2094f92 · hyper-note · pear-git-v3

Changed files (2) +63 -0

lib/multiwriter/notebook-apply.js +24 -0

…	…	@@ -119,3 +119,27 @@ export async function notebookApply (batch, view, base) {
119	119
120	120	await b.flush()
121	121	}
	122
	123	/**
	124	* Extract media hashes/identifiers from markdown note content.
	125	* Matches both old-style filenames and new SHA-256 hashes in media URLs.
	126	*/
	127	export function extractMediaHashes (content) {
	128	const hashes = new Set()
	129	if (!content) return hashes
	130
	131	// Match ![...](/api/notebooks/.../media/IDENTIFIER)
	132	const urlPattern = /!\[[^\]]\]\([^)]\/media\/([a-zA-Z0-9._-]+)\)/g
	133	let match
	134	while ((match = urlPattern.exec(content)) !== null) {
	135	hashes.add(match[1])
	136	}
	137
	138	// Match bare hash references ![...](HASH) where HASH is 64 hex chars
	139	const hashPattern = /!\[[^\]]*\]\(([a-f0-9]{64})\)/g
	140	while ((match = hashPattern.exec(content)) !== null) {
	141	hashes.add(match[1])
	142	}
	143
	144	return hashes
	145	}

test/unit/notebook-gc.test.js +39 -0

…	…	@@ -0,0 +1,39 @@
	1	import test from 'brittle'
	2	import { extractMediaHashes } from '../../lib/multiwriter/notebook-apply.js'
	3
	4	test('extractMediaHashes - finds image hashes in markdown', async (t) => {
	5	const content = 'Some text ![photo](/api/notebooks/general/media/abc123def456) and more ![vid](/api/notebooks/general/media/789012fed345)'
	6	const hashes = extractMediaHashes(content)
	7	t.is(hashes.size, 2)
	8	t.ok(hashes.has('abc123def456'))
	9	t.ok(hashes.has('789012fed345'))
	10	})
	11
	12	test('extractMediaHashes - returns empty set for no media', async (t) => {
	13	const content = 'Just text, no images'
	14	const hashes = extractMediaHashes(content)
	15	t.is(hashes.size, 0)
	16	})
	17
	18	test('extractMediaHashes - returns empty set for null/undefined', async (t) => {
	19	t.is(extractMediaHashes(null).size, 0)
	20	t.is(extractMediaHashes(undefined).size, 0)
	21	t.is(extractMediaHashes('').size, 0)
	22	})
	23
	24	test('extractMediaHashes - handles 64-char hex hash references', async (t) => {
	25	const hash = 'a'.repeat(64)
	26	const content = '![image](' + hash + ')'
	27	const hashes = extractMediaHashes(content)
	28	t.is(hashes.size, 1)
	29	t.ok(hashes.has(hash))
	30	})
	31
	32	test('extractMediaHashes - handles mixed old and new style refs', async (t) => {
	33	const hash = 'b'.repeat(64)
	34	const content = '![old](/api/notebooks/dev/media/lq3x2abc.jpg)\n![new](' + hash + ')'
	35	const hashes = extractMediaHashes(content)
	36	t.is(hashes.size, 2)
	37	t.ok(hashes.has('lq3x2abc.jpg'))
	38	t.ok(hashes.has(hash))
	39	})