[19058] in s-news-athena
[S] Summary: using splus for web server log analysis
daemon@ATHENA.MIT.EDU (Don Turnbull)
Fri Aug 27 11:20:03 1999
Message-Id: <37C6A954.427065A2@fis.utoronto.ca>
Date: Fri, 27 Aug 1999 10:05:57 -0500
From: Don Turnbull <donturn@fis.utoronto.ca>
Reply-To: donturn@fis.utoronto.ca
Mime-Version: 1.0
To: "Herbert M. Bryant Jr." <herb.bryant@swfwmd.state.fl.us>,
"s-news@wubios.wustl.edu" <s-news@wubios.wustl.edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
"Herbert M. Bryant Jr." wrote:
> Mr. Turnbull-
>
> I'm curious about any answers you get to the attached question.
>
> -Herb Bryant
Unfortunately, I only got one response (and I guess that's enough to share -
shame on me) from the kind Dr. Boles:
here it is:
Don:
I am looking at .log files produced by a chat program and examining
text strings using the liwc (linguistic word count program). I
preprocess the files using a perl script to prepare the data for liwc.
I assume you could do something similar to preprocess the data into a
format that could be imported into splus via read.table or scan. Below
is the perl script I use. You could modify or roll your own. Perl is
_fast_ and handles large files.
...
===================================================================
#!/usr/bin/perl
## dnet program to:
# read log file produced by chat program *.log
# write time user name, and text to *.log.liwc.txt file
# only write lines with message text, including emoticon produced text
# write message out in multiple lines with hard carriage returns
#########################
$arg = $ARGV[0]; # get file & path name
# print $arg;
@args = split(/\//,$arg); #print "@args\n";
$num_args = (@args); #print "$num_args\n";
$logfile = $args[$#args];
if ($num_args > 1) { $path = join('/', @args[0 .. $#args -1])."/";}
else {$path= "./";}
# print "$path\n$logfile\n";
##########################
$logout = $path.$logfile.".liwc.txt";
print "Writing $logout \n from: $logfile\n";
# process logfile
open(LOGOUT,"> $logout") || die "cant open LOGOUT $!\n";
while(<>)
{ chop; s/\r//; # using ascii files so strip carriage return
$message= /^.* {.*} .+$/;
# print "$message\n";
if ($message == 1)
{ # only convert records with messages
($weekday,$month,$day,$time,$year,$user, $braces, $rest)=
/(^\S+) (\S+) (\S+) (\S+) (\S+) (\S+) ({.*}) (.*)/;
$out = "$weekday $month $day $time $year $user $rest";
print LOGOUT "$out\r\n" ; # write records with id info
# print "$out\r\n" ;
}
elsif (!/^(Mon|Tue|Wed|Thu|Fri|Sat|Sun).*{.*}$/)
{
print LOGOUT "$_\r\n";
# print "$_\r\n";
}
}
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
