Skip to content

Docling Document

This is an automatic generated API reference of the DoclingDocument type.

doc

Package for models defined by the Document type.

Classes:

DoclingDocument

Bases: BaseModel

DoclingDocument.

Methods:

Attributes:

body 

body: GroupItem = GroupItem(
    name="_root_", self_ref="#/body"
)

furniture 

furniture: GroupItem = GroupItem(
    name="_root_", self_ref="#/furniture"
)

groups 

groups: List[GroupItem] = []

key_value_items 

key_value_items: List[KeyValueItem] = []

name 

name: str

origin 

origin: Optional[DocumentOrigin] = None

pages 

pages: Dict[int, PageItem] = {}

pictures 

pictures: List[PictureItem] = []

schema_name 

schema_name: Literal['DoclingDocument'] = 'DoclingDocument'

tables 

tables: List[TableItem] = []

texts 

texts: List[
    Union[SectionHeaderItem, ListItem, TextItem]
] = []

version 

version: Annotated[
    str,
    StringConstraints(pattern=VERSION_PATTERN, strict=True),
] = CURRENT_VERSION

add_group 

add_group(
    label: Optional[GroupLabel] = None,
    name: Optional[str] = None,
    parent: Optional[NodeItem] = None,
) -> GroupItem

add_group.

Parameters:

  • label (Optional[GroupLabel], default: None ) –

    Optional[GroupLabel]: (Default value = None)

  • name (Optional[str], default: None ) –

    Optional[str]: (Default value = None)

  • parent (Optional[NodeItem], default: None ) –

    Optional[NodeItem]: (Default value = None)

add_heading 

add_heading(
    text: str,
    orig: Optional[str] = None,
    level: LevelNumber = 1,
    prov: Optional[ProvenanceItem] = None,
    parent: Optional[NodeItem] = None,
)

add_heading.

Parameters:

  • label –

    DocItemLabel:

  • text (str) –

    str:

  • orig (Optional[str], default: None ) –

    Optional[str]: (Default value = None)

  • level (LevelNumber, default: 1 ) –

    LevelNumber: (Default value = 1)

  • prov (Optional[ProvenanceItem], default: None ) –

    Optional[ProvenanceItem]: (Default value = None)

  • parent (Optional[NodeItem], default: None ) –

    Optional[NodeItem]: (Default value = None)

add_list_item 

add_list_item(
    text: str,
    enumerated: bool = False,
    marker: Optional[str] = None,
    orig: Optional[str] = None,
    prov: Optional[ProvenanceItem] = None,
    parent: Optional[NodeItem] = None,
)

add_list_item.

Parameters:

  • label –

    str:

  • text (str) –

    str:

  • orig (Optional[str], default: None ) –

    Optional[str]: (Default value = None)

  • prov (Optional[ProvenanceItem], default: None ) –

    Optional[ProvenanceItem]: (Default value = None)

  • parent (Optional[NodeItem], default: None ) –

    Optional[NodeItem]: (Default value = None)

add_page 

add_page(
    page_no: int,
    size: Size,
    image: Optional[ImageRef] = None,
) -> PageItem

add_page.

Parameters:

  • page_no (int) –

    int:

  • size (Size) –

    Size:

add_picture 

add_picture(
    annotations: List[PictureDataType] = [],
    image: Optional[ImageRef] = None,
    caption: Optional[Union[TextItem, RefItem]] = None,
    prov: Optional[ProvenanceItem] = None,
    parent: Optional[NodeItem] = None,
)

add_picture.

Parameters:

  • data –

    List[PictureData]: (Default value = [])

  • caption (Optional[Union[TextItem, RefItem]], default: None ) –

    Optional[Union[TextItem:

  • RefItem]] –

    (Default value = None)

  • prov (Optional[ProvenanceItem], default: None ) –

    Optional[ProvenanceItem]: (Default value = None)

  • parent (Optional[NodeItem], default: None ) –

    Optional[NodeItem]: (Default value = None)

add_table 

add_table(
    data: TableData,
    caption: Optional[Union[TextItem, RefItem]] = None,
    prov: Optional[ProvenanceItem] = None,
    parent: Optional[NodeItem] = None,
    label: DocItemLabel = TABLE,
)

add_table.

