import apiFetch from '@wordpress/api-fetch';
import { parse } from '@wordpress/block-serialization-default-parser';
export default async ({ postId, postType }) => {
const type = postType === 'page' ? 'pages' : 'posts';
const response = await apiFetch({
path: `/wp/v2/${type}/${postId}?context=edit`,
});
const blocks = parse(response.content.raw);
return {
post_strings: dedupeStrings([
response.title.raw,
...extractTextFromBlocks(blocks),
]),
};
};
const newSet = (arr) => new Set(arr.filter(Boolean));
const dedupeStrings = (arr) => [...newSet(arr)];
const stripHtml = (html) =>
html
.replace(/<[^>]+>/g, '')
.replace(/\s+/g, ' ')
.trim();
// Handles image stuff
const extractAltAndTitleFromHtml = (html) => {
const matches = [];
const altMatch = html.match(/alt="([^"]*)"/);
if (altMatch && altMatch[1]) matches.push(altMatch[1].trim());
const titleMatch = html.match(/title="([^"]*)"/);
if (titleMatch && titleMatch[1]) matches.push(titleMatch[1].trim());
return matches;
};
const extractTextFromBlocks = (blocks) => {
if (!blocks || blocks.length === 0) return [];
return blocks.flatMap((block) => [
// Extract from innerContent (rendered HTML)
...(block.innerContent
? block.innerContent
.filter(Boolean)
.flatMap((html) => [
stripHtml(html),
...extractAltAndTitleFromHtml(html),
])
.filter(Boolean)
: []),
// Extract from relevant string attributes
...['content', 'caption', 'alt', 'title', 'value']
.map((key) =>
typeof block.attrs?.[key] === 'string' ? block.attrs[key].trim() : null,
)
.filter(Boolean),
// Recurse into innerBlocks
...extractTextFromBlocks(block.innerBlocks || []),
]);
};