docling-serve-client 0.0.0-stage → 0.4.0

package/dist/core/validate.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"validate.d.ts","sourceRoot":"","sources":["../../src/core/validate.ts"],"names":[],"mappings":"AAOA,2CAA2C;AAC3C,MAAM,MAAM,UAAU,GAAG,CAAC,KAAK,EAAE,OAAO,KAAK,OAAO,CAAC;AAMrD,iFAAiF;AACjF,eAAO,MAAM,uBAAuB,wCAAwC,CAAC;AAM7E;;;;;;;;GAQG;AACH,wBAAgB,YAAY,CAAC,CAAC,SAAS,MAAM,EACzC,KAAK,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,EAC9B,KAAK,EAAE,MAAM,CAAC,MAAM,CAAC,EAAE,UAAU,CAAC,EAClC,UAAU,SAAK,GAChB,OAAO,CAAC,CAAC,CAAC,CAiBZ;AAED,uDAAuD;AACvD,wBAAgB,YAAY,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAEjD;AAED,yDAAyD;AACzD,wBAAgB,YAAY,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAEjD;AAED;;;;GAIG;AACH,wBAAgB,aAAa,CAAC,KAAK,EAAE,OAAO,GAAG,KAAK,IAAI,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAM9E;AAED,mDAAmD;AACnD,wBAAgB,SAAS,CAAC,KAAK,EAAE,OAAO,GAAG,KAAK,IAAI,OAAO,CAE1D;AAED,qEAAqE;AACrE,wBAAgB,OAAO,CAAC,CAAC,SAAS,MAAM,EAAE,KAAK,EAAE,OAAO,EAAE,OAAO,EAAE,SAAS,CAAC,EAAE,GAAG,KAAK,IAAI,CAAC,CAE3F"}

package/dist/document.d.ts

@@ -0,0 +1,412 @@
+/**
+ * **DoclingDocument as sent by docling-serve in `json_content`**: schema `1.10.0`, docling-core 2.101.
+ *
+ * - items live in flat arrays (`texts`, `tables`, ...) and point to each other with `RefItem`s (`#/texts/12`)
+ * - reading order: walk `body.children` depth-first; `furniture` holds page headers and footers
+ * - the response is only checked for `schema_name` and `body`; deeper fields are typed, not validated
+ */
+export type DoclingDocument = {
+    schema_name: 'DoclingDocument';
+    /** Schema version, e.g. `'1.10.0'`. */
+    version: string;
+    name: string;
+    origin: DocumentOrigin | null;
+    /** Root of page headers and footers. */
+    furniture: GroupItem;
+    /** Root of the content in reading order. */
+    body: GroupItem;
+    groups: (ListGroup | InlineGroup | GroupItem)[];
+    texts: AnyTextItem[];
+    pictures: PictureItem[];
+    tables: TableItem[];
+    key_value_items: KeyValueItem[];
+    form_items: FormItem[];
+    /** Not sent by every docling version. */
+    field_regions?: FieldRegionItem[];
+    /** Not sent by every docling version. */
+    field_items?: FieldItem[];
+    /** PDF pages keyed by their 1-based number as string, e.g. `'1'`. */
+    pages: Record<string, PageItem>;
+};
+/** Source file of the document. */
+export type DocumentOrigin = {
+    mimetype: string;
+    binary_hash: number;
+    filename: string;
+    uri?: string | null;
+};
+/** One PDF page. */
+export type PageItem = {
+    size: Size;
+    /** Decoded page images become `null` (returned in `pageImages`); undecodable ones stay as sent. */
+    image: ImageRef | null;
+    page_no: number;
+};
+/** JSON pointer to another item, e.g. `{ $ref: '#/texts/12' }` or `{ $ref: '#/body' }`. */
+export type RefItem = {
+    $ref: string;
+};
+/**
+ * **Pointer to part of an item's text**.
+ *
+ * - `range`: `[start, end)` in Python code points; slice with `Array.from(text).slice(start, end).join('')`, not
+ *   `text.slice` (characters outside the BMP count as 1)
+ */
+export type FineRef = {
+    $ref: string;
+    range?: [number, number] | null;
+};
+/** Layer of an item; only `body` is document content. */
+export type ContentLayer = 'body' | 'furniture' | 'background' | 'invisible' | 'notes';
+/** Width and height in points (pages) or pixels (images). */
+export type Size = {
+    width: number;
+    height: number;
+};
+/** Rectangle in PDF points; `coord_origin` differs per item (`BOTTOMLEFT` for items, `TOPLEFT` for table cells). */
+export type BoundingBox = {
+    l: number;
+    t: number;
+    r: number;
+    b: number;
+    coord_origin: 'TOPLEFT' | 'BOTTOMLEFT';
+};
+/** Where an item sits in the PDF; one entry per page an item covers. */
+export type ProvenanceItem = {
+    /** 1-based PDF page. */
+    page_no: number;
+    bbox: BoundingBox;
+    /** `[start, end)` span within the item's text, in Python code points: slice via `Array.from(text)`, not `text.slice`. */
+    charspan: [number, number];
+};
+/** Image data; with `images: 'embedded'` `uri` is a `data:<mime>;base64,...` URI. */
+export type ImageRef = {
+    mimetype: string;
+    dpi: number;
+    size: Size;
+    uri: string;
+};
+/** Inline text formatting. */
+export type Formatting = {
+    bold: boolean;
+    italic: boolean;
+    underline: boolean;
+    strikethrough: boolean;
+    script: 'baseline' | 'sub' | 'super';
+};
+/** Media track span (audio/video sources); not used for PDFs. */
+export type TrackSource = {
+    kind: 'track';
+    start_time: number;
+    end_time: number;
+    identifier?: string | null;
+    voice?: string | null;
+};
+/** Value set by an enrichment model, with its confidence and creator. */
+type MetaField = {
+    confidence?: number | null;
+    created_by?: string | null;
+};
+/** Enrichment metadata of any item. */
+export type BaseMeta = {
+    summary?: (MetaField & {
+        text: string;
+    }) | null;
+    /** ISO 639-1 code, e.g. `'de'`. */
+    language?: (MetaField & {
+        code: string;
+    }) | null;
+    entities?: {
+        mentions: (MetaField & {
+            text: string;
+            orig?: string | null;
+            label?: string | null;
+            /** `[start, end)` in Python code points: slice via `Array.from(text)`, not `text.slice`. */
+            charspan?: [number, number] | null;
+        })[];
+    } | null;
+    keywords?: {
+        values: string[];
+    } | null;
+    topics?: {
+        values: string[];
+    } | null;
+};
+/** Enrichment metadata of tables, code and other floating items. */
+export type FloatingMeta = BaseMeta & {
+    description?: (MetaField & {
+        text: string;
+    }) | null;
+};