Parameters:

  • data (TableData) –

    TableData:

  • caption (Optional[Union[TextItem, RefItem]], default: None ) –

    Optional[Union[TextItem, RefItem]]: (Default value = None)

  • prov (Optional[ProvenanceItem], default: None ) –

    Optional[ProvenanceItem]: (Default value = None)

  • parent (Optional[NodeItem], default: None ) –

    Optional[NodeItem]: (Default value = None)

  • label (DocItemLabel, default: TABLE ) –

    DocItemLabel: (Default value = DocItemLabel.TABLE)

add_text 

add_text(
    label: DocItemLabel,
    text: str,
    orig: Optional[str] = None,
    prov: Optional[ProvenanceItem] = None,
    parent: Optional[NodeItem] = None,
)

add_text.

Parameters:

  • label (DocItemLabel) –

    str:

  • text (str) –

    str:

  • orig (Optional[str], default: None ) –

    Optional[str]: (Default value = None)

  • prov (Optional[ProvenanceItem], default: None ) –

    Optional[ProvenanceItem]: (Default value = None)

  • parent (Optional[NodeItem], default: None ) –

    Optional[NodeItem]: (Default value = None)

add_title 

add_title(
    text: str,
    orig: Optional[str] = None,
    prov: Optional[ProvenanceItem] = None,
    parent: Optional[NodeItem] = None,
)

add_title.

Parameters:

  • text (str) –

    str:

  • orig (Optional[str], default: None ) –

    Optional[str]: (Default value = None)

  • prov (Optional[ProvenanceItem], default: None ) –

    Optional[ProvenanceItem]: (Default value = None)

  • parent (Optional[NodeItem], default: None ) –

    Optional[NodeItem]: (Default value = None)

check_version_is_compatible 

check_version_is_compatible(v: str) -> str

Check if this document version is compatible with current version.

export_to_dict 

export_to_dict(
    mode: str = "json",
    by_alias: bool = True,
    exclude_none: bool = True,
) -> Dict

Export to dict.

export_to_document_tokens 

export_to_document_tokens(
    delim: str = "\n",
    from_element: int = 0,
    to_element: int = maxsize,
    labels: set[DocItemLabel] = DEFAULT_EXPORT_LABELS,
    xsize: int = 100,
    ysize: int = 100,
    add_location: bool = True,
    add_content: bool = True,
    add_page_index: bool = True,
    add_table_cell_location: bool = False,
    add_table_cell_label: bool = True,
    add_table_cell_text: bool = True,
    page_no: Optional[int] = None,
    with_groups: bool = True,
    newline: bool = True,
) -> str

Exports the document content to a DocumentToken format.

Operates on a slice of the document's body as defined through arguments from_element and to_element; defaulting to the whole main_text.

Parameters:

  • delim (str, default: '\n' ) –

    str: (Default value = "\n\n")

  • from_element (int, default: 0 ) –

    int: (Default value = 0)

  • to_element (int, default: maxsize ) –

    Optional[int]: (Default value = None)

  • labels (set[DocItemLabel], default: DEFAULT_EXPORT_LABELS ) –

    set[DocItemLabel]

  • xsize (int, default: 100 ) –

    int: (Default value = 100)

  • ysize (int, default: 100 ) –

    int: (Default value = 100)

  • add_location (bool, default: True ) –

    bool: (Default value = True)

  • add_content (bool, default: True ) –

    bool: (Default value = True)

  • add_page_index (bool, default: True ) –

    bool: (Default value = True)

  • add_table_cell_label (bool, default: True ) –

    bool: (Default value = True)

  • add_table_cell_text (bool, default: True ) –

    bool: (Default value = True)

Returns:

  • str –

    The content of the document formatted as a DocTags string.

export_to_element_tree 

export_to_element_tree() -> str

Export_to_element_tree.

export_to_html 

export_to_html(
    from_element: int = 0,
    to_element: int = maxsize,
    labels: set[DocItemLabel] = DEFAULT_EXPORT_LABELS,
    image_mode: ImageRefMode = PLACEHOLDER,
    page_no: Optional[int] = None,
    html_lang: str = "en",
    html_head: str = _HTML_DEFAULT_HEAD,
) -> str

Serialize to HTML.

export_to_markdown 

export_to_markdown(
    delim: str = "\n",
    from_element: int = 0,
    to_element: int = maxsize,
    labels: set[DocItemLabel] = DEFAULT_EXPORT_LABELS,
    strict_text: bool = False,
    image_placeholder: str = "<!-- image -->",
    image_mode: ImageRefMode = PLACEHOLDER,
    indent: int = 4,
    text_width: int = -1,
    page_no: Optional[int] = None,
) -> str

Serialize to Markdown.

