Skip to content

fix(utils): keep urls from the other sitemaps when one sitemap is malformed #4249

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
jayhemnani9910 wants to merge 1 commit into
base: master
Choose a base branch
from
+72 −34
Open

fix(utils): keep urls from the other sitemaps when one sitemap is malformed #4249

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
75 changes: 41 additions & 34 deletions packages/utils/src/internals/sitemap.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -385,47 +385,54 @@ export async function* parseSitemap<T extends ParseSitemapOptions>(
// the loop (per-entry warnings could flood the log; individual drops are logged at debug level).
let droppedUrlEntries = 0;

for await (const item of items) {
if (item.type === 'sitemapUrl' && !visitedSitemapUrls.has(item.url)) {
if (nestedSitemapFilter && !nestedSitemapFilter(item.url)) {
logger?.debug(`Skipping sitemap ${item.url} due to nestedSitemapFilter.`);
continue;
}

// Keep only nested sitemaps matching the strategy (and using http(s)) relative to the
// parent. Raw string sources have no parent URL, so the check is skipped.
if (source.type === 'url') {
const { allowed, reason } = filterUrl(item.url, sitemapUrl!, enqueueStrategy);
if (!allowed) {
logger?.warning(`Skipping nested sitemap ${item.url} (parent ${source.url}): ${reason}.`);
try {
for await (const item of items) {
if (item.type === 'sitemapUrl' && !visitedSitemapUrls.has(item.url)) {
if (nestedSitemapFilter && !nestedSitemapFilter(item.url)) {
logger?.debug(`Skipping sitemap ${item.url} due to nestedSitemapFilter.`);
continue;
}
}

sources.push({ type: 'url', url: item.url, depth: (source.depth ?? 0) + 1 });
if (emitNestedSitemaps) {
yield { loc: item.url, originSitemapUrl: null } as any;
}
}
// Keep only nested sitemaps matching the strategy (and using http(s)) relative to the
// parent. Raw string sources have no parent URL, so the check is skipped.
if (source.type === 'url') {
const { allowed, reason } = filterUrl(item.url, sitemapUrl!, enqueueStrategy);
if (!allowed) {
logger?.warning(`Skipping nested sitemap ${item.url} (parent ${source.url}): ${reason}.`);
continue;
}
}

if (item.type === 'url') {
// Keep only URL entries that match the enqueue strategy relative to the parent (see above).
if (source.type === 'url') {
const { allowed, reason } = filterUrl(item.loc, sitemapUrl!, enqueueStrategy);
if (!allowed) {
droppedUrlEntries++;
logger?.debug(`Skipping sitemap URL ${item.loc} (parent ${source.url}): ${reason}.`);
continue;
sources.push({ type: 'url', url: item.url, depth: (source.depth ?? 0) + 1 });
if (emitNestedSitemaps) {
yield { loc: item.url, originSitemapUrl: null } as any;
}
}

yield {
...item,
originSitemapUrl:
source.type === 'url'
? source.url
: `raw://${createHash('sha256').update(source.content).digest('base64')}`,
};
if (item.type === 'url') {
// Keep only URL entries that match the enqueue strategy relative to the parent (see above).
if (source.type === 'url') {
const { allowed, reason } = filterUrl(item.loc, sitemapUrl!, enqueueStrategy);
if (!allowed) {
droppedUrlEntries++;
logger?.debug(`Skipping sitemap URL ${item.loc} (parent ${source.url}): ${reason}.`);
continue;
}
}

yield {
...item,
originSitemapUrl:
source.type === 'url'
? source.url
: `raw://${createHash('sha256').update(source.content).digest('base64')}`,
};
}
}
} catch (e) {
// Skip a malformed sitemap without losing the other ones; raw sources already report the error above.
if (source.type === 'url') {
logger?.warning(`Malformed sitemap content: ${source.url} (${e})`);
}
}

Expand Down
31 changes: 31 additions & 0 deletions test/utils/sitemap.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -143,6 +143,21 @@ describe('Sitemap', () => {
'</sitemapindex>',
].join('\n'),
)
.get('/sitemap_parent_with_malformed_child.xml')
.reply(
200,
[
'<?xml version="1.0" encoding="UTF-8"?>',
'<sitemapindex xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">',
'<sitemap>',
'<loc>http://not-exists.com/not_actual_xml.xml</loc>',
'</sitemap>',
'<sitemap>',
'<loc>http://not-exists.com/sitemap_child.xml</loc>',
'</sitemap>',
'</sitemapindex>',
].join('\n'),
)
.get('/not_actual_xml.xml')
.reply(
200,
Expand Down Expand Up @@ -393,6 +408,22 @@ describe('Sitemap', () => {
expect(sitemap.urls).toEqual([]);
});

it('keeps urls from other sitemaps when one of them is malformed', async () => {
const sitemap = await Sitemap.load(
['http://not-exists.com/not_actual_xml.xml', 'http://not-exists.com/sitemap_child.xml'],
undefined,
{ httpClient: new FetchHttpClient() },
);
expect(sitemap.urls).toHaveLength(5);
});

it('keeps urls from other nested sitemaps when one of them is malformed', async () => {
const sitemap = await Sitemap.load('http://not-exists.com/sitemap_parent_with_malformed_child.xml', undefined, {
httpClient: new FetchHttpClient(),
});
expect(sitemap.urls).toHaveLength(5);
});

it('handles CDATA in loc tags', async () => {
const sitemap = await Sitemap.load('http://not-exists.com/sitemap_cdata.xml', undefined, {
httpClient: new FetchHttpClient(),
Expand Down
Loading