tlug: namazu update

[Tony Laszlo <laszlo@example.com>]

I am writing this for anyone who may 
think of using namazu in the future. 

Couple of good things can be said for 
this program. The readout that it sends 
to stdout is quite nice. Also, when 
updating the index, it doesn't start 
from scratch, unlike glimpse. And, of 
course, it can handle Japanese data, 
unlike the versions of glimpse that 
I have seen.  

However, there are at least two reasons 
to steer away from namazu. What took 
glimpse 6 hours to index is taking 
namazu 24+ (still going). 
Also, I found that, when running the 
program without recompiling, it does 
not index files bigger than 500K (not 
5M, as I had earlier though). I recompiled 
the program, allowing it to use more 
memory and index larger files. The 
program does this happily for a couple 
of hours and then seg faults. This is, 
as you can imagine, very frustrating. 
OK, I saw a few errors generated 
when I recompiled the program. That 
may be behind the seg faults. someone 
with more know-how may be able to avoid 
this pitfall. I got fed up and 
decided to split all these megabyte 
grade data files into 500K pieces. 
Now using the program as it was 
originally compiled. 
no seg faults yet, after just 1333 
minutes (!) . 
 

Tony Laszlo



--------------------------------------------------------------------
Next Nomikai Meeting: February 18 (Fri) 19:00 Tengu TokyoEkiMae
Next Technical Meeting:  March 11 (Sat) 13:00 Temple University Japan
* Topic: TBD
--------------------------------------------------------------------
more info: http://www.tlug.gr.jp        Sponsor: Global Online Japan