Operates on a slice of the document's body as defined through arguments from_element and to_element; defaulting to the whole document.

Parameters:

  • delim (str, default: '\n' ) –

    Delimiter to use when concatenating the various Markdown parts. (Default value = "\n").

  • from_element (int, default: 0 ) –

    Body slicing start index (inclusive). (Default value = 0).

  • to_element (int, default: maxsize ) –

    Body slicing stop index (exclusive). (Default value = maxint).

  • labels (set[DocItemLabel], default: DEFAULT_EXPORT_LABELS ) –

    The set of document labels to include in the export.

  • strict_text (bool, default: False ) –

    bool: Whether to only include the text content of the document. (Default value = False).

  • image_placeholder (str, default: '<!-- image -->' ) –

    The placeholder to include to position images in the markdown. (Default value = "\<!-- image -->").

  • image_mode (ImageRefMode, default: PLACEHOLDER ) –

    The mode to use for including images in the markdown. (Default value = ImageRefMode.PLACEHOLDER).

  • indent (int, default: 4 ) –

    The indent in spaces of the nested lists. (Default value = 4).

Returns:

  • str –

    The exported Markdown representation.

export_to_text 

export_to_text(
    delim: str = "\n\n",
    from_element: int = 0,
    to_element: int = 1000000,
    labels: set[DocItemLabel] = DEFAULT_EXPORT_LABELS,
) -> str

export_to_text.

iterate_items 

iterate_items(
    root: Optional[NodeItem] = None,
    with_groups: bool = False,
    traverse_pictures: bool = False,
    page_no: Optional[int] = None,
    _level: int = 0,
) -> Iterable[Tuple[NodeItem, int]]

iterate_elements.

Parameters:

  • root (Optional[NodeItem], default: None ) –

    Optional[NodeItem]: (Default value = None)

  • with_groups (bool, default: False ) –

    bool: (Default value = False)

  • traverse_pictures (bool, default: False ) –

    bool: (Default value = True)

  • page_no (Optional[int], default: None ) –

    Optional[int]: (Default value = None)

  • _level (int, default: 0 ) –

    (Default value = 0)

load_from_json 

load_from_json(filename: Path) -> DoclingDocument

load_from_json.

Parameters:

  • filename (Path) –

    The filename to load a saved DoclingDocument from a .json.

Returns:

  • DoclingDocument –

    The loaded DoclingDocument.

num_pages 

num_pages()

num_pages.

print_element_tree 

print_element_tree()

Print_element_tree.

save_as_document_tokens 

save_as_document_tokens(
    filename: Path,
    delim: str = "\n\n",
    from_element: int = 0,
    to_element: int = maxsize,
    labels: set[DocItemLabel] = DEFAULT_EXPORT_LABELS,
    xsize: int = 100,
    ysize: int = 100,
    add_location: bool = True,
    add_content: bool = True,
    add_page_index: bool = True,
    add_table_cell_location: bool = False,
    add_table_cell_label: bool = True,
    add_table_cell_text: bool = True,
    page_no: Optional[int] = None,
    with_groups: bool = True,
)

Save the document content to a DocumentToken format.

save_as_html 

save_as_html(
    filename: Path,
    artifacts_dir: Optional[Path] = None,
    from_element: int = 0,
    to_element: int = maxsize,
    labels: set[DocItemLabel] = DEFAULT_EXPORT_LABELS,
    image_mode: ImageRefMode = PLACEHOLDER,
    page_no: Optional[int] = None,
    html_lang: str = "en",
    html_head: str = _HTML_DEFAULT_HEAD,
)

Save to HTML.

save_as_json 

save_as_json(
    filename: Path,
    artifacts_dir: Optional[Path] = None,
    image_mode: ImageRefMode = EMBEDDED,
    indent: int = 2,
)

Save as json.

save_as_markdown 

save_as_markdown(
    filename: Path,
    artifacts_dir: Optional[Path] = None,
    delim: str = "\n",
    from_element: int = 0,
    to_element: int = maxsize,
    labels: set[DocItemLabel] = DEFAULT_EXPORT_LABELS,
    strict_text: bool = False,
    image_placeholder: str = "<!-- image -->",
    image_mode: ImageRefMode = PLACEHOLDER,
    indent: int = 4,
    text_width: int = -1,
    page_no: Optional[int] = None,
)

Save to markdown.

save_as_yaml 

save_as_yaml(
    filename: Path,
    artifacts_dir: Optional[Path] = None,
    image_mode: ImageRefMode = EMBEDDED,
    default_flow_style: bool = False,
)

Save as yaml.

validate_document 

validate_document(d: DoclingDocument)

validate_document.

validate_tree 

validate_tree(root) -> bool

validate_tree.

