To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=70244
------- Additional comments from [EMAIL PROTECTED] Tue Mar 11 10:31:42 +0000 2008 ------- @kohei: thanks for looking into this! When you look at the implementation of the method mentioned above you'll easily see what the problem is about: Not only are the available rows tested, but also all of the numerically possible rows. The method itself is not too expensive, but the cumulative cost of all the superfluous small text measurements can become significant. Since the cost of all these text measurements showed up prominently in a profile run it was originally seen as a performance problem in VCL, but a more detailed look revealed that this was not the case because of the root cause I outlined in this issues first comment. I'd rather not attach the document where I noticed the problem since I don't know the exact copyright status of it. I haven't looked into this problem for a long time, so it would cost me quite some effort to write another document which exhibits the issue. I'd probably do it though if you have too much trouble to get a document where ScDocument::GetNextDifferentChangedRow() is called indirectly while the document gets loaded. --------------------------------------------------------------------- Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
