Sylvain, so you intend something like the attached list ? I went in the first level. The procedure until now is a quick and dirty hack .. I can convert this one into percentage, then the results are more immediate. Once we have a stable procedure I can apply this on all log files and we have good statistics. This gives also as side effect a list of "strange" URLs we could analyze of what happened there.
cheers Reinhard /viewvc/ 152705 /viewcvs-doc/ 25531 /r/ 11144 /viewcvs/ 4913 /gitweb/ 1698 /cgi-bin/ 104 /bin/ 99 /admin/ 62 /cvsweb/ 56 /philosophy/ 48 /viewcvs.cgi/ 30 /graphics/ 29 /software/ 24 /includes/ 21 /sources/ 19 /icons/ 17 /components/ 14 /cvsweb.cgi/ 14 /cvsweb.pl/ 14 /administrator/ 12 /mhonarc/ 12 /gnu/ 11 /licenses/ 11 /modules/ 11 /src/ 11 /_vti_bin/ 9 /wwwes/ 9 /bbs/ 8 /server/ 8 /!scanhttp:/ 4 /cvs.savannah.nongnu.org/ 4 /fun/ 4 /phpgwapi/ 4 /alex_guestbook3/ 3 /include/ 3 /prep/ 3 /support/ 2 /app/ 2 /cvsroot/ 2 /ganesha/ 2 /ganesha2/ 2 /lib/ 2 /library/ 2 /oneadmin/ 2 /phpWebLog/ 2 /projects/ 2 /svn/ 2 /themes/ 2 /%22Warning:http:/ 1 /~papresco/ 1 /admin_modules/ 1 /b2-tools/ 1 /beacon/ 1 /directory/ 1 /freetype/ 1 /gnuschool/ 1 /help/ 1 /jobs/ 1 /libaffa/ 1 /music/ 1 /news/ 1 /olbookmarks-0.7.4/ 1 /patent-examp/ 1 /php.incs/ 1 /pngpp/ 1 /resources/ 1 /spanish/ 1 /testimonials/ 1 /v/ 1 /viewvcs/ 1 /yabbse/ 1 /zentrack/ 1
