On Sun, Oct 07, 2001 at 05:31:27PM -0400, Glenn F. Maynard wrote:
Easy enough: changed Combine(1) to Combine(0).
Change this back; it breaks SITE. I'll need to do it manually, I guess,
or combine differently for the different CMD(ls) types.
I did this:
--- commands.cc 2001/10/08 05:50:54
On Fri, Oct 05, 2001 at 03:09:44PM -0400, Glenn F. Maynard wrote:
Case-insensitive matching for non-filenames (ie. commands) would be
nice to have all the time, actually. (Why should commands be
case-sensitive?)
Religious matter :) Even if they were case insensitive, who would press Shift
Patch applied, BTW. Thanks!
--
Alexander.
A note about last patch:
GLOB_PERIOD is not defined on solaris. It is a GNU extension.
--
Alexander.
On Thu, Oct 04, 2001 at 11:26:03PM -0400, Glenn F. Maynard wrote:
I've moved both the fileset printing (moved it all into the printopts
structure, and renamed it FileSetPrint) and the FileCopyJob into
FileSetOutput.cc/FileSetOutput.h. So, the FileCopyJob stuff is in this
one, too; along with
On Fri, Oct 05, 2001 at 10:47:46AM +0400, Alexander V. Lukyanov wrote:
I have applied this patch and fixed two bugs I noticed:
* ColumnOutput.cc: fix memory leak.
* FileSetOutput.cc: fix coredump when LS_COLORS is unset.
Oops. Thanks.
BTW, I have noticed that cls -l does not
On Fri, Oct 05, 2001 at 03:03:36AM -0400, Glenn F. Maynard wrote:
I'll update my tree and get the last round going--completion.
Reupdated my address (got lost in the shuffle.)
Add cls, recls and rels to list of remote-completers.
Make local completion mimic remote completion by forming a
On Fri, Oct 05, 2001 at 04:00:51AM -0400, Glenn F. Maynard wrote:
Keep glob_res around after completion until the next completion or the
next command; this lets us use it to do our own completion result
display.
Maybe there is a callback in readline? If not, I think it is better
to call
On Fri, Oct 05, 2001 at 02:31:24PM +0400, Alexander V. Lukyanov wrote:
I think completion-ignore-case is only useful on windows.
Well, case-insensitive matching is just as useful in general in Unix.
For filenames, it's more useful in Windows, but CI matching for
filenames works fine in Unix.
On Thu, Oct 04, 2001 at 12:45:13PM +0400, Alexander V. Lukyanov wrote:
Don't give ls's job a blank name; results in unsightly empty `' in
status.
I'm not sure about this change. Maybe prepend `ls' always?
Easy enough: changed Combine(1) to Combine(0).
I've moved both the fileset printing
On Tue, Oct 02, 2001 at 10:22:59AM +0400, Alexander V. Lukyanov wrote:
What's the purpose of this?
+ ArgV arg(argc, argv);
Er, it doesn't seem to have one. Delete it.
misc.cc code to do the real work. (This could possibly be improved
to make it usable for the main tokenizer.)
I
On Sat, Sep 15, 2001 at 11:59:25PM -0400, Glenn F. Maynard wrote:
Originally I just output to an FD, honoring parent-output.
My code right now sets up a FileCopyPeer job to buffer output;
Look at how original ls is done. There is data source - FileCopyPeerDirList
and data destination which is
On Fri, Sep 14, 2001 at 06:58:51PM -0400, Glenn F. Maynard wrote:
There's a few extremely useful features of ls that would be great
to emulate in ls, and possibly file completion:
column width optimization
colorization
type suffixes
Agree, it would be nice to have an optional ls which
Originally I just output to an FD, honoring parent-output.
My code right now sets up a FileCopyPeer job to buffer output;
I'm not sure this is needed, and I had to do some modifications
to CmdExec I'm not sure about to get it working cleanly. Is it
OK to assume output is a local file or tty, and
There's a few extremely useful features of ls that would be great
to emulate in ls, and possibly file completion:
column width optimization
colorization
type suffixes
Width optimization is a generally useful feature, and would work for
all completions (though it might take some work to prevent
15 matches
Mail list logo