+/** Enrichment metadata of pictures. */
+export type PictureMeta = FloatingMeta & {
+    classification?: {
+        predictions: (MetaField & {
+            class_name: string;
+        })[];
+    } | null;
+    molecule?: (MetaField & {
+        smi: string;
+    }) | null;
+    tabular_chart?: (MetaField & {
+        title?: string | null;
+        chart_data: TableData;
+    }) | null;
+    code?: (MetaField & {
+        text: string;
+        language?: string | null;
+    }) | null;
+};
+/** Fields every item has. */
+type NodeItem = {
+    /** Pointer to this item, e.g. `'#/texts/12'`. */
+    self_ref: string;
+    /** `null` for `body` and `furniture`. */
+    parent: RefItem | null;
+    children: RefItem[];
+    content_layer: ContentLayer;
+};
+/** Fields of items placed on a page. */
+type DocItem = NodeItem & {
+    prov: ProvenanceItem[];
+    /** Not sent by every docling version. */
+    source?: TrackSource[];
+    /** Not sent by every docling version. */
+    comments?: FineRef[];
+};
+/** Fields of items with captions, references and footnotes. */
+type FloatingFields = {
+    /** Caption texts, also listed in `children`. */
+    captions: RefItem[];
+    references: RefItem[];
+    footnotes: RefItem[];
+    /** Rendered crop of the item; `null` unless requested. */
+    image: ImageRef | null;
+};
+/** Docling's group kinds. */
+export type GroupLabel = 'unspecified' | 'list' | 'ordered_list' | 'chapter' | 'section' | 'sheet' | 'slide' | 'form_area' | 'key_value_area' | 'comment_section' | 'inline' | 'picture_area';
+/** Container without own content, e.g. `body`, `furniture` or a key-value area. */
+export type GroupItem = NodeItem & {
+    meta: BaseMeta | null;
+    name: string;
+    label: GroupLabel;
+};
+/** List whose `children` are `ListItem`s; a nested list is a `ListGroup` child of a `ListItem`. */
+export type ListGroup = GroupItem & {
+    label: 'list';
+};
+/** Run of texts that belong to one line or paragraph, e.g. with mixed formatting. */
+export type InlineGroup = GroupItem & {
+    label: 'inline';
+};
+/** Labels of plain `TextItem`s; other text labels have their own item type. */
+export type TextLabel = 'caption' | 'checkbox_selected' | 'checkbox_unselected' | 'footnote' | 'page_footer' | 'page_header' | 'paragraph' | 'reference' | 'text' | 'empty_value' | 'field_key' | 'field_hint' | 'marker' | 'handwritten_text';
+/**
+ * **Fields of every text item**.
+ *
+ * - `orig`: text as read from the PDF; `text`: after docling's processing (formulas: LaTeX or empty)
+ */
+type TextFields = DocItem & {
+    meta: BaseMeta | null;
+    orig: string;
+    text: string;
+    formatting: Formatting | null;
+    hyperlink: string | null;
+};
+/** Paragraph, caption, footnote, page header or footer and other plain text. */
+export type TextItem = TextFields & {
+    label: TextLabel;
+};
+/** Document title. */
+export type TitleItem = TextFields & {
+    label: 'title';
+};
+/** Heading; `level` starts at 1. */
+export type SectionHeaderItem = TextFields & {
+    label: 'section_header';
+    level: number;
+};
+/** List entry; `marker` as printed (`'(1)'`, `'-'`, often `''`), `text` may still start with a bullet glyph. */
+export type ListItem = TextFields & {
+    label: 'list_item';
+    enumerated: boolean;
+    marker: string;
+};
+/** Code listing; `code_language` is `'unknown'` without code enrichment. */
+export type CodeItem = Omit<TextFields, 'meta'> & FloatingFields & {
+    label: 'code';
+    meta: FloatingMeta | null;
+    code_language: string;
+};
+/** Formula; `text` is LaTeX with formula enrichment, otherwise empty with the raw characters in `orig`. */
+export type FormulaItem = TextFields & {
+    label: 'formula';
+};
+/** Heading of a form field. */
+export type FieldHeadingItem = TextFields & {
+    label: 'field_heading';
+    level: number;
+};
+/** Value of a form field. */
+export type FieldValueItem = TextFields & {
+    label: 'field_value';
+    kind: 'read_only' | 'fillable';
+};
+/** Every item of `texts`, told apart by `label`. */
+export type AnyTextItem = TitleItem | SectionHeaderItem | ListItem | CodeItem | FormulaItem | FieldHeadingItem | FieldValueItem | TextItem;
+/** Table cell; offsets are 0-based and end-exclusive, spans repeat the cell in `grid`. */
+export type TableCell = {
+    bbox: BoundingBox | null;
+    row_span: number;
+    col_span: number;
+    start_row_offset_idx: number;
+    end_row_offset_idx: number;
+    start_col_offset_idx: number;
+    end_col_offset_idx: number;
+    text: string;
+    column_header: boolean;
+    row_header: boolean;
+    row_section: boolean;
+    fillable: boolean;
+    /** Set when the cell holds other items (rich cell); its `text` is then a summary. */
+    ref?: RefItem;
+};
+/** Cells of a table; `grid` is the same cells as rows x columns. */
+export type TableData = {
+    table_cells: TableCell[];
+    num_rows: number;
+    num_cols: number;
+    orientation: 'rot_0' | 'rot_90' | 'rot_180' | 'rot_270';
+    /** Computed by docling on export; `num_rows` arrays of `num_cols` cells. */
+    grid: TableCell[][];
+};
+/** Free-text description of a picture or table by a model. */
+export type DescriptionAnnotation = {
+    kind: 'description';
+    text: string;
+    provenance: string;
+};
+/** Annotation without a fixed shape. */
+export type MiscAnnotation = {
+    kind: 'misc';
+    content: Record<string, unknown>;
+};
+/** Table or table of contents (`document_index`). */
+export type TableItem = DocItem & FloatingFields & {
+    label: 'table' | 'document_index';
+    meta: FloatingMeta | null;
+    data: TableData;
+    annotations: (DescriptionAnnotation | MiscAnnotation)[];
+};
+/**
+ * **Model output on a picture**: description, classification, molecule or chart data.
