[Android WebView] Add a test for browser sign-in back navigation

[Android WebView] Add a test for browser sign-in back navigation

Add a regression test for a non-cacheable login page with a help link.

BUG=b/8888772

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=203986

[boliu, 2013-06-03 16:34:02]

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
File
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java
(right):

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java:235:
"    <form action='login.html' action='post'>" +
LOGIN_PAGE_PATH.substring(1) instead of login.html?

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java:236:
"      <input type='text' name='login'>" +
This 'login' is matches location.search.indexOf('login'), but is independent of
LOGIN_PAGE_PATH right? Can we use a different string like 'post_request' to
avoid confusion. Also use a static constant.

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java:244:
CommonResources.getTextHtmlHeaders(true));
You mentioned in original bug that the response had "Cache-control:no-cache,
no-store". CommonResources adds no-store only.

Might not matter but just pointing it out.

[mnaganov (inactive), 2013-06-03 16:46:25]

Thanks, Bo!

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
File
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java
(right):

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java:235:
"    <form action='login.html' action='post'>" +
On 2013/06/03 16:34:02, boliu wrote:
> LOGIN_PAGE_PATH.substring(1) instead of login.html?

Done.

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java:236:
"      <input type='text' name='login'>" +
On 2013/06/03 16:34:02, boliu wrote:
> This 'login' is matches location.search.indexOf('login'), but is independent
of
> LOGIN_PAGE_PATH right? Can we use a different string like 'post_request' to
> avoid confusion. Also use a static constant.

location.search doesn't contain the page name, only the query string, so there
is no confusion for the computer. Added a local constant to as a courtesy for
humans. Also, thanks to your comment, I've noticed that I'm referencing 'help'
from the test, so added a static constant for it.

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java:244:
CommonResources.getTextHtmlHeaders(true));
On 2013/06/03 16:34:02, boliu wrote:
> You mentioned in original bug that the response had "Cache-control:no-cache,
> no-store". CommonResources adds no-store only.
> 
> Might not matter but just pointing it out.

Yes, this doesn't matter, as Chromium ignores 'no-cache' according to Darin's
reply on my question on blink-dev.

[boliu, 2013-06-03 16:52:57]

lgtm

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
File
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java
(right):

https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java:236:
"      <input type='text' name='login'>" +
On 2013/06/03 16:46:25, Mikhail Naganov (Chromium) wrote:
> On 2013/06/03 16:34:02, boliu wrote:
> > This 'login' is matches location.search.indexOf('login'), but is independent
> of
> > LOGIN_PAGE_PATH right? Can we use a different string like 'post_request' to
> > avoid confusion. Also use a static constant.
> 
> location.search doesn't contain the page name, only the query string, so there
> is no confusion for the computer. Added a local constant to as a courtesy for
> humans. Also, thanks to your comment, I've noticed that I'm referencing 'help'
> from the test, so added a static constant for it.

Right, I was referring to confusion for humans, because the url would look like
"login.html?login"

[mnaganov (inactive), 2013-06-04 08:37:23]

On 2013/06/03 16:52:57, boliu wrote:
> lgtm
> 
>
https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
> File
>
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java
> (right):
> 
>
https://codereview.chromium.org/16337004/diff/1/android_webview/javatests/src...
>
android_webview/javatests/src/org/chromium/android_webview/test/NavigationHistoryTest.java:236:
> "      <input type='text' name='login'>" +
> On 2013/06/03 16:46:25, Mikhail Naganov (Chromium) wrote:
> > On 2013/06/03 16:34:02, boliu wrote:
> > > This 'login' is matches location.search.indexOf('login'), but is
independent
> > of
> > > LOGIN_PAGE_PATH right? Can we use a different string like 'post_request'
to
> > > avoid confusion. Also use a static constant.
> > 
> > location.search doesn't contain the page name, only the query string, so
there
> > is no confusion for the computer. Added a local constant to as a courtesy
for
> > humans. Also, thanks to your comment, I've noticed that I'm referencing
'help'
> > from the test, so added a static constant for it.
> 
> Right, I was referring to confusion for humans, because the url would look
like
> "login.html?login"

BTW, later I realized that if we were really using POST, there should be no
query string. And indeed, there was a typo in the form code--I wrote
"action='post'" instead of "method='post'". Fixed now.

[commit-bot: I haz the power, 2013-06-04 09:29:57]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/mnaganov@chromium.org/16337004/7001

[commit-bot: I haz the power, 2013-06-04 16:59:20]

Change committed as 203986