Shared `slotsTokenizer` for directive blocks

Abierto
#1,301 0 comentarios 0 reacciones 0 asignados Ver en GitHub

Los mantenedores suelen responder en 1 día

@makhnatkin ya está trabajando en esto.

Desde el 1/10/2026.

  • #1303 de @makhnatkin — abierto

Evaluación

Dificultad
5/5
Tiempo estimado
Más de una semana
Aptitud para principiantes
20/100
Tipo de issue
Nueva funcionalidad
Claridad
Bastante claro
Estado de actividad
Activo
Stack tecnológico
markdown, typescript
Área
frontend

Línea de trabajo

Start from the existing cut extension and container directive registration that call builder.configureMd; trace how each parses its body into Markdown-it tokens and where schema expressions are declared. Implement the shared router behind blockContentSlots and slotsTokenizer, then compare both token streams shown in the issue. Done when cut routes paragraphs/images and header routes actions/content, with preview and WYSIWYG specs accepting the same routed stream.

Escrito por el modelo de indexación a partir del texto del issue.

Descripción

Edited by AI
"Generated" means mostly the model, "Edited" means mostly the author.

Problem

Different blocks split their Markdown body into several parts. Each part limits which content it accepts and how many items it holds. Implemented in each block's own contentTokenizer, this selection and routing logic is duplicated.

Solution

A shared slotsTokenizer function takes rules for named parts (slots) and returns a tokenizer for the existing contentTokenizer field of container directives.

The block declares its slots and their limits. The shared function parses the body, selects content and emits slot wrappers.

Contract

The field in ContainerDirectiveConfig from @diplodoc/directive:

contentTokenizer?: (
    state: StateBlock,
    content: BlockContent,
    params: ContainerDirectiveParams,
) => void;

The directive handler calls it between the block's opening and closing tokens. The function adds the body tokens to state.tokens.

slotsTokenizer returns a function with this signature:

type SlotMatch = (group: readonly Token[]) => boolean;

type SlotRule = {
    slot: string;
    match: SlotMatch;
    limit?: number;
    className?: string;
};

declare function slotsTokenizer(
    name: string,
    rules: readonly SlotRule[],
): NonNullable<ContainerDirectiveConfig['contentTokenizer']>;

declare function byType(...types: string[]): SlotMatch;
  • name — directive name. The body is parsed with parentType <name>-directive, as by the default tokenizer.
  • slot — unique slot name. The wrapper is a <slot>_open / <slot>_close pair with the div tag.
  • match — condition for content the slot accepts.
  • limit — maximum number of groups in the slot per block instance: a positive integer, unlimited by default.
  • className — optional CSS class of the wrapper.
  • byType — matches the type of the group's first token.

A group is a standalone top-level token, or an opening token with everything up to its matching closing token. A group is moved as a whole.

Processing rules

  1. Groups are processed in source order.
  2. A group goes to the first slot whose match returns true and whose limit is not reached.
  3. A group no slot accepts is dropped. A last slot with match: () => true keeps the remaining content.
  4. Slots are emitted in rule order, and groups keep source order within a slot. Empty slots emit their wrappers too.
  5. Moved tokens get level + 1. A wrapper's map spans its groups.
  6. match receives the whole group with inline.children filled, so it can check nested markup. Markdown-it fills children in the core inline rule after block parsing, so slotsTokenizer fills them itself without duplicates after that rule runs.
  7. Preview and WYSIWYG use the same token stream. Editor slot nodes accept the content their rules accept, including an empty slot.

Usage

registerContainerDirective(md, {
    name: 'block',
    match: () => true,
    container: {tag: 'div', token: 'block'},
    contentTokenizer: slotsTokenizer('block', [
        {slot: 'block-a', match: byType('paragraph_open'), limit: 1},
        {slot: 'block-b', match: byType('paragraph_open', 'bullet_list_open')},
    ]),
});

The first paragraph goes to block-a. Later paragraphs and bullet lists go to block-b. Everything else is dropped.

Lenguaje dominante
TypeScript
Estrellas
428
Forks
48
Merge medio
5 d 15 h
PR fusionados (30 d)
47

Preparar el entorno

Este proyecto no incluye contenedor de desarrollo, Dockerfile ni guía de contribución, así que la configuración corre por tu cuenta: empieza por su README y consulta nuestra guía para la primera contribución para los pasos generales.

Primeros pasos

  1. Lee el issue completo y luego la guía de contribución del proyecto.
  2. Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
  3. Haz un fork del repositorio y trabaja en una rama.
  4. Abre un pull request que haga referencia al número del issue.

Más de gravity-ui/markdown-editor

Todos los issues de gravity-ui/markdown-editor

Issues similares

Más issues de TypeScript

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.