+ *
+ * - chart kinds are filled with `chartExtraction`
+ */
+export type PictureAnnotation = DescriptionAnnotation | MiscAnnotation | {
+    kind: 'classification';
+    provenance: string;
+    predicted_classes: {
+        class_name: string;
+        confidence: number;
+    }[];
+} | {
+    kind: 'molecule_data';
+    smi: string;
+    confidence: number;
+    class_name: string;
+    segmentation: [number, number][];
+    provenance: string;
+} | {
+    kind: 'tabular_chart_data';
+    title: string;
+    chart_data: TableData;
+} | (ChartAxes<'line_chart_data'> & {
+    lines: {
+        label: string;
+        values: [number, number][];
+    }[];
+}) | (ChartAxes<'bar_chart_data'> & {
+    bars: {
+        label: string;
+        values: number;
+    }[];
+}) | (ChartAxes<'stacked_bar_chart_data'> & {
+    stacked_bars: {
+        label: string[];
+        values: [string, number][];
+    }[];
+}) | {
+    kind: 'pie_chart_data';
+    title: string;
+    slices: {
+        label: string;
+        value: number;
+    }[];
+} | (ChartAxes<'scatter_chart_data'> & {
+    points: {
+        value: [number, number];
+    }[];
+});
+/** Title and axis labels of a chart annotation. */
+type ChartAxes<K extends string> = {
+    kind: K;
+    title: string;
+    x_axis_label: string;
+    y_axis_label: string;
+};
+/**
+ * **Picture or chart**.
+ *
+ * - `image`: data URI with `images: 'embedded'`, `null` with `'placeholder'`
+ * - `children`: captions and texts docling found inside the picture
+ */
+export type PictureItem = DocItem & FloatingFields & {
+    label: 'picture' | 'chart';
+    meta: PictureMeta | null;
+    annotations: PictureAnnotation[];
+};
+/** Cell of a key-value or form graph. */
+export type GraphCell = {
+    label: 'unspecified' | 'key' | 'value' | 'checkbox';
+    cell_id: number;
+    text: string;
+    orig: string;
+    prov: ProvenanceItem | null;
+    item_ref: RefItem | null;
+};
+/** Link between two graph cells. */
+export type GraphLink = {
+    label: 'unspecified' | 'to_value' | 'to_key' | 'to_parent' | 'to_child';
+    source_cell_id: number;
+    target_cell_id: number;
+};
+/** Key-value pairs as a graph of cells and links. */
+export type GraphData = {
+    cells: GraphCell[];
+    links: GraphLink[];
+};
+/** Region of key-value pairs. */
+export type KeyValueItem = DocItem & FloatingFields & {
+    label: 'key_value_region';
+    meta: FloatingMeta | null;
+    graph: GraphData;
+};
+/** Form region. */
+export type FormItem = DocItem & FloatingFields & {
+    label: 'form';
+    meta: FloatingMeta | null;
+    graph: GraphData;
+};
+/** Region holding form fields. */
+export type FieldRegionItem = DocItem & {
+    label: 'field_region';
+    meta: BaseMeta | null;
+};
+/** One form field. */
+export type FieldItem = DocItem & {
+    label: 'field_item';
+    meta: BaseMeta | null;
+};
+export {};
+//# sourceMappingURL=document.d.ts.map

