Re: Update from remote - reflect in history file

[Larry Jones]

Dmitry Ryzhkov writes:
> 
> I can't find any way to register 'cvs update ..' command in the file if this
> update is requested from the remote machine.

This appears to be a long-standing bug in client/server mode and an
unintended consequence of setting the command_name global variable to
"server" whenever running in client/server mode rather than to the
actual command name.  There are already a number of places in the server
code where command_name is reset to the actual command name so that
things work right; I'm thinking we should just do it all the time.  The
major user-visible effect of this change would be that most messages in
client/server mode would contain the actual command name rather than the
generic "server".  It would also remove some gratuitous differences
between local and client/server modes (e.g., the "file was lost" warning
would show up in client/server mode).  Any objections?

-Larry Jones

Even my FRIENDS don't do what I want. -- Calvin