chrome/test/data/extensions/webstore_blocked_by_frame_src.html - Issue 2736863003: Test that no crash happens with the CWS error page.

Side by Side Diff: chrome/test/data/extensions/webstore_blocked_by_frame_src.html

Issue 2736863003: Test that no crash happens with the CWS error page.

Patch Set: Created 3 years, 9 months ago

Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.

Jump to:

OLD	NEW
(Empty)
	1 <!DOCTYPE html>

	2 <html>

	3 <head>

	4 <meta http-equiv="Content-Security-Policy" content="child-src 'none';">

	5 </head>

	6 <body>

	7 <iframe src="https://chrome.google.com/webstore"> </iframe>
	nasko 2017/03/09 05:07:22 Instead of hardcoding this URL in the HTML file, w Instead of hardcoding this URL in the HTML file, why not navigate the iframe from the browser process, where we can use the actual string constant that the browser code will be using to check. If left like that, change in the CWS URL will mean that this test will no longer be valid. Also, that means that you can just use chrome/test/data/iframe_blank.html or some other file. arthursonzogni 2017/03/09 15:40:52 Yes, good idea. I applied your suggestion, but the Show quoted text On 2017/03/09 05:07:22, nasko (slow) wrote: > Instead of hardcoding this URL in the HTML file, why not navigate the iframe > from the browser process, where we can use the actual string constant that the > browser code will be using to check. If left like that, change in the CWS URL > will mean that this test will no longer be valid. Yes, good idea. I applied your suggestion, but the problem is that if I try content::NavigateIframeToURL, then it is waiting for a navigation to happen, but in the current code, there is still no error page and so content::NavigateIframeToURL is waiting forever. I made something that rely on the 'load' event in the next patch. That works in every case. Show quoted text > Also, that means that you can just use chrome/test/data/iframe_blank.html or > some other file. Yes, modulo the CSP. I still need one new file. nasko 2017/03/09 17:18:10 You could use the TestNavigationManager and WaitFo Show quoted text On 2017/03/09 15:40:52, arthursonzogni wrote: > On 2017/03/09 05:07:22, nasko (slow) wrote: > > Instead of hardcoding this URL in the HTML file, why not navigate the iframe > > from the browser process, where we can use the actual string constant that the > > browser code will be using to check. If left like that, change in the CWS URL > > will mean that this test will no longer be valid. > Yes, good idea. > I applied your suggestion, but the problem is that if I try > content::NavigateIframeToURL, then it is waiting for a navigation to happen, but > in the current code, there is still no error page and so > content::NavigateIframeToURL is waiting forever. You could use the TestNavigationManager and WaitForRequestStart(), which could be good enough for you to know the request was made. You don't need to wait for the full commit, though it might be better longer term. Show quoted text > I made something that rely on the 'load' event in the next patch. That works in > every case. > > > Also, that means that you can just use chrome/test/data/iframe_blank.html or > > some other file. > Yes, modulo the CSP. I still need one new file. Oh yeah, good point :). arthursonzogni 2017/03/10 09:25:47 If a crash happens, it will be in RenderFrameHostI Show quoted text On 2017/03/09 17:18:10, nasko (slow) wrote: > On 2017/03/09 15:40:52, arthursonzogni wrote: > > On 2017/03/09 05:07:22, nasko (slow) wrote: > > > Instead of hardcoding this URL in the HTML file, why not navigate the iframe > > > from the browser process, where we can use the actual string constant that > the > > > browser code will be using to check. If left like that, change in the CWS > URL > > > will mean that this test will no longer be valid. > > Yes, good idea. > > I applied your suggestion, but the problem is that if I try > > content::NavigateIframeToURL, then it is waiting for a navigation to happen, > but > > in the current code, there is still no error page and so > > content::NavigateIframeToURL is waiting forever. > > You could use the TestNavigationManager and WaitForRequestStart(), which could > be good enough for you to know the request was made. You don't need to wait for > the full commit, though it might be better longer term. If a crash happens, it will be in RenderFrameHostImpl::OnDidCommitProvisionnalLoad. So the crash occurs after WaitForRequestStart(). I need to wait for the full commit.
	8 </body>

	9 </html>

OLD	NEW

« chrome/browser/extensions/process_management_browsertest.cc ('K') | « chrome/browser/extensions/process_management_browsertest.cc ('k') | no next file » | no next file with comments »