Hi Michael, > maybe I need to create a simple test case that can replicate this?
Without digging much into the problem, I'd say that creating a procedure to reproduce is always a helper. If you are working with the trunk version, you may want to take a look at the recently implemented window methods (bug 46072 [1]), for creating a self-contained test case using a single SVG file (or two files, counting with an external image). This would also be good as an exercise to confirm that the issue isn't in your application. ;-) I'd suggest something like: * A simple SVG embedding one (PNG?) image; * Use the "window.reload()" script method (which, AFAIK uses "setDocument" internally) to cause the document to reload (probably in the document's onload event or using a short timer); * Open it and watch for memory usage... If you want to cut the set (SVG+image) to a single file, you may also encode the image using a data URI (there are online converters available [2]). Of course this may bump other sources of memory leaks, so I'd propose that for an alternative/later version of the test case. Hope this helps, Helder [1] https://issues.apache.org/bugzilla/show_bug.cgi?id=46072 [2] http://scalora.org/projects/uriencoder/ --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