DocumentOrigin

Bases: BaseModel

FileSource.

Methods:

Attributes:

binary_hash 

binary_hash: Uint64

filename 

filename: str

mimetype 

mimetype: str

uri 

uri: Optional[AnyUrl] = None

parse_hex_string 

parse_hex_string(value)

parse_hex_string.

validate_mimetype 

validate_mimetype(v)

validate_mimetype.

DocItem

Bases: NodeItem

DocItem.

Methods:

Attributes:

children 

children: List[RefItem] = []

label 

label: DocItemLabel

model_config 

model_config = ConfigDict(extra='forbid')

parent 

parent: Optional[RefItem] = None

prov 

prov: List[ProvenanceItem] = []

self_ref 

self_ref: str = Field(pattern=_JSON_POINTER_REGEX)

get_image 

get_image(doc: DoclingDocument) -> Optional[Image]

Returns the image of this DocItem.

The function returns None if this DocItem has no valid provenance or if a valid image of the page containing this DocItem is not available in doc.

get_location_tokens 

get_location_tokens(
    doc: DoclingDocument,
    new_line: str,
    xsize: int = 100,
    ysize: int = 100,
    add_page_index: bool = True,
) -> str

Get the location string for the BaseCell.

get_ref 

get_ref()

get_ref.

DocItemLabel

Bases: str, Enum

DocItemLabel.

Methods:

  • get_color –

    Return the RGB color associated with a given label.

Attributes:

CAPTION 

CAPTION = 'caption'

CHECKBOX_SELECTED 

CHECKBOX_SELECTED = 'checkbox_selected'

CHECKBOX_UNSELECTED 

CHECKBOX_UNSELECTED = 'checkbox_unselected'

CODE 

CODE = 'code'

DOCUMENT_INDEX 

DOCUMENT_INDEX = 'document_index'

FOOTNOTE 

FOOTNOTE = 'footnote'

FORM 

FORM = 'form'

FORMULA 

FORMULA = 'formula'

KEY_VALUE_REGION 

KEY_VALUE_REGION = 'key_value_region'

LIST_ITEM 

LIST_ITEM = 'list_item'

PAGE_FOOTER = 'page_footer'

PAGE_HEADER 

PAGE_HEADER = 'page_header'

PARAGRAPH 

PARAGRAPH = 'paragraph'

PICTURE 

PICTURE = 'picture'

REFERENCE 

REFERENCE = 'reference'

SECTION_HEADER 

SECTION_HEADER = 'section_header'

TABLE 

TABLE = 'table'

TEXT 

TEXT = 'text'

TITLE 

TITLE = 'title'

get_color 

get_color(label: DocItemLabel) -> Tuple[int, int, int]

Return the RGB color associated with a given label.

ProvenanceItem

Bases: BaseModel

ProvenanceItem.

Attributes:

bbox 

bbox: BoundingBox

charspan 

charspan: Tuple[int, int]

page_no 

page_no: int

GroupItem

Bases: NodeItem

GroupItem.

Methods:

Attributes:

children 

children: List[RefItem] = []

label 

label: GroupLabel = UNSPECIFIED

model_config 

model_config = ConfigDict(extra='forbid')

name 

name: str = 'group'

parent 

parent: Optional[RefItem] = None

self_ref 

self_ref: str = Field(pattern=_JSON_POINTER_REGEX)

get_ref 

get_ref()

get_ref.

GroupLabel

Bases: str, Enum

GroupLabel.

