Deutsch

Current Texts Comic Imprint Calendar Search PHP-Classes Container-Wizard main.s21

Categories

Bad Attitude II

On the same day I had encountered Bad Attitude, I encountered even again, and this time, it's even worse. Not so much in what they actually do, but the situation and their "prey".

I made an update query for a database which was a little bit more complicated (not much) to correct some entries. As I was not quite sure whether it would really work out as expected, I put in a "LIMIT 1" to have a test with only one entry. PostreSQL didn't take that, so I did a quick lookup whether PostgreSQL did even support LIMIT in UPDATE queries.

It does not, as I found out in a discussion on their mailing list: http://archives.postgresql.org//pgsql-patches/2002-09/msg00255.php.

Now this is even worse, as the guy even submitted a patch. It is completely OK that they didn't accept it, because they're somewhat right that this is MySQL only behaviour which is not conforming to standards (but PostgreSQL got a lot of features which don't conform either).
But it would be very advised to treat that guy with some respect, because people who are capable of submitting patches to a database systems aren't that numerous. This is not some other PHP guestbook, but real C/CC++ stuff which requires a higher skill. That guy - and others - may have second thoughts on commiting another patch, some that would be welcomed more warmly. I mean, submitting a patch, even submitting a bug report, means that I spend my time to improve your project.

Bug reports are the next. I once reported a bug to Apache, which was about contradicting manual information about the build process on OS X. Actually, I have to say that I got a good answer because they told me that I was basically right, but could not expect it to be fixed anymore, because it was Apache 1.3.ancient. No problem.
I then found several glitches in PHP which I would consider to be a bug. The most serious one was a reproducable segmentation fault. A segfaulting PHP is especially evil when it happens to the Apache module, as it will just present a white screen. Actually, the segfault is triggered by a programming error in PHP, yet I classify it as a bug as I think that the interpreter should recognize it, such as it recognizes other errors. Debugging a script which produces a segfault can be very tedious as there is actually no way to debug it instead of selectively commenting out whole sections / includes.
Now I went to the PHP page, were they classified this Bug (among others I would have reported) as "Bogus" and quenched discussion with a short "refer to the manual".

As said in the predecessor, open communication is vital to Open Source. Well, it should be vital if one buys often made statements about Open Source, especially it's benefits and it's alleged superiority.

Comments

Please note: comments posted won't be visible immediately, as they will be checked for harmful content.