Macro is a unified workspace for teams: email, chat, docs, tasks, agents, calls, and CRM — @-linked together with shared AI memory.
TypeScript agentaiai-agentsall-in-onecrmcrm-systememaillinearmcpmessagingnotesnotionofficerustslackslack-alternativestartupstartup-toolstypescriptworkspace
/**
* Generalized utilities for extracting potential tasks from markdown content.
* Used by Task Mode in channel input to preview and create tasks from checkboxes.
*/
import {
extractDateMention,
extractTitleFromMarkdown,
extractUserMentions,
} from '@core/component/LexicalMarkdown/plugins/checkbox-to-task/checkboxParsing';
export { extractUserMentions };
/**
* Represents a potential task extracted from markdown text.
* Unlike ParsedCheckbox (which requires Lexical node context),
* this works with raw markdown strings.
*/
type PotentialTask = {
/** Line index in the source markdown (0-based) */
lineIndex: number;
/** Cleaned title without mentions */
title: string;
/** Original line for replacement matching */
rawLine: string;
/** User IDs extracted from @mentions */
assigneeUserIds: string[];
/** ISO date string from date mention, or null */
dueDate: Date | null;
/** Whether the checkbox is checked [x] or unchecked [ ] */
isChecked: boolean;
};
// Pattern to match checkbox lines: "- [ ] text" or "- [x] text"
// Captures: (1) leading whitespace, (2) check state, (3) content after checkbox
const CHECKBOX_LINE_PATTERN = /^(\s*)-\s*\[([ xX])\]\s*(.*)$/;
const LEADING_USER_MENTION_PATTERN =
/^<m-user-mention>[^<]*<\/m-user-mention>\s*/;
const TRAILING_USER_MENTION_PATTERN =
/\s*<m-user-mention>[^<]*<\/m-user-mention>$/;
/**
* Trim user @mention tags from the start and end of a markdown string.
* Mentions in the middle are preserved.
*/
export function trimEdgeUserMentions(markdown: string): string {
let result = markdown.trim();
let prev: string;
do {
prev = result;
result = result
.replace(LEADING_USER_MENTION_PATTERN, '')
.replace(TRAILING_USER_MENTION_PATTERN, '')
.trim();
} while (result !== prev);
return result;
}
/**
* Extract all checkbox items from a markdown string as potential tasks.
*
* @param markdown - Raw markdown content
* @returns Array of potential tasks with extracted metadata
*/
export function extractCheckboxesFromMarkdown(
markdown: string
): PotentialTask[] {
const lines = markdown.split('\n');
const tasks: PotentialTask[] = [];
for (let i = 0; i < lines.length; i++) {
const line = lines[i];
const match = CHECKBOX_LINE_PATTERN.exec(line);
if (match) {
const [, leadingWhitespace, checkState, content] = match;
const isChecked = checkState.toLowerCase() === 'x';
// Strip leading whitespace before passing to extractTitleFromMarkdown
// since it expects the line to start with "- [ ]"
const lineWithoutIndent = line.slice(leadingWhitespace.length);
const title = extractTitleFromMarkdown(lineWithoutIndent);
if (!title.trim()) {
continue;
}
tasks.push({
lineIndex: i,
title,
rawLine: line,
assigneeUserIds: extractUserMentions(content),
dueDate: extractDateMention(content),
isChecked,
});
}
}
return tasks;
}