Attributes:

CHAPTER 

CHAPTER = 'chapter'

COMMENT_SECTION 

COMMENT_SECTION = 'comment_section'

FORM_AREA 

FORM_AREA = 'form_area'

KEY_VALUE_AREA 

KEY_VALUE_AREA = 'key_value_area'

LIST 

LIST = 'list'

ORDERED_LIST 

ORDERED_LIST = 'ordered_list'

SECTION 

SECTION = 'section'

SHEET 

SHEET = 'sheet'

SLIDE 

SLIDE = 'slide'

UNSPECIFIED 

UNSPECIFIED = 'unspecified'

NodeItem

Bases: BaseModel

NodeItem.

Methods:

Attributes:

children 

children: List[RefItem] = []

model_config 

model_config = ConfigDict(extra='forbid')

parent 

parent: Optional[RefItem] = None

self_ref 

self_ref: str = Field(pattern=_JSON_POINTER_REGEX)

get_ref 

get_ref()

get_ref.

PageItem

Bases: BaseModel

PageItem.

Attributes:

image 

image: Optional[ImageRef] = None

page_no 

page_no: int

size 

size: Size

FloatingItem

Bases: DocItem

FloatingItem.

Methods:

Attributes:

captions 

captions: List[RefItem] = []

children 

children: List[RefItem] = []

footnotes 

footnotes: List[RefItem] = []

image 

image: Optional[ImageRef] = None

label 

label: DocItemLabel

model_config 

model_config = ConfigDict(extra='forbid')

parent 

parent: Optional[RefItem] = None

prov 

prov: List[ProvenanceItem] = []

references 

references: List[RefItem] = []

self_ref 

self_ref: str = Field(pattern=_JSON_POINTER_REGEX)

caption_text 

caption_text(doc: DoclingDocument) -> str

Computes the caption as a single text.

get_image 

get_image(doc: DoclingDocument) -> Optional[Image]

Returns the image corresponding to this FloatingItem.

This function returns the PIL image from self.image if one is available. Otherwise, it uses DocItem.get_image to get an image of this FloatingItem.

In particular, when self.image is None, the function returns None if this FloatingItem has no valid provenance or the doc does not contain a valid image for the required page.

get_location_tokens 

get_location_tokens(
    doc: DoclingDocument,
    new_line: str,
    xsize: int = 100,
    ysize: int = 100,
    add_page_index: bool = True,
) -> str

Get the location string for the BaseCell.

get_ref 

get_ref()

get_ref.

TextItem

Bases: DocItem

TextItem.

Methods:

Attributes:

children 

children: List[RefItem] = []

label 

label: Literal[
    CAPTION,
    CHECKBOX_SELECTED,
    CHECKBOX_UNSELECTED,
    CODE,
    FOOTNOTE,
    FORMULA,
    PAGE_FOOTER,
    PAGE_HEADER,
    PARAGRAPH,
    REFERENCE,
    TEXT,
    TITLE,
]

model_config 

model_config = ConfigDict(extra='forbid')

orig 

orig: str

parent 

parent: Optional[RefItem] = None

prov 

prov: List[ProvenanceItem] = []

self_ref 

self_ref: str = Field(pattern=_JSON_POINTER_REGEX)

text 

text: str

export_to_document_tokens 

export_to_document_tokens(
    doc: DoclingDocument,
    new_line: str = "\n",
    xsize: int = 100,
    ysize: int = 100,
    add_location: bool = True,
    add_content: bool = True,
    add_page_index: bool = True,
)

Export text element to document tokens format.

Parameters:

  • doc (DoclingDocument) –

    "DoclingDocument":

  • new_line (str, default: '\n' ) –

    str: (Default value = "\n")

  • xsize (int, default: 100 ) –

    int: (Default value = 100)

  • ysize (int, default: 100 ) –

    int: (Default value = 100)

  • add_location (bool, default: True ) –

    bool: (Default value = True)

  • add_content (bool, default: True ) –

    bool: (Default value = True)

  • add_page_index (bool, default: True ) –

    bool: (Default value = True)

get_image 

get_image(doc: DoclingDocument) -> Optional[Image]

Returns the image of this DocItem.

The function returns None if this DocItem has no valid provenance or if a valid image of the page containing this DocItem is not available in doc.

get_location_tokens 

get_location_tokens(
    doc: DoclingDocument,
    new_line: str,
    xsize: int = 100,
    ysize: int = 100,
    add_page_index: bool = True,
) -> str

Get the location string for the BaseCell.

get_ref 

get_ref()

get_ref.

TableItem

Bases: FloatingItem

TableItem.

