I found a bug on the view implementation.

According to the manual some view is updatable if the view satisfies the “one-to-one relationship” between the rows in the view definition and the rows in the underlying tables. To be more specific a view is not updatable if it contains any of the following:

• aggregate functions (AVG(), SUM(), MIN(), …)
• DISTINCT
• GROUP BY
• HAVING
• UNION (ALL)
• subquery in select list
• JOIN (with some exception)
• non-updatable view in the FROM clause
• algorithm=temptable
• refers only to literal values

So, let’s try this simple example:


mysql> create table t1 (a int);
Query OK, 0 rows affected (0.21 sec)



mysql> create view v1 as select a from t1;
Query OK, 0 rows affected (0.00 sec) …

Last week while visiting Lars I had a moment of embarrassment. While going over my system of "Getting Things Done" with Lars, I revealed that my inboxes were huge.

On the flight back home I took what was once thousands of pieces of email divided into two email account (one my MySQL account, and the other Tangent) and got my inboxes down to under 150 total. I did this through a combination of "delete this entire mailing list" and "read it all".

So how did I have such a large email inbox? A few things I noticed while deleting the email:

1) There were emails in my inbox I should have replied too. Most email I can answer in minutes, anything else gets deleted and added to my to-do list. It looks like I will occasionally make an assumption of "this will take just a couple of minutes" get interrupted by a phone call, and never return to it. Not answering or deleting just two pieces of email a day, will leave …

A common problem MySQL users have with Crystal Reports 9 is the inability to browse tables and fields. While you can define a MySQL DSN, when you open it in Crystal Reports 9 you have no tables or fields in the field browser. This of course makes it very difficult to create a report.

Up until recently, the only two workarounds have been to either create the report in an older version of Crystal Reports and then open it in Crystal Reports 9, or use custom commands to return field data. Neither of these are particularly effective for creating reports.

This has been a known issue with Crystal Decisions and has now been solved. The full article is available in a knowledgebase article located at http://support.crystaldecisions.com/library/kbase/new_articles/c2013269.asp”

The short version is that you need to download …

If you like compiling from source then a collection of scripts may save you a considerable amount of effort, especially when it come to repeating the process when you upgrade.

OK, so I'm starting to get really excited about MySQL Camp in a couple weeks. Things are shaping up to be a fantastic event. Later today, I'll be cleaning up the SessionSchedule wiki page with more information on exact times and rooms for everything.

I will be arriving in San Jose airport on Thursday (November the 9th) evening at 7:30pm and will be getting a rental car at the airport (thanks Blue Chip program at Thrifty!). I'll be hanging out at the airport at the Internet cafe until 10pm so that I can give Baron Schwartz and Sheeri Kritzer a ride to the hotels in Palo Alto (we're staying at Motel 6, which AFAIK, still has available rooms). That leaves an extra seat in the car for another camper, so shoot me an email to …

I wish I could say that Europe is free from software patents, but it is not. In contravention of the existing statutory law, the European Patent Office (EPO) continues to bend and break the law every day by issuing European software patents. And if the EPLA were to be ratified in its current form or any form near the present proposal, tens of thousands of existing European software patents would become strongly enforceable overnight.

Earlier this year, the NoSoftwarePatents Award campaign started in order to draw attention to the EPO’s unacceptable granting practice. That award is not linked to the original NoSoftwarePatents.com campaign I founded in 2004 and handed to the FFII …

Saturn’s autoweb

I’ve hacked my scripts that generate doxygen docs to also build MySQL 4.1, 5.0 and 5.1 for AMD64 (the box that it’s running on) with Cluster. This is to help my idea of running Gallery at home with NDB disk data tables in very recent MySQL builds.

How’s it going so far? Well… I’ve found some bugs and some seemingly strange behaviour here and there. However, bug reports will come, and I’m currently running a bit of an older build.

I’ll make the URL of the Gallery public at some point too

Zend is hosting it's second annual Zend / PHP Conference & Expo on Oct 30 - Nov 2 in San Jose at the Doubletree Hotel.  The program has expanded from last year and includes a full day of tutorials on Monday.  The conference features in-depth technical sessions on topics such as high volume scaling techniques for PHP and MySQL, performance tuning, web services as well as sessions on new capabilities in forthcoming PHP 6. There are also keynotes, case studies and panel sessions.  Zend is one of the major contributors to PHP and they have also developed a great framework that greatly simplifies the …

mysqlreport v2.7 has been released. This version sports the following changes:

• Option “password” in ~/.my.cnf now recognized (used to be that only “pass” was recognized)
• Options host, port, and socket now recognized in ~/.my.cnf
• Given those two changes, the order of option precendece is such that ~/.my.cnf is read first, then command line options override ~/.my.cnf options, and if nothing else default values are assigned
• –all actually shows all reports now (i.e. no more need to –all –tab)
• Like mytop, mysqlreport tries to connect via a socket first, otherwise it tries via a network connection

Thank you to those who provided patches for these changes, although I wound up writing the mods myself for various reasons. To …

I’ve been writing a lot lately about Perl, MySQL and so forth, and neglecting another topic that interests me: the Web. I’m going to finish and publish some of the drafts I have on the Web before I continue with another massive database marathon. This one is about someone who read my mind and built exactly the Firefox extension I was about to build myself: an easy way to block all cookies and allow them on a case-by-case basis without nagging.