PR/51518: Jay West: prompt is interleaved with client output
Both these issues are caused by rl_restore_handler not DTRT; fix
it so that it kills the internal libedit state completely. This is
inefficient, but it works.
Also fix:
1. add append_history()/H_NSAVE_FP
2. call the rl_startup_hook before printing the first prompt as documented.
callint it from rl_initialize breaks python, because the callback ends
up being invoked before the readline module is installed, and we end up
dereferencing a NULL pointer.
3. add el_resize_terminal.
With those changes, s/lreadline/ledit/g in python works.
- Put the data type el_rfunc_t into the public header <histedit.h>.
- Make el_read in struct editline an opaque pointer rather
than an embedded struct.
- Do not include "read.h" everywhere, but only in the two files
needing access to el_read, read.c and el.c.
- To functions that don't need more, pass the struct el_read_t *
rather than the full EditLine *.
- Of course, that means that read_init() can now fail from
memory exhaustion, but it's easy to clean up after that.
Pass in loads of parameters instead of relying on shed-loads of global
variables to modify the behaviour.
The filename completion code can now be enabled by code that uses el_gets().
(eg /bin/sh)
* Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str()
publically available in <histedit.h>
* Documented the public functions in editline(3)
* Renamed tok_line() -> tok_str()
* Added new tok_line() which takes a "const LineInfo *" instead of
"const char *" (the former has "cursor" information), and optionally
return the argv index ("int *cursorc") and offset within that index
("int *cursorv"). This means that completion routines can use the
tokenization code to crack the line and easily find which word the
cursor is at. (mmm, context sensitive completion :)
* Fixed TEST/test.c when using "continuation" lines (unmatched quote
or \ at EOL), and added some more DEBUG messages including highlighting
where the cursor is (with a `_').
The posix 'sh' specification defines vi-mode editing quite tightly.
The netbsd libedit code (used by sh to do this) was missing several
features, there were also minor errors in others.
Compare netbsd sh to the definition available from:
http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html
In particular the following were not supported:
U - undo all changes to line
| - goto column
Y - yank to end of line
y - yank
# - comment out current line
@ - take input from shell alias [1]
G - goto numbered line in history buffer
v - edit history line with vi
_ - append word from last input line
. - redo last command
Other minor changes have also been made.
[1] This needs the shell to define an appropriate routine to
return the text of the alias. There is no requirement that
such a function exist.
I wrote for tcsh(1) three years ago.
- implement EL_RPROMPT, which allows a setting/getting of a function which
returns a string to be used as the right-side prompt.
- improve HISTORY and AUTHORS sections in editline(3).
- bump shlib minor version for EL_RPROMPT.
XXX: due to an implementation issue, the rprompt has a 1 space gap before the
edge of the logical screen. editline's logical screen is 1 space less
than the full screen width, so there's a 2 space gap between the rprompt
and the right end of the physical screen. i'm not concerned about this.
* implement el_get(EditLine *, int op, void *result), which does the
inverse of el_set()
* add EL_EDITMODE operation to el_set and el_get; if non zero editing
is enabled (the default).
* add "edit on | off" editrc command, which modifies EL_EDITMODE.
users can now add '*:edit off' in ~/.editrc as an advisory to
disable editing.
NOTE: at this time EL_EDITMODE is just an indication of the
state of the 'edit' command. It's up to the application to check
this after el_source() or el_parse() to determine if editing is still
required.
christos