On Fri, May 13, 2005 at 12:18:04PM +1000, [EMAIL PROTECTED] wrote: > Can this be right ? I checked out DBD-mysql-2.9007 and ran it through > Devel::Cover. Apart from skipping 15 tests to do with leaks and 1 test > to do with transactions, the overall coverage figure from Devel::Cover > is 56%
<snip> > ----------------------------------- ------ ------ ------ ------ ------ > ------ > File stmt branch cond sub time > total > ----------------------------------- ------ ------ ------ ------ ------ > ------ > blib/lib/DBD/mysql.pm 71.9 42.3 38.6 75.9 12.0 > 60.1 > blib/lib/DBD/mysql/GetInfo.pm 70.6 n/a n/a 62.5 0.9 > 68.0 > blib/lib/Mysql.pm 67.3 42.3 40.0 59.5 82.6 > 60.2 > blib/lib/Mysql/Statement.pm 38.5 36.1 0.0 70.8 4.5 > 40.3 > Total 62.5 41.1 33.8 67.3 100.0 > 56.0 > ----------------------------------- ------ ------ ------ ------ ------ > ------ The "total" produced by Devel::Cover can be deceiving. Its a simple average and not taking into account things like the fact that Mysql::Statement is 171 lines while DBD::mysql is 1753. So Mysql::Statement's poor coverage is having a much greater effect on the apparent coverage than it really should. I'm also not sure what voodoo Devel::Cover uses to create a "total coverage" out of four different types of coverage stats. That said, removing it only brings things up to about 63%. And yes, its entirely possible it is that badly covered. Posting the coverage database would help to determine what's going on better, but its not unlikely the tests just aren't that great. The good news from your perspective is you can do a lot of good with a little effort. Hop to it, hopilte! :) -- Michael G Schwern [EMAIL PROTECTED] http://www.pobox.com/~schwern Just call me 'Moron Sugar'. http://www.somethingpositive.net/sp05182002.shtml