package/dist/document.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"document.d.ts","sourceRoot":"","sources":["../src/document.ts"],"names":[],"mappings":"AASA;;;;;;GAMG;AACH,MAAM,MAAM,eAAe,GAAG;IAC1B,WAAW,EAAE,iBAAiB,CAAC;IAC/B,uCAAuC;IACvC,OAAO,EAAE,MAAM,CAAC;IAChB,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,EAAE,cAAc,GAAG,IAAI,CAAC;IAC9B,wCAAwC;IACxC,SAAS,EAAE,SAAS,CAAC;IACrB,4CAA4C;IAC5C,IAAI,EAAE,SAAS,CAAC;IAChB,MAAM,EAAE,CAAC,SAAS,GAAG,WAAW,GAAG,SAAS,CAAC,EAAE,CAAC;IAChD,KAAK,EAAE,WAAW,EAAE,CAAC;IACrB,QAAQ,EAAE,WAAW,EAAE,CAAC;IACxB,MAAM,EAAE,SAAS,EAAE,CAAC;IACpB,eAAe,EAAE,YAAY,EAAE,CAAC;IAChC,UAAU,EAAE,QAAQ,EAAE,CAAC;IACvB,yCAAyC;IACzC,aAAa,CAAC,EAAE,eAAe,EAAE,CAAC;IAClC,yCAAyC;IACzC,WAAW,CAAC,EAAE,SAAS,EAAE,CAAC;IAC1B,qEAAqE;IACrE,KAAK,EAAE,MAAM,CAAC,MAAM,EAAE,QAAQ,CAAC,CAAC;CACnC,CAAC;AAEF,mCAAmC;AACnC,MAAM,MAAM,cAAc,GAAG;IACzB,QAAQ,EAAE,MAAM,CAAC;IACjB,WAAW,EAAE,MAAM,CAAC;IACpB,QAAQ,EAAE,MAAM,CAAC;IACjB,GAAG,CAAC,EAAE,MAAM,GAAG,IAAI,CAAC;CACvB,CAAC;AAEF,oBAAoB;AACpB,MAAM,MAAM,QAAQ,GAAG;IACnB,IAAI,EAAE,IAAI,CAAC;IACX,mGAAmG;IACnG,KAAK,EAAE,QAAQ,GAAG,IAAI,CAAC;IACvB,OAAO,EAAE,MAAM,CAAC;CACnB,CAAC;AAMF,2FAA2F;AAC3F,MAAM,MAAM,OAAO,GAAG;IAAE,IAAI,EAAE,MAAM,CAAA;CAAE,CAAC;AAEvC;;;;;GAKG;AACH,MAAM,MAAM,OAAO,GAAG;IAAE,IAAI,EAAE,MAAM,CAAC;IAAC,KAAK,CAAC,EAAE,CAAC,MAAM,EAAE,MAAM,CAAC,GAAG,IAAI,CAAA;CAAE,CAAC;AAExE,yDAAyD;AACzD,MAAM,MAAM,YAAY,GAAG,MAAM,GAAG,WAAW,GAAG,YAAY,GAAG,WAAW,GAAG,OAAO,CAAC;AAEvF,6DAA6D;AAC7D,MAAM,MAAM,IAAI,GAAG;IAAE,KAAK,EAAE,MAAM,CAAC;IAAC,MAAM,EAAE,MAAM,CAAA;CAAE,CAAC;AAErD,oHAAoH;AACpH,MAAM,MAAM,WAAW,GAAG;IACtB,CAAC,EAAE,MAAM,CAAC;IACV,CAAC,EAAE,MAAM,CAAC;IACV,CAAC,EAAE,MAAM,CAAC;IACV,CAAC,EAAE,MAAM,CAAC;IACV,YAAY,EAAE,SAAS,GAAG,YAAY,CAAC;CAC1C,CAAC;AAEF,wEAAwE;AACxE,MAAM,MAAM,cAAc,GAAG;IACzB,wBAAwB;IACxB,OAAO,EAAE,MAAM,CAAC;IAChB,IAAI,EAAE,WAAW,CAAC;IAClB,yHAAyH;IACzH,QAAQ,EAAE,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CAC9B,CAAC;AAEF,qFAAqF;AACrF,MAAM,MAAM,QAAQ,GAAG;IACnB,QAAQ,EAAE,MAAM,CAAC;IACjB,GAAG,EAAE,MAAM,CAAC;IACZ,IAAI,EAAE,IAAI,CAAC;IACX,GAAG,EAAE,MAAM,CAAC;CACf,CAAC;AAEF,8BAA8B;AAC9B,MAAM,MAAM,UAAU,GAAG;IACrB,IAAI,EAAE,OAAO,CAAC;IACd,MAAM,EAAE,OAAO,CAAC;IAChB,SAAS,EAAE,OAAO,CAAC;IACnB,aAAa,EAAE,OAAO,CAAC;IACvB,MAAM,EAAE,UAAU,GAAG,KAAK,GAAG,OAAO,CAAC;CACxC,CAAC;AAEF,iEAAiE;AACjE,MAAM,MAAM,WAAW,GAAG;IACtB,IAAI,EAAE,OAAO,CAAC;IACd,UAAU,EAAE,MAAM,CAAC;IACnB,QAAQ,EAAE,MAAM,CAAC;IACjB,UAAU,CAAC,EAAE,MAAM,GAAG,IAAI,CAAC;IAC3B,KAAK,CAAC,EAAE,MAAM,GAAG,IAAI,CAAC;CACzB,CAAC;AAEF,yEAAyE;AACzE,KAAK,SAAS,GAAG;IAAE,UAAU,CAAC,EAAE,MAAM,GAAG,IAAI,CAAC;IAAC,UAAU,CAAC,EAAE,MAAM,GAAG,IAAI,CAAA;CAAE,CAAC;AAE5E,uCAAuC;AACvC,MAAM,MAAM,QAAQ,GAAG;IACnB,OAAO,CAAC,EAAE,CAAC,SAAS,GAAG;QAAE,IAAI,EAAE,MAAM,CAAA;KAAE,CAAC,GAAG,IAAI,CAAC;IAChD,mCAAmC;IACnC,QAAQ,CAAC,EAAE,CAAC,SAAS,GAAG;QAAE,IAAI,EAAE,MAAM,CAAA;KAAE,CAAC,GAAG,IAAI,CAAC;IACjD,QAAQ,CAAC,EAAE;QACP,QAAQ,EAAE,CAAC,SAAS,GAAG;YACnB,IAAI,EAAE,MAAM,CAAC;YACb,IAAI,CAAC,EAAE,MAAM,GAAG,IAAI,CAAC;YACrB,KAAK,CAAC,EAAE,MAAM,GAAG,IAAI,CAAC;YACtB,4FAA4F;YAC5F,QAAQ,CAAC,EAAE,CAAC,MAAM,EAAE,MAAM,CAAC,GAAG,IAAI,CAAC;SACtC,CAAC,EAAE,CAAC;KACR,GAAG,IAAI,CAAC;IACT,QAAQ,CAAC,EAAE;QAAE,MAAM,EAAE,MAAM,EAAE,CAAA;KAAE,GAAG,IAAI,CAAC;IACvC,MAAM,CAAC,EAAE;QAAE,MAAM,EAAE,MAAM,EAAE,CAAA;KAAE,GAAG,IAAI,CAAC;CACxC,CAAC;AAEF,oEAAoE;AACpE,MAAM,MAAM,YAAY,GAAG,QAAQ,GAAG;IAClC,WAAW,CAAC,EAAE,CAAC,SAAS,GAAG;QAAE,IAAI,EAAE,MAAM,CAAA;KAAE,CAAC,GAAG,IAAI,CAAC;CACvD,CAAC;AAEF,uCAAuC;AACvC,MAAM,MAAM,WAAW,GAAG,YAAY,GAAG;IACrC,cAAc,CAAC,EAAE;QAAE,WAAW,EAAE,CAAC,SAAS,GAAG;YAAE,UAAU,EAAE,MAAM,CAAA;SAAE,CAAC,EAAE,CAAA;KAAE,GAAG,IAAI,CAAC;IAChF,QAAQ,CAAC,EAAE,CAAC,SAAS,GAAG;QAAE,GAAG,EAAE,MAAM,CAAA;KAAE,CAAC,GAAG,IAAI,CAAC;IAChD,aAAa,CAAC,EAAE,CAAC,SAAS,GAAG;QAAE,KAAK,CAAC,EAAE,MAAM,GAAG,IAAI,CAAC;QAAC,UAAU,EAAE,SAAS,CAAA;KAAE,CAAC,GAAG,IAAI,CAAC;IACtF,IAAI,CAAC,EAAE,CAAC,SAAS,GAAG;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,QAAQ,CAAC,EAAE,MAAM,GAAG,IAAI,CAAA;KAAE,CAAC,GAAG,IAAI,CAAC;CAC1E,CAAC;AAEF,6BAA6B;AAC7B,KAAK,QAAQ,GAAG;IACZ,iDAAiD;IACjD,QAAQ,EAAE,MAAM,CAAC;IACjB,yCAAyC;IACzC,MAAM,EAAE,OAAO,GAAG,IAAI,CAAC;IACvB,QAAQ,EAAE,OAAO,EAAE,CAAC;IACpB,aAAa,EAAE,YAAY,CAAC;CAC/B,CAAC;AAEF,wCAAwC;AACxC,KAAK,OAAO,GAAG,QAAQ,GAAG;IACtB,IAAI,EAAE,cAAc,EAAE,CAAC;IACvB,yCAAyC;IACzC,MAAM,CAAC,EAAE,WAAW,EAAE,CAAC;IACvB,yCAAyC;IACzC,QAAQ,CAAC,EAAE,OAAO,EAAE,CAAC;CACxB,CAAC;AAEF,+DAA+D;AAC/D,KAAK,cAAc,GAAG;IAClB,gDAAgD;IAChD,QAAQ,EAAE,OAAO,EAAE,CAAC;IACpB,UAAU,EAAE,OAAO,EAAE,CAAC;IACtB,SAAS,EAAE,OAAO,EAAE,CAAC;IACrB,0DAA0D;IAC1D,KAAK,EAAE,QAAQ,GAAG,IAAI,CAAC;CAC1B,CAAC;AAMF,6BAA6B;AAC7B,MAAM,MAAM,UAAU,GAChB,aAAa,GACb,MAAM,GACN,cAAc,GACd,SAAS,GACT,SAAS,GACT,OAAO,GACP,OAAO,GACP,WAAW,GACX,gBAAgB,GAChB,iBAAiB,GACjB,QAAQ,GACR,cAAc,CAAC;AAErB,mFAAmF;AACnF,MAAM,MAAM,SAAS,GAAG,QAAQ,GAAG;IAC/B,IAAI,EAAE,QAAQ,GAAG,IAAI,CAAC;IACtB,IAAI,EAAE,MAAM,CAAC;IACb,KAAK,EAAE,UAAU,CAAC;CACrB,CAAC;AAEF,mGAAmG;AACnG,MAAM,MAAM,SAAS,GAAG,SAAS,GAAG;IAAE,KAAK,EAAE,MAAM,CAAA;CAAE,CAAC;AAEtD,qFAAqF;AACrF,MAAM,MAAM,WAAW,GAAG,SAAS,GAAG;IAAE,KAAK,EAAE,QAAQ,CAAA;CAAE,CAAC;AAM1D,+EAA+E;AAC/E,MAAM,MAAM,SAAS,GACf,SAAS,GACT,mBAAmB,GACnB,qBAAqB,GACrB,UAAU,GACV,aAAa,GACb,aAAa,GACb,WAAW,GACX,WAAW,GACX,MAAM,GACN,aAAa,GACb,WAAW,GACX,YAAY,GACZ,QAAQ,GACR,kBAAkB,CAAC;AAEzB;;;;GAIG;AACH,KAAK,UAAU,GAAG,OAAO,GAAG;IACxB,IAAI,EAAE,QAAQ,GAAG,IAAI,CAAC;IACtB,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IACb,UAAU,EAAE,UAAU,GAAG,IAAI,CAAC;IAC9B,SAAS,EAAE,MAAM,GAAG,IAAI,CAAC;CAC5B,CAAC;AAEF,gFAAgF;AAChF,MAAM,MAAM,QAAQ,GAAG,UAAU,GAAG;IAAE,KAAK,EAAE,SAAS,CAAA;CAAE,CAAC;AAEzD,sBAAsB;AACtB,MAAM,MAAM,SAAS,GAAG,UAAU,GAAG;IAAE,KAAK,EAAE,OAAO,CAAA;CAAE,CAAC;AAExD,oCAAoC;AACpC,MAAM,MAAM,iBAAiB,GAAG,UAAU,GAAG;IAAE,KAAK,EAAE,gBAAgB,CAAC;IAAC,KAAK,EAAE,MAAM,CAAA;CAAE,CAAC;AAExF,gHAAgH;AAChH,MAAM,MAAM,QAAQ,GAAG,UAAU,GAAG;IAAE,KAAK,EAAE,WAAW,CAAC;IAAC,UAAU,EAAE,OAAO,CAAC;IAAC,MAAM,EAAE,MAAM,CAAA;CAAE,CAAC;AAEhG,4EAA4E;AAC5E,MAAM,MAAM,QAAQ,GAAG,IAAI,CAAC,UAAU,EAAE,MAAM,CAAC,GAC3C,cAAc,GAAG;IAAE,KAAK,EAAE,MAAM,CAAC;IAAC,IAAI,EAAE,YAAY,GAAG,IAAI,CAAC;IAAC,aAAa,EAAE,MAAM,CAAA;CAAE,CAAC;AAEzF,2GAA2G;AAC3G,MAAM,MAAM,WAAW,GAAG,UAAU,GAAG;IAAE,KAAK,EAAE,SAAS,CAAA;CAAE,CAAC;AAE5D,+BAA+B;AAC/B,MAAM,MAAM,gBAAgB,GAAG,UAAU,GAAG;IAAE,KAAK,EAAE,eAAe,CAAC;IAAC,KAAK,EAAE,MAAM,CAAA;CAAE,CAAC;AAEtF,6BAA6B;AAC7B,MAAM,MAAM,cAAc,GAAG,UAAU,GAAG;IAAE,KAAK,EAAE,aAAa,CAAC;IAAC,IAAI,EAAE,WAAW,GAAG,UAAU,CAAA;CAAE,CAAC;AAEnG,oDAAoD;AACpD,MAAM,MAAM,WAAW,GACjB,SAAS,GACT,iBAAiB,GACjB,QAAQ,GACR,QAAQ,GACR,WAAW,GACX,gBAAgB,GAChB,cAAc,GACd,QAAQ,CAAC;AAMf,0FAA0F;AAC1F,MAAM,MAAM,SAAS,GAAG;IACpB,IAAI,EAAE,WAAW,GAAG,IAAI,CAAC;IACzB,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,MAAM,CAAC;IACjB,oBAAoB,EAAE,MAAM,CAAC;IAC7B,kBAAkB,EAAE,MAAM,CAAC;IAC3B,oBAAoB,EAAE,MAAM,CAAC;IAC7B,kBAAkB,EAAE,MAAM,CAAC;IAC3B,IAAI,EAAE