On 2/2/07, Velko Hristov <[EMAIL PROTECTED]> wrote: > Axel, > > if you run "mimedb *fish" on the four files from the previous post > you'll get > > text/x-matlab > application/octet-stream > application/octet-stream > application/octet-stream > > Do you have any idea why the last three of them are not recognized as > text files?
Hmm, partially, yes. I made a mistake in my previous mail, which also explains why the mimedb command was so slow. The deafult behaviour for mimedb is to determine mimetype by looking at the filename if that is enough, and try to open the file and check for a magic cookie at the begining if that fails. One has to use the -f switch to force mimedb to only use the name of the file, which is much faster. At least on my system, using '-f' has the side effect of making all .fish have the same mimetype, namely application/octet-stream, which is the default mimetype. There is no entry in the mimetype database for fish files yet. For shame... My guess as to why you get those inconsistent results is that the code for checking for magic cookies is a bit optimistic and generates a few false positives. I don't know, really, mimedb internally relies on the xdgmime library for looking up mimetypes. > > Regards - Velko -- Axel ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Fish-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/fish-users
