| || |
Robot with Japanese web interface
I've been having some problems using Robot with a Japanese web-based application. When I get one of the innerText or innerHtml properties of an html object on a page (e.g. innerText of an HTMLTable) with SQAGetPropertyAsString or with Inspector, some Japanese characters appear corrupted. They do not appear as all question marks, but appear to be interpreted with the wrong text encoding method. All of the pages on our web interface are encoded as UTF-8, but Robot seems to try to read parts of some pages as Shift-JIS; if I force Internet Explorer to view the page as Shift-JIS, it matches what Robot seems in the corrupted areas. Even more frustrating: the problem shows up inconsistently; the innerText of one HTMLTable on a page will be read correctly, but the innerText of another HTMLTable on the same page will be corrupted. Does anyone have any idea what could be causing this? Is there any way to tell Robot what encoding method to use?
[ 06-16-2004, 08:38 AM: Message edited by: Michael LS ]