Pierangelo Masarati wrote:
Howard Chu wrote:
It sounds like you've now run into ITS#4638, a bug in the rwm overlay. It maps the filter for searching in the first database, but doesn't restore the original value before proceeding to the next database, so it's using the wrong filter there. I usually don't touch the rwm code since Ando mainly works on it, but can take a look later.
I'm not going to fix it in the very short term; however I will, at some point (could be right after Xmas or so).
I've patched this in HEAD. This should resolve ITS#4638 and #4689. Please test it and report back...