Community technical support mailing list was retired 2010 and replaced with a professional technical support team. For assistance please contact: Pre-sales Technical support via email to sales@march-hare.com.
Randy McCharles wrote: > We are doing cross-platform development for Windows, Mac, and Linux. > > We've run into a problem where our Unicode files are stored in CVSNT (2.0.51c) as UTF-8. > Windows and Mac will check these out fine into UTF-16 (Little Endian), but Linux does not recognize the BYTE ORDER MARK (U+FEFF) at the beginning of the file, causing the checked-out version to be incorrectly formatted. > > Is there a Linux client that will check out the CVSNT Unicode files correctly? Or do we need to store all Unicode files as binary? > (Currently we are doing all Linux Unicode check-in/out in Windows and porting the files, but this is becoming a productivity hit.) Yes, the linux cvsnt client will checkout in any format you want. 2.0.51c is extremely old - there have been a number of enhancements to the unicode support since (including the ability to optionally checkout without the BOM if necessary). Tony