[cvsnt] Non-ascii characters go missing with update -p

Neil Roberts bpeeluk at yahoo.co.uk
Mon Jan 29 11:47:43 GMT 2007


I have a small problem using CVSNT regarding non-ascii characters. I
am using version 2.0.58d on a Linux box as the server and until
recently I have been using 2.0.51d (the one that comes with WinCVS) on
a Windows box as the client.

Some of the source code I work with uses non-ascii characters (such as
the copyright symbol) stored in the default Windows-1252 character
encoding (eg, not Unicode). This has been working fine until I
upgraded my client to 2.5.03. Now if I do an update using the -p
option to pipe to the standard out then all of the non-ascii
characters disappear in the output. It doesn't seem to have this
problem if I do a regular update or checkout.

This is a nuisance because WinCVS uses the -p option to do an external
diff, so all of the non-ascii characters appear as differences.

I will probably just downgrade back to the old version, but I was just
wondering if this is a known problem or if I'm doing something wrong?

Thanks for your help

- Neil

More information about the cvsnt mailing list