transforms.ts

Utility functions for converting extracted content into formats suitable for Umbraco document properties.

What it does

Provides four exported functions for content transformation:

1. markdownToHtml — converts Markdown text to HTML using the marked library
2. normalizeToKebabCase — normalizes text to kebab-case for section IDs (mirrors C# NormalizeToKebabCase)
3. stripMarkdown — strips Markdown formatting, returning plain text (for text/textArea fields)
4. buildRteValue — wraps HTML content in the Umbraco Rich Text Editor (RTE) value structure

Functions

markdownToHtml

export function markdownToHtml(markdown: string): string

Converts a Markdown string to HTML using the marked library with GitHub Flavored Markdown (GFM) enabled:

const html = marked.parse(markdown, {
    gfm: true,      // GitHub Flavored Markdown
    breaks: false,   // Don't convert \n to <br>
});

Fallback handling:

• Returns an empty string if the input is falsy
• If marked.parse returns a Promise (unexpected with sync config), falls back to wrapping in <p> tags
• If an exception occurs, falls back to manual conversion: double newlines become </p><p>, single newlines become <br>

normalizeToKebabCase

export function normalizeToKebabCase(text: string): string

Converts text to kebab-case for use as section IDs. Mirrors the C# NormalizeToKebabCase in ContentTransformService:

• Lowercases, trims, replaces non-alphanumeric runs with hyphens, strips leading/trailing hyphens
• "FEATURES" → "features", "WHAT WE WILL SEE" → "what-we-will-see"

stripMarkdown

export function stripMarkdown(markdown: string): string

Strips Markdown formatting from text, returning plain text. Used when mapping markdown content to plain text fields (text, textArea). Removes:

• Heading prefixes (# ## ###)
• Bold (**text**) and italic (*text*) markers
• Strikethrough (~~text~~) and inline code (`text`)
• Bullet list prefixes (- * +), numbered list prefixes (1. 2.), blockquote prefixes (>)

Returns an empty string if the input is falsy.

buildRteValue

export function buildRteValue(htmlContent: string)

Builds the value object expected by Umbraco’s Rich Text Editor property:

return {
    blocks: {
        contentData: [],
        settingsData: [],
        expose: [],
        Layout: {},
    },
    markup: htmlContent,
};

The blocks structure is required by Umbraco’s RTE even when no blocks are used. The markup field contains the actual HTML content.

Key concepts

Extracted from action

These functions were originally the #convertToHtml private method on UpDocEntityAction. They were extracted to a separate module so both the action’s #applyBlockMapping method and any future consumers can reuse them.

Used together

The typical usage pattern is to pipe content through both functions:

const htmlContent = markdownToHtml(sectionValue);
targetValue.value = buildRteValue(htmlContent);

This converts extracted Markdown content into the full RTE value structure expected by Umbraco’s Management API.

Imports

import { marked } from 'marked';

The marked library is a dependency of the UpDoc npm package.

Used by

• up-doc-action.ts — markdownToHtml/buildRteValue for richText fields, stripMarkdown for text/textArea fields
• up-doc-collection-action.element.ts — same as above (bridge files kept in sync)
• up-doc-modal.element.ts — stripMarkdown for document name pre-fill