Hi Mark, since this was my issue a couple of weeks (or is it months?) ago, I might get back on this.
Actually it is a bug in the way, the row and column is calculated, leading to strange stretching of the image. I corrected the bug in my code, recompiled it and now it works fine. I sent an e-mail to the dev-list with the modifications I did. However, I did not follow, if the changes found their way into the current beta release. Cheers, Hannes PS: I sent the e-mail with the bug correction on 12 Feb 2011 17:42 - I can send it again, if required PPS: So it is "months" after all ... Am 15.04.2011 13:42, schrieb Mark Beardsley: > Some time ago, another post exactly like this one surfaced and we found that > the problem was with the anchor. It actually 'attaches' for want of a better > word, the bottom left hand corner of the image to the top left hand corner > of an adjacent cell. To explain this a little better, imagine that you > inserted an image into cell A1, then the bottom left hand corner of that > image would actually be 'attached' - again for waht of a better word - to > the top left hand corner of cell B2. This allows Excel to stretch the image > if the width of column A is increased or the heigth of row one is likewise > increased. It also seemed to be the reason why the image did not move when > the data was sorted - it was anchored to the top left hand corners of two > cells and Excel did not know how to handle this problem. The solution seemed > to be to change the way the images anchor was created so that you specified > the width and height of the image using Excels co-ordinate system; and I > cannot remember off hand exactly what these values were and so need to look > up some old code. When I get the full information, I will post again later. > > The test I conducted to get to this answer was simple. I used POI to insert > images into cells just as you did and confirmed the 'sort proble'. Next, I > simply used Excel to modify the sizes of the images so that they no longer > overlapped the edges of the cell and found that the sort problem was > resolved. It might be worthwhile repeating this test yourself just to make > sure my recollection is correct. --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
