[chromedriver] Generate as much filler text as needed for testMoveToElementAndClick

[chromedriver] Generate as much filler text as needed for testMoveToElementAndClick

BUG=

Committed: https://crrev.com/6bf3192b1dc84ab54371e80e13dce93c90cd3e4a
Cr-Commit-Position: refs/heads/master@{#313106}

[samuong, 2015-01-23 20:23:45]

samuong@chromium.org changed reviewers:
+ stgao@chromium.org

[samuong, 2015-01-23 20:23:46]

crbug.com/866113002 fixed it on win7, but broke it on android. I don't think
there's a way to use the same page everywhere, since fonts get rendered
differently depending on the platform and location configuration. So I've added
some JavaScript to generate as much filler text as needed, which should
everywhere.

[stgao, 2015-01-24 01:57:56]

lgtm if the cl is well-tested.

[samuong, 2015-01-26 18:54:30]

On 2015/01/24 01:57:56, Shuotao wrote:
> lgtm if the cl is well-tested.

I've tested this on:

- Windows 7 (local desktop with antialiasing, and remote to buildbot without)
- Linux (with antialiasing and without, and inside mobile emulation)
- Mac
- Android (inside the Android emulator)

It's difficult to test this exhaustively, as it's not just platform-dependent
but also dependent on the host's font settings. In particular, hosts that don't
have antialiased text (e.g. our Win7 bot over rdesktop, and the default
configuration of certain Linux distros) render and reflow text differently from
those that do.

[samuong, 2015-01-26 18:54:34]

The CQ bit was checked by samuong@chromium.org

[commit-bot: I haz the power, 2015-01-26 18:55:34]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/872883002/20001

[stgao, 2015-01-26 19:09:24]

On 2015/01/26 18:54:30, samuong wrote:
> On 2015/01/24 01:57:56, Shuotao wrote:
> > lgtm if the cl is well-tested.
> 
> I've tested this on:
> 
> - Windows 7 (local desktop with antialiasing, and remote to buildbot without)
> - Linux (with antialiasing and without, and inside mobile emulation)
> - Mac
> - Android (inside the Android emulator)
> 
> It's difficult to test this exhaustively, as it's not just platform-dependent
> but also dependent on the host's font settings. In particular, hosts that
don't
> have antialiased text (e.g. our Win7 bot over rdesktop, and the default
> configuration of certain Linux distros) render and reflow text differently
from
> those that do.

Thanks for all the testing.
Yes, I agree this testcase is kind of depending on host/platform.

[commit-bot: I haz the power, 2015-01-26 20:04:44]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-01-26 20:05:32]

Patchset 2 (id:??) landed as
https://crrev.com/6bf3192b1dc84ab54371e80e13dce93c90cd3e4a
Cr-Commit-Position: refs/heads/master@{#313106}