In message <567bd604.7050...@smoothartist.com>
Michael Drake <m...@smoothartist.com> wrote:
>Hi Dave,
>
>I replied on IRC, but from your subsequent comments, you didn't see that.
>
>On 30/11/15 22:15, Dave Higton wrote:
>> I also have some questions:
>>
>> 1) Are the first two additions necessary or desirable?
>
>They shouldn't be needed but should do no harm. When you call
>browser_window_search the state of those buttons should be set as
>appropriate via callbacks anyway.
>
>> 2) Do other platforms have the same problem?
>
>The core search implementation is a bit broken, and that is where the
>fix belongs. However I'm not really interested in fixing it at the
>moment, since the core search code will be thrown away when we do the
>new layout engine.
>
>If your workaround improves things on RISC OS, then I don't mind
>applying it.
I'm happy that this works as well as can be expected, and gives
a satisfactory solution to issue 2339.
The revised patch (I removed the two redundant statements) is
attached.
Thank you!
Dave
____________________________________________________________
Can't remember your password? Do you need a strong and secure password?
Use Password manager! It stores your passwords & protects your account.
Check it out at http://mysecurelogon.com/manager
diff --git a/riscos/search.c b/riscos/search.c
index 60b3716..9c438bf 100644
--- a/riscos/search.c
+++ b/riscos/search.c
@@ -173,6 +173,8 @@ bool ro_gui_search_click(wimp_pointer *pointer)
case ICON_SEARCH_CASE_SENSITIVE:
flags = SEARCH_FLAG_FORWARDS |
ro_gui_search_update_flags();
+ browser_window_search_clear(
+ search_data.search_window);
browser_window_search(search_data.search_window, NULL,
flags,
ro_gui_get_icon_string(dialog_search,
