[webkit-reviews] review denied: [Bug 38195] Bring CLIENT_BASED_GEOLOCATION more in line with non-client based: remove Geolocation::setPosition and use lastPosition() : [Attachment 54413] Patch

[bugzilla-daemon at webkit.org]

Steve Block <steveblock at google.com> has denied Marcus Bulach
<bulach at chromium.org>'s request for review:
Bug 38195: Bring CLIENT_BASED_GEOLOCATION more in line with non-client based:
remove Geolocation::setPosition and use lastPosition()
https://bugs.webkit.org/show_bug.cgi?id=38195

Attachment 54413: Patch
https://bugs.webkit.org/attachment.cgi?id=54413&action=review

------- Additional Comments from Steve Block <steveblock at google.com>
> - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 
> + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
I think it's best to leave the whitespace changes out of this patch for
clarity.

> +    if (m_lastPosition.get())
> +	   return m_lastPosition.get();
Is this caching required for correct behaviour or is it just an optimisation?
GeolocationController::lastPosition() should only be called once the client has
called GeolocationController::positionChanged(), and it seems that the client
should always have a last position available by this time.