Index: [Article Count Order] [Thread]

Date:  Fri, 28 Jun 2002 23:35:08 +0900
From:  Takayuki Matsuki <tmatsukijp@ybb.ne.jp>
Subject:  [analog-jp:01229] Re: SEARCHQUERY: Japanese displayed as hex problem
Sender:  tmatsukijp@ybb.ne.jp
To:  analog-jp@jp.analog.cx
Message-Id:  <mid-1229-analog-jp@jp.analog.cx>
X-Mail-Count: 01229

Thank you for the information!

>I guess you are using analog of ver.5.22 or later. Dr. Turner told me
>that he set as a default "SEARCHCHARCONVERT OFF" for analog when using
>multibyte character set, and even setting "SEARCHCHARCONVERT ON" this
>doesn't work. He does't remember why he coded like that way.
>
I am using the newest version 5.24. I tried as you mention using 
SEARCHARCHCONVERT (ON and OFF) and neither setting add any effect. I 
hope Dr. Turner can one day build language independent search query 
functionality.

>
>
>One solution is given by Tsuji Norihisa who gave a perl code which is
>on our home page; urldecode.pl in "Helper Applications and their
>instructions" section. 
>
It would't work because I am missing some Perl module. I need to figure 
out to get the missing Perl module.

>
>Another might be given by using logkf-1.0beta2.tar.gz which was given
>by Masashi Tachibana. This is also in the same section. I haven't
>tried yet though. This program is like the program nkf to convert any
>Japanese character codes in a log file into EUC as well as converting
>hex codes into EUC. You may specify the final output code other than
>EUC.
>
This program did not work at all. I ran logkf on my apache log file and 
it did not seem to do anything. I ran analog on the new output file from 
logkf but I still got the same results. I think this program is broken 
or assumes something about the OS or log file format that isn't true 
about my system.

Thanks for the help!

Jc