alex/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2026-04-06 07:47:53 +10:00
Code Issues Packages Projects Releases Wiki Activity

tons of fixes with codex

Browse Source
This commit is contained in:
Nick Sweeting
2026-01-19 01:00:53 -08:00
parent eaf7256345
commit c7b2217cd6
184 changed files with 3943 additions and 2420 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
archivebox/plugins/parse_dom_outlinks/on_Snapshot__75_parse_dom_outlinks.js
View File

@@ -193,6 +193,9 @@ async function extractOutlinks(url) {
type: 'Snapshot',
url: href,
plugin: PLUGIN_NAME,
depth: depth + 1,
parent_snapshot_id: snapshotId || undefined,
crawl_id: crawlId || undefined,
})).join('\n');
if (urlsJsonl) {
@@ -214,6 +217,8 @@ async function main() {
const args = parseArgs();
const url = args.url;
const snapshotId = args.snapshot_id;
const crawlId = args.crawl_id || process.env.CRAWL_ID;
const depth = parseInt(args.depth || process.env.SNAPSHOT_DEPTH || '0', 10) || 0;
if (!url || !snapshotId) {
console.error('Usage: on_Snapshot__75_parse_dom_outlinks.js --url=<url> --snapshot-id=<uuid>');

2
archivebox/plugins/parse_dom_outlinks/templates/icon.html
View File

@@ -1 +1 @@
🔗
<span class="abx-output-icon abx-output-icon--parse_dom_outlinks" title="Outlinks"><svg width="16" height="16" viewBox="0 0 24 24" aria-hidden="true" focusable="false" fill="none" stroke="currentColor" stroke-width="1.8" stroke-linecap="round" stroke-linejoin="round"><path d="M10 13a4 4 0 0 1 0-6l2-2a4 4 0 0 1 6 6l-1 1"/><path d="M14 11a4 4 0 0 1 0 6l-2 2a4 4 0 0 1-6-6l1-1"/></svg></span>

1
archivebox/plugins/parse_dom_outlinks/tests/__init__.py
View File

@@ -1 +0,0 @@
"""Tests for the parse_dom_outlinks plugin."""

3
archivebox/plugins/parse_dom_outlinks/tests/test_parse_dom_outlinks.py
View File

@@ -79,8 +79,7 @@ class TestParseDomOutlinksWithChrome(TestCase):
# Run outlinks hook with the active Chrome session
result = subprocess.run(
['node', str(OUTLINKS_HOOK), f'--url={test_url}', f'--snapshot-id={snapshot_id}'],
cwd=str(snapshot_chrome_dir,
env=get_test_env()),
cwd=str(snapshot_chrome_dir),
capture_output=True,
text=True,
timeout=60,