Methods:

Attributes:

captions 

captions: List[RefItem] = []

children 

children: List[RefItem] = []

data 

data: TableData

footnotes 

footnotes: List[RefItem] = []

image 

image: Optional[ImageRef] = None

label 

label: Literal[DOCUMENT_INDEX, TABLE] = TABLE

model_config 

model_config = ConfigDict(extra='forbid')

parent 

parent: Optional[RefItem] = None

prov 

prov: List[ProvenanceItem] = []

references 

references: List[RefItem] = []

self_ref 

self_ref: str = Field(pattern=_JSON_POINTER_REGEX)

caption_text 

caption_text(doc: DoclingDocument) -> str

Computes the caption as a single text.

export_to_dataframe 

export_to_dataframe() -> DataFrame

Export the table as a Pandas DataFrame.

export_to_document_tokens 

export_to_document_tokens(
    doc: DoclingDocument,
    new_line: str = "\n",
    xsize: int = 100,
    ysize: int = 100,
    add_location: bool = True,
    add_caption: bool = True,
    add_content: bool = True,
    add_cell_location: bool = True,
    add_cell_label: bool = True,
    add_cell_text: bool = True,
    add_page_index: bool = True,
)

Export table to document tokens format.

Parameters:

  • doc (DoclingDocument) –

    "DoclingDocument":

  • new_line (str, default: '\n' ) –

    str: (Default value = "\n")

  • xsize (int, default: 100 ) –

    int: (Default value = 100)

  • ysize (int, default: 100 ) –

    int: (Default value = 100)

  • add_location (bool, default: True ) –

    bool: (Default value = True)

  • add_caption (bool, default: True ) –

    bool: (Default value = True)

  • add_content (bool, default: True ) –

    bool: (Default value = True)

  • add_cell_location (bool, default: True ) –

    bool: (Default value = True)

  • add_cell_label (bool, default: True ) –

    bool: (Default value = True)

  • add_cell_text (bool, default: True ) –

    bool: (Default value = True)

  • add_page_index (bool, default: True ) –

    bool: (Default value = True)

export_to_html 

export_to_html(
    doc: Optional[DoclingDocument] = None,
    add_caption: bool = True,
) -> str

Export the table as html.

export_to_markdown 

export_to_markdown() -> str

Export the table as markdown.

export_to_otsl 

export_to_otsl(
    doc: DoclingDocument,
    add_cell_location: bool = True,
    add_cell_text: bool = True,
    xsize: int = 100,
    ysize: int = 100,
) -> str

Export the table as OTSL.

get_image 

get_image(doc: DoclingDocument) -> Optional[Image]

Returns the image corresponding to this FloatingItem.

This function returns the PIL image from self.image if one is available. Otherwise, it uses DocItem.get_image to get an image of this FloatingItem.

In particular, when self.image is None, the function returns None if this FloatingItem has no valid provenance or the doc does not contain a valid image for the required page.

get_location_tokens 

get_location_tokens(
    doc: DoclingDocument,
    new_line: str,
    xsize: int = 100,
    ysize: int = 100,
    add_page_index: bool = True,
) -> str

Get the location string for the BaseCell.

get_ref 

get_ref()

get_ref.

TableCell

Bases: BaseModel

TableCell.

Methods:

Attributes:

bbox 

bbox: Optional[BoundingBox] = None

col_span 

col_span: int = 1

column_header 

column_header: bool = False

end_col_offset_idx 

end_col_offset_idx: int

end_row_offset_idx 

end_row_offset_idx: int

row_header 

row_header: bool = False

row_section 

row_section: bool = False

row_span 

row_span: int = 1

start_col_offset_idx 

start_col_offset_idx: int

start_row_offset_idx 

start_row_offset_idx: int

text 

text: str

from_dict_format 

from_dict_format(data: Any) -> Any

from_dict_format.

TableData

Bases: BaseModel

BaseTableData.

Attributes:

grid 

grid: List[List[TableCell]]

grid.

num_cols 

num_cols: int = 0

num_rows 

num_rows: int = 0

table_cells 

table_cells: List[TableCell] = []

TableCellLabel

Bases: str, Enum

TableCellLabel.

Attributes:

BODY 

BODY = 'body'

COLUMN_HEADER 

COLUMN_HEADER = 'col_header'

ROW_HEADER 

ROW_HEADER = 'row_header'

ROW_SECTION 

ROW_SECTION = 'row_section'

KeyValueItem

Bases: DocItem

KeyValueItem.

