monline

Tom has written a pretty nice article on how to avoid the (in)famous "ORA-04091 table xy is mutating, trigger/function may not see it" errors. It's worth to read if you don't know the answer by yourself.

I came upon this article by searching on the net for the conditions where you cannot/mustn't query the parent table in a trigger (I always forget that you can do this only in "before insert" row level triggers ... or in statement level triggers of course).

I use this very rarely so I always forget it again and again. You can list all the subdomains of eg. a top level domain like this ...

1. host -t NS example.com
   Find a nameserver in the output of the DNS lookup.
2. host -t AXFR example.com <name_server>
   where <name_server> comes from the first step.

Of course most of the DNS servers deny such requests ... so don't be surprised if you get a "Host example.com not found: 5(REFUSED)" error in the second step (or optionally you do not get any response at all).

Today I was confronted with the fact that Windows XP does not allow one to format drives larger than 32GB with a FAT/FAT-32 filesystem.

However my new Vosonic VP8360 PSD (Photo Storage Device) supports only FAT. I've tried my luck wth Google and came upon "Linux mkdosfs for Windows NT/2K/XP".

This is a Windows port of the Linux mkdosfs command and makes up for XP's handicap.

PS: of course I can format the PSD using its builtin format option ... I just wanted to see whether I could format a drive as FAT if chose to.

People tend to stick a "This is valid XHTML" logo on their site (probably a trend today?), however in 99% of all cases the page does not validate as a standard XHTML document. They even used to put a link on the page pointing to http://validator.w3.org/check/referer and if you click to check, you get at least half a dozen errors (or warnings in the best case). Where's the sense in that? I've got that link too, but pages on this site truely validate as XHTML Strict.

IMDb, PORT.hu

A film egy elválaszthatatlan testvérpárról szól: Rose csendes, visszafogott, "jó" kislány, míg huga Maggie az örök bajkeverő, számára az élet csupa szórakozás. Kettejük kapcsolata, mint tűz és víz, de ennek ellenére felbonthatatlannak látszik.

This is a feature introduced in Oracle 9i. The point is that the RDBMS can cache results of subqueries and provide a significant performance gain in certain cases.

I was quite thrilled, when I first read about the "DETERMINISTIC" compiler hint introduced in 8i. Unfortunately it works not as one would hope/expect.

The tricky part here is that the default UI language is determined by a setting in a file that is not present by default in the OO directory.
It should be at the following path: <OO_DIR>\share\registry\data\org\openoffice\Office\Linguistic.xcu

The name of the setting is UILocale in the General section. It's easier if you first start OO with any user, set the UI language and default locale in OO itself, then copy the Linguistic.xcu file from the user's OO settings folder ("<profile_base>\Application Data\OpenOffice.org2\user\registry\data\org\openoffice\Office\Linguistic.xcu") to the previously described path in the OO directory. Now you can edit this Linguistic.xcu to include only those settings that you want to have a default for all users.

One would think that finding langpacks for the popular OpenOffice.org office suite takes just a few clicks with Google. But it does not! I had to do quite some digging to find oootranslation.services.openoffice.org at last. Another good source is ftp.funet.fi.

Most láttam egy rövid reklámot, mely a Dove által létrehozott ún. Önbecsülési Alapot népszerűsítette. Eme kezdeményezés a fiatal lányok és nők önértékelési problémáin hivatott segíteni.