[docs]    parser_object_set = fields.List(fields.Nested(ParserFieldSchema))