Regression: Error page is not seen on navigating to 'chrome://,./' page.
Reported by
vvishwak...@etouch.net,
Oct 26 2016
|
||||||
Issue descriptionVersion: 56.0.2900.0 (Official Build) canary 1e4fbec3b5701d9f605e88f14dafa442be98b648-refs/heads/master@{#427219} (32/64-bit) OS: Windows (7,8,10), Mac(10.10.5, 10.11.4), Linux (14.04 LTS) What steps will reproduce the problem? 1) Launch chrome, open NTP, go to ‘chrome://,./‘ and observe. Error page is not seen on navigating to chrome://,./ page. Error page should be seen on navigating to chrome://,./ page. This is a Regression issue broken in M-56, will soon update other info Manual bisect: Good build: 56.0.2888.0 Bad build: 56.0.2890.0
,
Oct 26 2016
This looks reproducible even with "chrome://,/" +nick It looks like the URL is properly rewriting to chrome://%2C/, but then rewriting that to an about:blank navigation (so the net stack doesn't give us an INVALID_URL error). Not sure how my change is involved yet but it probably is. Will try to pull up a debugger today and investigate.
,
Oct 26 2016
Adding RB label as this is a recent regression.
,
Oct 26 2016
Actually this hits a DCHECK on SchemeHostPort assuming data coming from the security origin is already canonicalized. In particular, it looks like SecurityOrigin is sending lowercased escape characters, where we consider "canonical urls" to be uppercased escape characters! Brett, this is exactly what we were talking about hours ago in this revision: https://codereview.chromium.org/2447293002/ In fact, if we stop calling lower() on host() in constructing SecurityOrigins, we fix this bug :) Some other weirdness if happening that I'll try to figure out, but I think we can solve this problem in the linked revision, with more test cases.
,
Oct 31 2016
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/2874fa423262879ed27036f716f3e96a5f8e863f commit 2874fa423262879ed27036f716f3e96a5f8e863f Author: csharrison <csharrison@chromium.org> Date: Mon Oct 31 18:52:03 2016 Don't call lower() on KURL protocol/host KURL should already canonicalize the scheme and host. This is just needless work and causes confusion to readers who may get the wrong impression about code invariants. Additionally, calling lower() on the host is a bug, because we canonicalize escape characters to uppercase, but lower() reverts that and brings them back to lowercase, effecively rendering the host not canonical anymore! This is a pre-req for removing CaseFoldingHash from SchemeRegistry. BUG= 659550 ,348655 Review-Url: https://codereview.chromium.org/2447293002 Cr-Commit-Position: refs/heads/master@{#428762} [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/content/child/blink_platform_impl_unittest.cc [add] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/LayoutTests/http/tests/security/document-domain-canonicalizes.html [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/LayoutTests/http/tests/security/document-domain-invalid.html [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/Source/core/dom/Document.cpp [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/Source/platform/weborigin/KURL.cpp [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/Source/platform/weborigin/KURLTest.cpp [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/Source/platform/weborigin/KnownPorts.cpp [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/Source/platform/weborigin/OriginAccessEntry.cpp [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/Source/platform/weborigin/SecurityOrigin.cpp [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/Source/platform/weborigin/SecurityOrigin.h [modify] https://crrev.com/2874fa423262879ed27036f716f3e96a5f8e863f/third_party/WebKit/Source/platform/weborigin/SecurityOriginTest.cpp
,
Nov 1 2016
Tested the issue on Windows 7,Mac 10.11.4,Linux Ubuntu 14.04 using chrome Dev version-56.0.2906.0 with the steps mentioned in comment #0.Issue working as expected. Please find the attached screen cast for the same. Adding TE-Verified labels. Thanks,
,
Nov 7 2016
|
||||||
►
Sign in to add a comment |
||||||
Comment 1 by nyerramilli@chromium.org
, Oct 26 2016Owner: csharrison@chromium.org
Status: Assigned (was: Unconfirmed)