microlinkhq / browserless / 42095928827

77%

push

github

fix(browserless): a call after destroyContext rejects instead of reopening (#1023)

* fix(browserless): a call after destroyContext rejects instead of reopening

A call made after destroyContext() reopened the context: a new Chromium
browser context and, with a credentialed proxyServer, a new loopback
forwarder. Nothing closed either one until browser.close(), which a
long-lived pool never calls.

destroyContext() is now final. Every later call that needs the context
rejects: html/text/screenshot/pdf/evaluate/withPage/report with an
AbortError, page/context with EBRWSRCONTEXTCONNRESET. The retry path
already checks isAbandoned() synchronously before newBrowserContext(),
and that check now sees the destroyed flag.

Closes #1018

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix(browserless): lock the error from a call after destroyContext

The nullish create in getBrowserContext cannot run, and the tests accepted either rejection. Page-loading calls must reject with AbortError, and page and context with EBRWSRCONTEXTCONNRESET.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix(goto): wait for the fingerprintjs demo identifier

The demo no longer includes `.giant` in the initial HTML, so reading it once throws before the visitor identifier is rendered.

Co-authored-by: Cursor <cursoragent@cursor.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Cursor <cursoragent@cursor.com>

2208 of 2488 branches covered (88.75%)

Branch coverage included in aggregate %.

5 of 5 new or added lines in 1 file covered. (100.0%)

7820 of 10414 relevant lines covered (75.09%)

57716.36 hits per line