Fernando

I really shouldn't have wrote that so late at night, I sent you the wrong file! This is the correct one: http://www.tesouro.fazenda.gov.br/documents/10180/327844/pge_finbra_2010.zip/4560c066-10a1-483c-9067-7f1bb60b4152 That other one may have the issue or not, I haven't really tested it. This one is the one I've been testing.

I really have no idea why it takes so much time, but I would be glad if you could take a look at it :)

Hello! Again, dealing with a government database, it seems like they put in a very nefarious view into the database. It contains a SUM of every field possible, and for some reason, when the driver tries to execute that (presumably to get the metadata), it takes absolutely forever. My current fix for now is to just disable views, as they aren't currently needed, until we can get better performance on it. The database in question is the one available under this link: http://www.tesouro.fazenda.gov.br/documents/10180/137713/Finbra2002v0809.zip...

Hello! I've came across a government database that used columns with commas. It seems like UCanAccess doesn't quote them properly. I took a look at the code and the preEscapingIdentifier function doesn't bat an eye at commas! I'm seeing if just monkey patching this function to transform commas to underscores will work, but I would like this to be fixed in the next version :)