Gérard Milmeister wrote: >>> Meta-data (using extended file attributes) would be faster, since it >>> wouldn't involve opening the file and trying to guess the file type from >>> the file content. If you have selinux enabled, each file access involves >>> accessing extended file attributes. You still have to read that extra information, a read is a read. And now you're dependent on the filing system. File data should be part of the file itself. Mikkel L. Ellertson: > One way would be to have a utility that would go through the files, > determine the mime-type, and set the attributes. Run it as a nightly > cron job to index new files. Ugh, rather than store data in a sane way in the first place, do more work on top of it, repeatedly. I'm reminded of mail programs that continually re-index their mail folders, rather than update their indexes when things get changed, leaving them alone in the meantime. -- (This box runs FC6, my others run FC4 & FC5, in case that's important to the thread.) Don't send private replies to my address, the mailbox is ignored. I read messages from the public lists.
