On Tue, Sep 07, 2004 at 02:41:02AM +0300, Amir Hardon wrote: > After Didi gave me the solution for the encoding problem with unzip I decided > to write a script for converting the filenames (Maybe I'll patch unzip in the > future but that's my temporary solution). > > The script I wrote just change the filenames and directories names using > iconv, but if it tries to handle an already iso8859-8 encoded filename iconv > gives an error (The validity check fails, it expects the input string to be > encoded as iso8859-1 but it is not). > Is there any standard unix tool I can use to first check the string's encoding > and then decide whether to convert it or not?
You do realize this is pure guesswork, do you? Noone can tell you the correct answer for sure. That said, a tool that tries to do this is called 'enca'. I never tried it myself, though. If all you want is to know if a given string is iso8859-1 or 8859-8 than I would personally simply do a range check - Hebrew should have bytes below 127 or between 224 and 250. Note!!! That this isn't accurate, and I'll never do this on a large mission critical tree without a good backup. > > Thanks, > -Amir. -- Didi ================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