,MAAM,CAAC;IACb,aAAa,EAAE,OAAO,CAAC;IACvB,UAAU,EAAE,OAAO,CAAC;IACpB,WAAW,EAAE,OAAO,CAAC;IACrB,QAAQ,EAAE,OAAO,CAAC;IAClB,qFAAqF;IACrF,GAAG,CAAC,EAAE,OAAO,CAAC;CACjB,CAAC;AAEF,oEAAoE;AACpE,MAAM,MAAM,SAAS,GAAG;IACpB,WAAW,EAAE,SAAS,EAAE,CAAC;IACzB,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,MAAM,CAAC;IACjB,WAAW,EAAE,OAAO,GAAG,QAAQ,GAAG,SAAS,GAAG,SAAS,CAAC;IACxD,4EAA4E;IAC5E,IAAI,EAAE,SAAS,EAAE,EAAE,CAAC;CACvB,CAAC;AAEF,8DAA8D;AAC9D,MAAM,MAAM,qBAAqB,GAAG;IAAE,IAAI,EAAE,aAAa,CAAC;IAAC,IAAI,EAAE,MAAM,CAAC;IAAC,UAAU,EAAE,MAAM,CAAA;CAAE,CAAC;AAE9F,wCAAwC;AACxC,MAAM,MAAM,cAAc,GAAG;IAAE,IAAI,EAAE,MAAM,CAAC;IAAC,OAAO,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAA;CAAE,CAAC;AAEhF,qDAAqD;AACrD,MAAM,MAAM,SAAS,GAAG,OAAO,GAC3B,cAAc,GAAG;IACb,KAAK,EAAE,OAAO,GAAG,gBAAgB,CAAC;IAClC,IAAI,EAAE,YAAY,GAAG,IAAI,CAAC;IAC1B,IAAI,EAAE,SAAS,CAAC;IAChB,WAAW,EAAE,CAAC,qBAAqB,GAAG,cAAc,CAAC,EAAE,CAAC;CAC3D,CAAC;AAMN;;;;GAIG;AACH,MAAM,MAAM,iBAAiB,GACvB,qBAAqB,GACrB,cAAc,GACd;IACI,IAAI,EAAE,gBAAgB,CAAC;IACvB,UAAU,EAAE,MAAM,CAAC;IACnB,iBAAiB,EAAE;QAAE,UAAU,EAAE,MAAM,CAAC;QAAC,UAAU,EAAE,MAAM,CAAA;KAAE,EAAE,CAAC;CACnE,GACD;IACI,IAAI,EAAE,eAAe,CAAC;IACtB,GAAG,EAAE,MAAM,CAAC;IACZ,UAAU,EAAE,MAAM,CAAC;IACnB,UAAU,EAAE,MAAM,CAAC;IACnB,YAAY,EAAE,CAAC,MAAM,EAAE,MAAM,CAAC,EAAE,CAAC;IACjC,UAAU,EAAE,MAAM,CAAC;CACtB,GACD;IAAE,IAAI,EAAE,oBAAoB,CAAC;IAAC,KAAK,EAAE,MAAM,CAAC;IAAC,UAAU,EAAE,SAAS,CAAA;CAAE,GACpE,CAAC,SAAS,CAAC,iBAAiB,CAAC,GAAG;IAAE,KAAK,EAAE;QAAE,KAAK,EAAE,MAAM,CAAC;QAAC,MAAM,EAAE,CAAC,MAAM,EAAE,MAAM,CAAC,EAAE,CAAA;KAAE,EAAE,CAAA;CAAE,CAAC,GAC3F,CAAC,SAAS,CAAC,gBAAgB,CAAC,GAAG;IAAE,IAAI,EAAE;QAAE,KAAK,EAAE,MAAM,CAAC;QAAC,MAAM,EAAE,MAAM,CAAA;KAAE,EAAE,CAAA;CAAE,CAAC,GAC7E,CAAC,SAAS,CAAC,wBAAwB,CAAC,GAAG;IAAE,YAAY,EAAE;QAAE,KAAK,EAAE,MAAM,EAAE,CAAC;QAAC,MAAM,EAAE,CAAC,MAAM,EAAE,MAAM,CAAC,EAAE,CAAA;KAAE,EAAE,CAAA;CAAE,CAAC,GAC3G;IAAE,IAAI,EAAE,gBAAgB,CAAC;IAAC,KAAK,EAAE,MAAM,CAAC;IAAC,MAAM,EAAE;QAAE,KAAK,EAAE,MAAM,CAAC;QAAC,KAAK,EAAE,MAAM,CAAA;KAAE,EAAE,CAAA;CAAE,GACrF,CAAC,SAAS,CAAC,oBAAoB,CAAC,GAAG;IAAE,MAAM,EAAE;QAAE,KAAK,EAAE,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;KAAE,EAAE,CAAA;CAAE,CAAC,CAAC;AAEpF,mDAAmD;AACnD,KAAK,SAAS,CAAC,CAAC,SAAS,MAAM,IAAI;IAAE,IAAI,EAAE,CAAC,CAAC;IAAC,KAAK,EAAE,MAAM,CAAC;IAAC,YAAY,EAAE,MAAM,CAAC;IAAC,YAAY,EAAE,MAAM,CAAA;CAAE,CAAC;AAE1G;;;;;GAKG;AACH,MAAM,MAAM,WAAW,GAAG,OAAO,GAC7B,cAAc,GAAG;IACb,KAAK,EAAE,SAAS,GAAG,OAAO,CAAC;IAC3B,IAAI,EAAE,WAAW,GAAG,IAAI,CAAC;IACzB,WAAW,EAAE,iBAAiB,EAAE,CAAC;CACpC,CAAC;AAMN,yCAAyC;AACzC,MAAM,MAAM,SAAS,GAAG;IACpB,KAAK,EAAE,aAAa,GAAG,KAAK,GAAG,OAAO,GAAG,UAAU,CAAC;IACpD,OAAO,EAAE,MAAM,CAAC;IAChB,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,cAAc,GAAG,IAAI,CAAC;IAC5B,QAAQ,EAAE,OAAO,GAAG,IAAI,CAAC;CAC5B,CAAC;AAEF,oCAAoC;AACpC,MAAM,MAAM,SAAS,GAAG;IACpB,KAAK,EAAE,aAAa,GAAG,UAAU,GAAG,QAAQ,GAAG,WAAW,GAAG,UAAU,CAAC;IACxE,cAAc,EAAE,MAAM,CAAC;IACvB,cAAc,EAAE,MAAM,CAAC;CAC1B,CAAC;AAEF,qDAAqD;AACrD,MAAM,MAAM,SAAS,GAAG;IAAE,KAAK,EAAE,SAAS,EAAE,CAAC;IAAC,KAAK,EAAE,SAAS,EAAE,CAAA;CAAE,CAAC;AAEnE,iCAAiC;AACjC,MAAM,MAAM,YAAY,GAAG,OAAO,GAC9B,cAAc,GAAG;IAAE,KAAK,EAAE,kBAAkB,CAAC;IAAC,IAAI,EAAE,YAAY,GAAG,IAAI,CAAC;IAAC,KAAK,EAAE,SAAS,CAAA;CAAE,CAAC;AAEhG,mBAAmB;AACnB,MAAM,MAAM,QAAQ,GAAG,OAAO,GAAG,cAAc,GAAG;IAAE,KAAK,EAAE,MAAM,CAAC;IAAC,IAAI,EAAE,YAAY,GAAG,IAAI,CAAC;IAAC,KAAK,EAAE,SAAS,CAAA;CAAE,CAAC;AAEjH,kCAAkC;AAClC,MAAM,MAAM,eAAe,GAAG,OAAO,GAAG;IAAE,KAAK,EAAE,cAAc,CAAC;IAAC,IAAI,EAAE,QAAQ,GAAG,IAAI,CAAA;CAAE,CAAC;AAEzF,sBAAsB;AACtB,MAAM,MAAM,SAAS,GAAG,OAAO,GAAG;IAAE,KAAK,EAAE,YAAY,CAAC;IAAC,IAAI,EAAE,QAAQ,GAAG,IAAI,CAAA;CAAE,CAAC"}

