markdown-formatter.ts•5.61 kB
/**
* Markdown to Plain Text Formatter for Attio Notes
*
* Converts markdown content to well-formatted plain text that displays
* nicely in Attio's note interface, which doesn't support rich text.
*/
export interface MarkdownFormatterOptions {
/** Character to use for bullet points (default: '•') */
bulletChar?: string;
/** Maximum line length before wrapping (default: 80) */
maxLineLength?: number;
/** Whether to preserve original spacing (default: false) */
preserveSpacing?: boolean;
/** Custom section separators */
sectionSeparators?: {
h1?: string;
h2?: string;
h3?: string;
};
}
const DEFAULT_OPTIONS: Required<MarkdownFormatterOptions> = {
bulletChar: '•',
maxLineLength: 80,
preserveSpacing: false,
sectionSeparators: {
h1: '=',
h2: '-',
h3: '~',
},
};
/**
* Converts markdown content to formatted plain text suitable for Attio notes
*
* @param markdown - The markdown content to convert
* @param options - Optional formatting configuration
* @returns Formatted plain text with proper spacing and structure
*/
export function convertMarkdownToPlainText(
markdown: string,
options: MarkdownFormatterOptions = {}
): string {
if (!markdown || typeof markdown !== 'string') {
return '';
}
const opts = { ...DEFAULT_OPTIONS, ...options };
let text = markdown;
// Step 1: Handle headers with proper spacing and separators
text = text.replace(/^# (.+)$/gm, (match, title) => {
const separator = opts.sectionSeparators.h1!.repeat(
Math.min(title.length, 60)
);
return `\n\n${separator}\n${title.toUpperCase()}\n${separator}\n`;
});
text = text.replace(/^## (.+)$/gm, (match, title) => {
const separator = opts.sectionSeparators.h2!.repeat(
Math.min(title.length, 50)
);
return `\n\n${title}\n${separator}\n`;
});
text = text.replace(/^### (.+)$/gm, (match, title) => {
return `\n\n${title}:\n`;
});
// Step 2: Handle bold text markers - convert to section headers
text = text.replace(/^\*\*([^*]+)\*\*:?\s*$/gm, '\n\n$1:\n');
text = text.replace(/\*\*([^*]+)\*\*/g, '$1'); // Remove remaining bold
// Step 3: Handle bullet points with proper indentation
text = text.replace(/^- (.+)$/gm, `${opts.bulletChar} $1`);
text = text.replace(/^\* (.+)$/gm, `${opts.bulletChar} $1`);
// Step 4: Handle numbered lists
text = text.replace(/^(\d+)\. (.+)$/gm, '$1. $2');
// Step 5: Handle code blocks and inline code
text = text.replace(/```[\s\S]*?```/g, ''); // Remove code blocks
text = text.replace(/`([^`]+)`/g, '"$1"'); // Convert inline code to quotes
// Step 6: Handle links - extract URL and show it cleanly
text = text.replace(/\[([^\]]+)\]\(([^)]+)\)/g, '$1 ($2)');
text = text.replace(/https?:\/\/[^\s)]+/g, (url) => {
// Keep URLs readable but not too long
return url.length > 60 ? url.substring(0, 57) + '...' : url;
});
// Step 7: Clean up excessive whitespace while preserving intentional spacing
if (!opts.preserveSpacing) {
// Remove trailing spaces from lines
text = text.replace(/ +$/gm, '');
// Normalize multiple blank lines to at most 2
text = text.replace(/\n{3,}/g, '\n\n');
// Ensure sections have proper spacing
text = text.replace(/(\n[A-Z\-=~]{3,}\n)/g, '\n$1');
}
// Step 8: Handle special formatting for key-value pairs
text = text.replace(/^([A-Z][^:\n]+):\s*(.+)$/gm, (match, key, value) => {
if (value.length > 60) {
return `${key}:\n ${value}`;
}
return `${key}: ${value}`;
});
// Step 9: Final cleanup
text = text.trim();
// Ensure the text starts cleanly
text = text.replace(/^[\n\s]+/, '');
return text;
}
/**
* Specialized formatter for business research notes
* Optimized for the common patterns in business qualification content
*/
export function formatBusinessResearchNote(markdown: string): string {
return convertMarkdownToPlainText(markdown, {
bulletChar: '•',
maxLineLength: 90,
sectionSeparators: {
h1: '=',
h2: '-',
h3: '',
},
});
}
/**
* Compact formatter for shorter notes
* Uses minimal spacing for concise presentation
*/
export function formatCompactNote(markdown: string): string {
let text = convertMarkdownToPlainText(markdown, {
bulletChar: '→',
preserveSpacing: true,
});
// Further compress spacing for compact format
text = text.replace(/\n\n+/g, '\n');
text = text.replace(/^[\n\s]+/, '');
return text;
}
/**
* Validates if content appears to be markdown and would benefit from formatting
*/
export function isMarkdownContent(content: string): boolean {
if (!content || typeof content !== 'string') {
return false;
}
const markdownPatterns = [
/^#{1,6}\s+.+$/m, // Headers
/^\*\*[^*]+\*\*$/m, // Bold text on its own line
/^[-*]\s+.+$/m, // Bullet points
/^\d+\.\s+.+$/m, // Numbered lists
/\[.+\]\(.+\)/, // Links
/```[\s\S]*?```/, // Code blocks
];
return markdownPatterns.some((pattern) => pattern.test(content));
}
/**
* Auto-detects the best formatting approach based on content
*/
export function autoFormatContent(content: string): string {
if (!isMarkdownContent(content)) {
return content;
}
// If content is very long, use business research formatting
if (content.length > 1000) {
return formatBusinessResearchNote(content);
}
// For shorter content, use compact formatting
if (content.length < 300) {
return formatCompactNote(content);
}
// Default formatting for medium-length content
return convertMarkdownToPlainText(content);
}