Methods:

Attributes:

children 

children: List[RefItem] = []

label 

label: Literal[KEY_VALUE_REGION] = KEY_VALUE_REGION

model_config 

model_config = ConfigDict(extra='forbid')

parent 

parent: Optional[RefItem] = None

prov 

prov: List[ProvenanceItem] = []

self_ref 

self_ref: str = Field(pattern=_JSON_POINTER_REGEX)

get_image 

get_image(doc: DoclingDocument) -> Optional[Image]

Returns the image of this DocItem.

The function returns None if this DocItem has no valid provenance or if a valid image of the page containing this DocItem is not available in doc.

get_location_tokens 

get_location_tokens(
    doc: DoclingDocument,
    new_line: str,
    xsize: int = 100,
    ysize: int = 100,
    add_page_index: bool = True,
) -> str

Get the location string for the BaseCell.

get_ref 

get_ref()

get_ref.

SectionHeaderItem

Bases: TextItem

SectionItem.

Methods:

Attributes:

children 

children: List[RefItem] = []

label 

label: Literal[SECTION_HEADER] = SECTION_HEADER

level 

level: LevelNumber = 1

model_config 

model_config = ConfigDict(extra='forbid')

orig 

orig: str

parent 

parent: Optional[RefItem] = None

prov 

prov: List[ProvenanceItem] = []

self_ref 

self_ref: str = Field(pattern=_JSON_POINTER_REGEX)

text 

text: str

export_to_document_tokens 

export_to_document_tokens(
    doc: DoclingDocument,
    new_line: str = "\n",
    xsize: int = 100,
    ysize: int = 100,
    add_location: bool = True,
    add_content: bool = True,
    add_page_index: bool = True,
)

Export text element to document tokens format.

Parameters:

  • doc (DoclingDocument) –

    "DoclingDocument":

  • new_line (str, default: '\n' ) –

    str: (Default value = "\n")

  • xsize (int, default: 100 ) –

    int: (Default value = 100)

  • ysize (int, default: 100 ) –

    int: (Default value = 100)

  • add_location (bool, default: True ) –

    bool: (Default value = True)

  • add_content (bool, default: True ) –

    bool: (Default value = True)

  • add_page_index (bool, default: True ) –

    bool: (Default value = True)

get_image 

get_image(doc: DoclingDocument) -> Optional[Image]

Returns the image of this DocItem.

The function returns None if this DocItem has no valid provenance or if a valid image of the page containing this DocItem is not available in doc.

get_location_tokens 

get_location_tokens(
    doc: DoclingDocument,
    new_line: str,
    xsize: int = 100,
    ysize: int = 100,
    add_page_index: bool = True,
) -> str

Get the location string for the BaseCell.

get_ref 

get_ref()

get_ref.

PictureItem

Bases: FloatingItem

PictureItem.

Methods:

Attributes:

annotations 

annotations: List[PictureDataType] = []

captions 

captions: List[RefItem] = []

children 

children: List[RefItem] = []

footnotes 

footnotes: List[RefItem] = []

image 

image: Optional[ImageRef] = None

label 

label: Literal[PICTURE] = PICTURE

model_config 

model_config = ConfigDict(extra='forbid')

parent 

parent: Optional[RefItem] = None

prov 

prov: List[ProvenanceItem] = []

references 

references: List[RefItem] = []

self_ref 

self_ref: str = Field(pattern=_JSON_POINTER_REGEX)

caption_text 

caption_text(doc: DoclingDocument) -> str

Computes the caption as a single text.

export_to_document_tokens 

export_to_document_tokens(
    doc: DoclingDocument,
    new_line: str = "\n",
    xsize: int = 100,
    ysize: int = 100,
    add_location: bool = True,
    add_caption: bool = True,
    add_content: bool = True,
    add_page_index: bool = True,
)

Export picture to document tokens format.

