diff options
| author | rowanbeentje <rowan@beent.je> | 2008-11-18 10:34:23 +0000 |
|---|---|---|
| committer | rowanbeentje <rowan@beent.je> | 2008-11-18 10:34:23 +0000 |
| commit | 7cb7092431a98c369993608f504318fbd28ef5a6 (patch) | |
| tree | 4e1ac27fba465c5797654ed2e0ddbdb40941fd82 /TableContent.m | |
| parent | c55412ca23dfba852f8bf4eae0c3249db32e90c8 (diff) | |
| download | sequelpro-7cb7092431a98c369993608f504318fbd28ef5a6.tar.gz sequelpro-7cb7092431a98c369993608f504318fbd28ef5a6.tar.bz2 sequelpro-7cb7092431a98c369993608f504318fbd28ef5a6.zip | |
Experimental change: store the table content viewport before a refresh, and then restore it afterwards - ie suggestion (2) in http://groups.google.com/group/sequel-pro/browse_frm/thread/71541c6267fc08c1/2f4aeeaa7f15b268 .
I was initially in two minds as to whether this would be a good thing, but with r195/r196 removing the need to refresh for structural changes, this change does actually feel right to me - in fact, I now wish it didn't clear any filters/limit, and I'd like to preserve them as well. Any objections?
Diffstat (limited to 'TableContent.m')
| -rw-r--r-- | TableContent.m | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/TableContent.m b/TableContent.m index 6c9c7341..8c72b724 100644 --- a/TableContent.m +++ b/TableContent.m @@ -319,7 +319,13 @@ reloads the table (performing a new mysql-query) */ { + // Store the current viewport location + NSRect viewRect = [tableContentView visibleRect]; + [self loadTable:selectedTable]; + + // Restore the viewport + [tableContentView scrollRectToVisible:viewRect]; } - (IBAction)reloadTableValues:(id)sender |