package/dist/images.d.ts

@@ -0,0 +1,67 @@
+import type { DoclingDocument, ProvenanceItem } from './document.js';
+import type { ExtractedImage, ImageOptions, ImageSize, ImageWarning, PageImage, PageImageBox, PageImageWarning } from './types.js';
+/** Rewritten md, extracted image files, and `image_not_decoded` warnings. */
+export type ExtractedMarkdown = {
+    markdown: string;
+    images: ExtractedImage[];
+    warnings: ImageWarning[];
+};
+/** Document with decoded `pages[].image` set to `null`, the decoded page images and `page_image_not_decoded` warnings. */
+export type ExtractedPageImages = {
+    document: DoclingDocument;
+    pageImages: PageImage[];
+    warnings: PageImageWarning[];
+};
+/**
+ * **Pixels per PDF point of pictures and page images**: `images_scale` (144 dpi).
+ *
+ * - fixed, so page images and `toPageImageBox` always agree
+ * - docling-serve's default and its maximum (`max_images_scale`); OCR runs at its own scale and is not affected
+ */
+export declare const PAGE_IMAGE_SCALE = 2;
+/**
+ * **Size heuristic behind `omitSmall`**: icons, bullets, small logos and thin rules count as small.
+ *
+ * - small: area below 15 000 px, longer side below 200 px, or shorter side below 24 px
+ * - works on `ExtractedImage` and on `pictures[].image.size` of a DoclingDocument
+ * - `false` when width or height is not a finite number
+ */
+export declare function isSmallImage(size: ImageSize): boolean;
+/** Reads pixel size from PNG IHDR chunk; returns `undefined` for non-PNG or truncated data. */
+export declare function readPngSize(bytes: Uint8Array): ImageSize | undefined;
+/**
+ * **Extracts base64 images into raw bytes and replaces them with file links**.
+ *
+ * - links: `<prefix><n>.<ext>`, numbered over kept images only; extension mapped from the MIME subtype
+ * - small images (with `omitSmall`) and undecodable images are removed like placeholders
+ * - undecodable image: `image_not_decoded` warning with its 1-based index among all embedded images
+ * - docling's own `<!-- image -->` placeholders (pictures sent without data) are kept
+ */
+export declare function extractImages(markdown: string, options: Required<ImageOptions>): ExtractedMarkdown;
+/**
+ * **Moves page images out of the DoclingDocument into raw bytes**.
+ *
+ * - every decoded `pages[].image` becomes `null`; the input document is not changed
+ * - page number from `page_no`, else from the `pages` key; result sorted by page
+ * - a URI that is not a base64 image data URI, or has invalid or empty data: `page_image_not_decoded` warning,
+ *   and the page keeps its `image` as sent
+ * - a document without a `pages` object is returned as it is
+ */
+export declare function extractPageImages(document: DoclingDocument): ExtractedPageImages;
+/**
+ * **Maps a docling bounding box onto its page image, in pixels with a top-left origin**.