Parameters:

  • doc (DoclingDocument) –

    "DoclingDocument":

  • new_line (str, default: '\n' ) –

    str: (Default value = "\n")

  • xsize (int, default: 100 ) –

    int: (Default value = 100)

  • ysize (int, default: 100 ) –

    int: (Default value = 100)

  • add_location (bool, default: True ) –

    bool: (Default value = True)

  • add_caption (bool, default: True ) –

    bool: (Default value = True)

  • add_content (bool, default: True ) –

    bool: (Default value = True)

export_to_html 

export_to_html(
    doc: DoclingDocument,
    add_caption: bool = True,
    image_mode: ImageRefMode = PLACEHOLDER,
) -> str

Export picture to HTML format.

export_to_markdown 

export_to_markdown(
    doc: DoclingDocument,
    add_caption: bool = True,
    image_mode: ImageRefMode = EMBEDDED,
    image_placeholder: str = "<!-- image -->",
) -> str

Export picture to Markdown format.

get_image 

get_image(doc: DoclingDocument) -> Optional[Image]

Returns the image corresponding to this FloatingItem.

This function returns the PIL image from self.image if one is available. Otherwise, it uses DocItem.get_image to get an image of this FloatingItem.

In particular, when self.image is None, the function returns None if this FloatingItem has no valid provenance or the doc does not contain a valid image for the required page.

get_location_tokens 

get_location_tokens(
    doc: DoclingDocument,
    new_line: str,
    xsize: int = 100,
    ysize: int = 100,
    add_page_index: bool = True,
) -> str

Get the location string for the BaseCell.

get_ref 

get_ref()

get_ref.

ImageRef

Bases: BaseModel

ImageRef.

Methods:

Attributes:

dpi 

dpi: int

mimetype 

mimetype: str

pil_image 

pil_image: Optional[Image]

Return the PIL Image.

size 

size: Size

uri 

uri: Union[AnyUrl, Path] = Field(union_mode="left_to_right")

from_pil 

from_pil(image: Image, dpi: int) -> Self

Construct ImageRef from a PIL Image.

validate_mimetype 

validate_mimetype(v)

validate_mimetype.

PictureClassificationClass

Bases: BaseModel

PictureClassificationData.

Attributes:

class_name 

class_name: str

confidence 

confidence: float

PictureClassificationData

Bases: BasePictureData

PictureClassificationData.

Attributes:

kind 

kind: Literal['classification'] = 'classification'

predicted_classes 

predicted_classes: List[PictureClassificationClass]

provenance 

provenance: str

RefItem

Bases: BaseModel

RefItem.

Methods:

Attributes:

cref 

cref: str = Field(alias="$ref", pattern=_JSON_POINTER_REGEX)

model_config 

model_config = ConfigDict(populate_by_name=True)

get_ref 

get_ref()

get_ref.

resolve 

resolve(doc: DoclingDocument)

resolve.

BoundingBox

Bases: BaseModel

BoundingBox.

Methods:

Attributes:

b 

b: float

coord_origin 

coord_origin: CoordOrigin = TOPLEFT

height 

height

height.

l 

l: float

r 

r: float

t 

t: float

width 

width

width.

area 

area() -> float

area.

as_tuple 

as_tuple()

as_tuple.

from_tuple 

from_tuple(coord: Tuple[float, ...], origin: CoordOrigin)

from_tuple.

Parameters:

  • coord (Tuple[float, ...]) –

    Tuple[float:

  • ...] –
  • origin (CoordOrigin) –

    CoordOrigin:

intersection_area_with 

intersection_area_with(other: BoundingBox) -> float

intersection_area_with.

Parameters:

normalized 

normalized(page_size: Size) -> BoundingBox

normalized.

Parameters:

  • page_size (Size) –

    Size:

scaled 

scaled(scale: float) -> BoundingBox

scaled.

Parameters:

  • scale (float) –

    float:

to_bottom_left_origin 

to_bottom_left_origin(page_height) -> BoundingBox

to_bottom_left_origin.

Parameters:

  • page_height –

to_top_left_origin 

to_top_left_origin(page_height)

to_top_left_origin.

Parameters:

  • page_height –

CoordOrigin

Bases: str, Enum

CoordOrigin.

Attributes:

BOTTOMLEFT 

BOTTOMLEFT = 'BOTTOMLEFT'

TOPLEFT 

TOPLEFT = 'TOPLEFT'

ImageRefMode

Bases: str, Enum

ImageRefMode.

Attributes:

EMBEDDED 

EMBEDDED = 'embedded'

PLACEHOLDER 

PLACEHOLDER = 'placeholder'

REFERENCED 

REFERENCED = 'referenced'

Size

Bases: BaseModel

Size.

Methods:

Attributes:

height 

height: float = 0.0

width 

width: float = 0.0

as_tuple 

as_tuple()

as_tuple.