Hi,
some time ago I encountered the same stupid lines in /etc/inputrc also
on a red hat linux system. I think it is simply a typo and was
originally meant to mean
if ( NOT -f $HOME/.inputrc ) then
setenv INPUTRC /etc/inputrc
endif
(I don't know the proper csh syntax for this)
this way, it could set the readline library to some reasonable default
in /etc/inputrc if the user does not have his own .inputrc file.
regards Thorsten
Mike Miller wrote:
On Sun, 16 Oct 2005, John W. Eaton wrote:
Looking at the definition of
rl_read_init_file in bind.c from
readline, I see
/* Do key bindings from a file. If FILENAME is NULL it defaults
to the first non-null filename from this list:
1. the filename used for the previous call
2. the value of the shell variable `INPUTRC'
3. ~/.inputrc
If the file existed and could be opened and read, 0 is returned,
otherwise errno is returned. */
int
rl_read_init_file (filename)
const char *filename;
{
/* Default the filename. */
if (filename == 0)
{
filename = last_readline_init_file;
if (filename == 0)
filename = sh_get_env_value ("INPUTRC");
if (filename == 0)
filename = DEFAULT_INPUTRC;
}
if (*filename == 0)
filename = DEFAULT_INPUTRC;
#if defined (__MSDOS__)
if (_rl_read_init_file (filename, 0) == 0)
return 0;
filename = "~/_inputrc";
#endif
return (_rl_read_init_file (filename, 0));
}
This function is called with filename == NULL from
readline_initialize_everything, which is called the first time
readline is called. My guess is you have INPUTRC set to
"/etc/inputrc" in your environment . Maybe that is done in
/etc/profile on your system?
I had assumed that readline was looking in /etc/inputrc then
~/.inputrc, but that's apparently not what happens. Instead, it is
only reading one file at startup time.
It looks like you figured this one out too, John! I'm using tcsh on
Red Hat Linux and this is in the default /etc/csh.login file that came
with the system:
if ( -f $HOME/.inputrc ) then
setenv INPUTRC /etc/inputrc
endif
I guess I don't understand the reason for that section because it looks
like it is saying that if ~/.inputrc is a plain file, the environment
variable INPUTRC should be set to /etc/inputrc so that ~/.inputrc will
not be read.
This does explain something strange that I noticed before. When I
installed a ~/.inputrc earlier, it worked fine. Later, after the
machine was rebooted due to a power failure, I installed the new Octave
and it did not use the ~/.inputrc, but neither did the old Octave. How
could that be? Well, restarting meant sourcing the /etc/csh.login file
which then detected ~/.inputrc and set INPUTRC to /etc/inputrc. It
seems crazy.
Can anyone see the need for that section of /etc/csh.login above?
Anyway, tcsh is good but I'm working on switching everything to bash
now.
Best,
Mike
-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.
Octave's home on the web: http://www.octave.org
How to fund new projects: http://www.octave.org/funding.html
Subscription information: http://www.octave.org/archive.html
-------------------------------------------------------------
|