架设网站的apache日志使用cronolog进行轮循或者能够获得的时间做为有规律文件名：
/home/apache/logs/access_log.20040415
/home/apache/logs/access_log.20040416
/home/apache/logs/access_log.20040417
/home/apache/logs/access_log.20040418
#!/bin/sh
#$Id: spider_stats.sh,v 1.9 2004/05/15 16:52:44 chedong Exp $
YESTERDAY=`date -d yesterday +%Y%m%d`
# for FreeBSD: YESTERDAY=`date -v-1d +%Y%m%d`
THISMONTH=`date -d yesterday +%m%Y`
LOG_FILE=’/home/apache/logs/access_log’
grep -i Googlebot $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort | uniq -c | sort -rn > spider/$YESTERDAY.googlebot.txt
grep -i baiduspider $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort | uniq -c | sort -rn>spider/$YESTERDAY.baiduspider.txt
grep -i msnbot $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort | uniq -c | sort -rn>spider/$YESTERDAY.msnbot.txt
grep -i slurp $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort | uniq -c | sort -rn>spider/$YESTERDAY.inktomi.txt
grep -i openbot $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort |uniq -c | sort -rn>spider/$YESTERDAY.openbot.txt
# for search entry stats
grep -i www.google.com/search $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort | uniq -c | sort -rn > search/$YESTERDAY.google.txt
grep -i www.baidu.com/baidu $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort | uniq -c | sort -rn > search/$YESTERDAY.baidu.txt
grep -i 3721.com $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort | uniq -c | sort -rn > search/$YESTERDAY.3721.txt
grep -i search.sohu.com $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort | uniq -c | sort -rn > search/$YESTERDAY.sohu.txt
grep -i search.sina.com.cn $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort |uniq -c | sort -rn > search/$YESTERDAY.sina.txt
grep -i search.yahoo.com $LOG_FILE.$YESTERDAY|awk ’{print $7}’ |sort |uniq -c | sort -rn > search/$YESTERDAY.yahoo.txt
# for search keywords stats
grep www.baidu.com/baidu $LOG_FILE.$YESTERDAY | awk ’{print $11}’ | perl -pe ’s/\\x(\w+)/%\1/gi’ |perl -p -e ’s/%(..)/pack("c", hex($1))/eg’ | perl -pe ’s/(.*)?(word=(.*?))[&"].*/$3/gi’ |sort|uniq -c|sort -rn > keywords/$YESTERDAY.baidu.txt
grep www.google.com/search $LOG_FILE.$YESTERDAY | awk ’{print $11}’ | perl -pe ’s/\\x(\w+)/%\1/gi’ |perl -p -e ’s/%(..)/pack("c", hex($1))/eg’|perl -pe ’s/(.*)?(q=(.*?))[&"].*/$3/gi’ |sort|uniq -c|sort -rn > keywords/$YESTERDAY.google.txt
grep 3721.com $LOG_FILE.$YESTERDAY | awk ’{print $11}’| perl -pe ’s/\\x(\w+)/%\1/gi’ |perl -p -e ’s/%(..)/pack("c", hex($1))/eg’|perl -pe ’s/(.*)?((p|name)=(.*?))[&"].*/$3/gi’ |sort|uniq -c|sort -rn > keywords/$YESTERDAY.3721.txt
grep search.sohu.com $LOG_FILE.$YESTERDAY | awk ’{print $11}’| perl -pe ’s/\\x(\w+)/%\1/gi’ |perl -p -e ’s/%(..)/pack("c", hex($1))/eg’|perl -pe ’s/(.*)?((key_word|word)=(.*?))[&"].*/$3/gi’ |sort|uniq -c|sort -rn > keywords/$YESTERDAY.sohu.txt
grep search.sina.com.cn $LOG_FILE.$YESTERDAY | awk ’{print $11}’| perl -pe ’s/\\x(\w+)/%\1/gi’ |perl -p -e ’s/%(..)/pack("c", hex($1))/eg’|perl -pe ’s/(.*)?((_searchkey|word)=(.*?))[&"].*/$3/gi’ |sort|uniq -c|sort -rn > keywords/$YESTERDAY.sina.txt
grep search.yahoo.com $LOG_FILE.$YESTERDAY | awk ’{print $11}’| perl -pe ’s/\\x(\w+)/%\1/gi’ |perl -p -e ’s/%(..)/pack("c", hex($1))/eg’|perl -pe ’s/(.*)?(p=(.*?))[&"].*/$3/gi’ |sort|uniq -c|sort -rn > keywords/$YESTERDAY.yahoo.txt
perl -pe ’s/\\x(\w+)/%\1/gi’ : 用于转换： \xe4\x23 这样的转码
perl -p -e ’s/%(..)/pack("c", hex($1))/eg’ : 进行UrlDecode
sort|uniq -c|sort -rn : 用于排序，汇聚计数并按照次数排序输出
