On 06/23/15 06:14, Craig Skinner wrote: > On 2015-06-22 Mon 12:39 PM |, Noah wrote: >> On Mon, Jun 22, 2015 at 11:58 AM, Craig Skinner <skin...@britvault.co.uk> >> wrote: >> > >> > *) either/both .txt/.html >> > *) .txt output something like: ls [-l[h]] | fgrep -v index.txt >> > >> >> Does auto index do the trick? It doesn't make an index.html/txt file, but >> it does provide file names and links as you'd expect. >> > > Ummm.... I was thinking of something that could generate $RELEASE > index.txt files, including siteXX.tgz & siteXX-<hostname>.tgz files.
oh please no. That is NOT the web server's job! This is a task easily -- and properly! -- done outside the web server. Your script is the right idea, but it should be run whenever whatever updates the contents of the directory runs, rather than periodically from cron. Even if the idea of generating a /very/ special purpose file in the web server wasn't a bad idea all around, you will note that this also violates the design of the chrooted webserver -- not only do you now have the webserver CREATING files, you have it doing it in the content area. (actually, if you are intent on doing this wrong, you might be able to use the "location" key word to call a CGI script when you try to fetch index.txt...but again, this is (in my opinion) the wrong way to do THIS task) Nick.