+ *
+ * - `prov`: any `prov[]` entry; for a table cell pass `{ page_no: table.prov[0].page_no, bbox: cell.bbox }`
+ * - flips `BOTTOMLEFT` boxes using the page height from `document.pages`, then scales points by `PAGE_IMAGE_SCALE`
+ * - works without `pageImages: true`; the boxes fit the page images of any conversion by this client
+ * - returns `undefined` when page_no or a coordinate is not a finite number, or page is missing from `document.pages`
+ */
+export declare function toPageImageBox(document: DoclingDocument, prov: Pick<ProvenanceItem, 'page_no' | 'bbox'>): PageImageBox | undefined;
+/**
+ * **Removes `<!-- image -->` placeholders**.
+ *
+ * - a placeholder line goes; blank lines separating the text around it stay, without leaving a blank-line run
+ * - a placeholder inside a line is cut out of it
+ */
+export declare function removeImagePlaceholders(markdown: string): string;
+//# sourceMappingURL=images.d.ts.map

package/dist/images.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"images.d.ts","sourceRoot":"","sources":["../src/images.ts"],"names":[],"mappings":"AAIA,OAAO,KAAK,EAAE,eAAe,EAAY,cAAc,EAAE,MAAM,eAAe,CAAC;AAC/E,OAAO,KAAK,EACR,cAAc,EACd,YAAY,EACZ,SAAS,EACT,YAAY,EACZ,SAAS,EACT,YAAY,EACZ,gBAAgB,EACnB,MAAM,YAAY,CAAC;AAMpB,6EAA6E;AAC7E,MAAM,MAAM,iBAAiB,GAAG;IAC5B,QAAQ,EAAE,MAAM,CAAC;IACjB,MAAM,EAAE,cAAc,EAAE,CAAC;IACzB,QAAQ,EAAE,YAAY,EAAE,CAAC;CAC5B,CAAC;AAEF,0HAA0H;AAC1H,MAAM,MAAM,mBAAmB,GAAG;IAC9B,QAAQ,EAAE,eAAe,CAAC;IAC1B,UAAU,EAAE,SAAS,EAAE,CAAC;IACxB,QAAQ,EAAE,gBAAgB,EAAE,CAAC;CAChC,CAAC;AAMF;;;;;GAKG;AACH,eAAO,MAAM,gBAAgB,IAAI,CAAC;AAmDlC;;;;;;GAMG;AACH,wBAAgB,YAAY,CAAC,IAAI,EAAE,SAAS,GAAG,OAAO,CAOrD;AAED,+FAA+F;AAC/F,wBAAgB,WAAW,CAAC,KAAK,EAAE,UAAU,GAAG,SAAS,GAAG,SAAS,CAYpE;AAED;;;;;;;GAOG;AACH,wBAAgB,aAAa,CAAC,QAAQ,EAAE,MAAM,EAAE,OAAO,EAAE,QAAQ,CAAC,YAAY,CAAC,GAAG,iBAAiB,CAoClG;AAED;;;;;;;;GAQG;AACH,wBAAgB,iBAAiB,CAAC,QAAQ,EAAE,eAAe,GAAG,mBAAmB,CAgChF;AAED;;;;;;;GAOG;AACH,wBAAgB,cAAc,CAC1B,QAAQ,EAAE,eAAe,EACzB,IAAI,EAAE,IAAI,CAAC,cAAc,EAAE,SAAS,GAAG,MAAM,CAAC,GAC/C,YAAY,GAAG,SAAS,CA2B1B;AAED;;;;;GAKG;AACH,wBAAgB,uBAAuB,CAAC,QAAQ,EAAE,MAAM,GAAG,MAAM,CAEhE"}

package/dist/index.d.ts

@@ -0,0 +1,6 @@
+export type * from './document.js';
+export type { CallOptions, ClientOptions, ConversionFailure, ConvertError, ConvertedDocument, ConvertedMarkdown, ConvertInput, DoclingClient, DoclingConfig, DoclingOptions, DocumentImageMode, DocumentJob, DocumentOptions, DocumentWarning, ErrorCode, ExtractedImage, ImageOptions, ImageSize, ImageWarning, Job, JobState, MarkdownImageMode, MarkdownJob, MarkdownJobOutput, MarkdownOptions, MarkdownWarning, PageImage, PageImageBox, PageImageWarning, PdfBackend, PdfInput, Pipeline, Result, ServerDetails, ServerError, ServerVersions, ServerWarning, TaskFailure, } from './types.js';
+export { PAGE_IMAGE_SCALE } from './images.js';
+export { createDoclingClient } from './client.js';
+export { isSmallImage, readPngSize, toPageImageBox } from './images.js';
+//# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAOA,mBAAmB,eAAe,CAAC;AACnC,YAAY,EACR,WAAW,EACX,aAAa,EACb,iBAAiB,EACjB,YAAY,EACZ,iBAAiB,EACjB,iBAAiB,EACjB,YAAY,EACZ,aAAa,EACb,aAAa,EACb,cAAc,EACd,iBAAiB,EACjB,WAAW,EACX,eAAe,EACf,eAAe,EACf,SAAS,EACT,cAAc,EACd,YAAY,EACZ,SAAS,EACT,YAAY,EACZ,GAAG,EACH,QAAQ,EACR,iBAAiB,EACjB,WAAW,EACX,iBAAiB,EACjB,eAAe,EACf,eAAe,EACf,SAAS,EACT,YAAY,EACZ,gBAAgB,EAChB,UAAU,EACV,QAAQ,EACR,QAAQ,EACR,MAAM,EACN,aAAa,EACb,WAAW,EACX,cAAc,EACd,aAAa,EACb,WAAW,GACd,MAAM,YAAY,CAAC;AAMpB,OAAO,EAAE,gBAAgB,EAAE,MAAM,aAAa,CAAC;AAM/C,OAAO,EAAE,mBAAmB,EAAE,MAAM,aAAa,CAAC;AAClD,OAAO,EAAE,YAAY,EAAE,WAAW,EAAE,cAAc,EAAE,MAAM,aAAa,CAAC"}