path: root/ksh
AgeCommit message (Collapse)AuthorFilesLines
2018-07-25Stamp down three new ShellCheck 0.5.0 errorsTom Ryder1-0/+1
All of these are false positives.
2018-06-29Break sh noglob guard into two lines for clarityTom Ryder1-1/+2
They're nicer to read this way.
2017-06-27Inline some switch casesTom Ryder1-3/+1
2017-06-19Remove user@ prefix from promptTom Ryder1-2/+2
2017-06-16Allow setting a prompt mode for the shellTom Ryder1-2/+2
2017-05-24Remove Yash supportTom Ryder1-1/+1
I never use it
2017-03-29Remove SC2154 ignore for $.sh from ksh startupTom Ryder1-1/+0
2017-02-17Don't re-set HISTSIZE in kshrcTom Ryder1-1/+0
Already set in shrc
2017-02-17Condense ENV_EXT hack, unset itTom Ryder1-4/+1
2017-02-17shellcheck suggested a more efficient assignmentTom Ryder1-1/+1
2017-01-28Correct a typeset callTom Ryder1-1/+1
2017-01-27Extend "fallback" prompt a bitTom Ryder1-1/+4
Include username/hostname if no SSH variables
2017-01-26Fixed prompt alignment hack for kshTom Ryder1-3/+3
The manual page for mksh hints that the escaped initial character should in fact be a carriage return, not a newline. That seems to work really well. The newline variable was an empty string before this commit anyway because it was stripped by the subshell expansion.
2017-01-25Show prompt prefix if a shell is exoticTom Ryder1-0/+14
That is, include e.g. "ksh:" as a prefix to the prompt if the user appears to have Bash or Zsh (or anything else) as their login shell. This is probably imperfect, but it's a start.
2017-01-25Be more specific in ref name trimmingTom Ryder1-1/+1
e.g. for a namespaced branch "foo/bar/bar", don't strip the leading part off
2017-01-24Remove SHLVL nested shell feature for nowTom Ryder1-8/+0
Causes a bit too much curliness in configuration; may need a more comprehensive approach.
2017-01-11Have ls() func undo aliases in its wayTom Ryder1-2/+0
2017-01-11Don't let OpenBSD alias `ls` in kshTom Ryder1-0/+2
2017-01-07Tolerate C-M-l for clear window in kshTom Ryder1-14/+3
Already works in ksh93 and mksh. Trapping C-l in ksh93 interferes with the builtin SIGWINCH handling, clearing the screen every time the window resizes, and I can't find a good way to work around it. Probably best not to fight this.
2017-01-06Rename keyboard trap func less ambiguouslyTom Ryder1-2/+2
2017-01-06Add tab completion settings to kshTom Ryder1-2/+8
2017-01-06Appease ShellCheckTom Ryder1-2/+3
2017-01-06Found workable ^L bindings in all three kshesTom Ryder1-8/+6
2017-01-06Clean binding of ^L in ksh93 to clear screenTom Ryder1-18/+7
2017-01-06Experimental/not-very-good-yet ksh bindingsTom Ryder1-0/+34
Just starting with what I know and seeing if I can make ^L work the same way it does in Bash. Once I understand this a bit better I intend to have a crack at writing some dynamic completion for ksh93.
2017-01-06Add "trackall" option to kshTom Ryder1-0/+3
2017-01-06Add a couple of ksh optionsTom Ryder1-1/+12
2016-12-30Remove unneeded typeset callTom Ryder1-1/+0
2016-12-28Deal with the kshes' varying treatment of ! in PS1Tom Ryder1-2/+22
2016-12-28Quiet a shellcheck errorTom Ryder1-0/+1
Like in 3828a1f, this variable actually does get set before use, but not in a place ShellCheck could be reasonably expected to find
2016-12-27Change quoting mechanismTom Ryder1-1/+1
2016-12-27Fix an issueTom Ryder1-1/+1
End the prompt subshell with a : to reset $?
2016-12-27Preserve job count quickly for kshTom Ryder1-3/+1
2016-12-27Tell ShellCheck to ignore an errorTom Ryder1-0/+1
ret actually is assigned, just not in any place ShellCheck could reasonably be expected to find
2016-12-27Realised easy win avoiding subshellsTom Ryder1-5/+1
And also the PROMPT_RETURN fudge in Bash/Zsh
2016-12-27Put exit value of previous command in ksh PS1Tom Ryder1-11/+7
Found a way to do this that seems to work in all the KSHes I've tried
2016-12-22Use more compatible checkTom Ryder1-1/+1
2016-12-22Limit keep() to ksh93Tom Ryder1-2/+9
FreeBSD pdksh doesn't even have complete typeset -p output, nor printf %q, which I think makes this impractical.
2016-12-22Use . not sourceTom Ryder1-1/+1
2016-12-22Fork keep() to kshTom Ryder1-0/+152
Mostly works.
2016-12-21Use clearer logic for prompt color settingTom Ryder1-8/+7
2016-12-19Attempt at properly escaping non-printing charsTom Ryder1-1/+6
This is supposed to be the equivalent of \[ \] in Bash PS1... >Note that since the command line editors try to figure out how long the >prompt is (so they know how far it is to edge of the screen), escape >codes in the prompt tend to mess things up. You can tell the shell not >to count certain sequences (such as escape codes) by prefixing your >prompt with a non-printing char- acter (such as control-A) followed by >a carriage return and then delimiting the escape codes with this >non-printing character. If you don't have any non-printing characters, >you're out of luck... BTW, don't blame me for this hack; it's in the >original ksh. <http://blog.0xpebbles.org/ksh-prompt-coloring-example>
2016-12-19Block a well-intended shellcheck errorTom Ryder1-0/+1
2016-12-19Request version with "ver" shortcut not promptTom Ryder2-12/+6
2016-12-19Attempt a much saner approach to managing SHLVLTom Ryder1-7/+4
2016-12-18Defer kshrc loading until after shrc all loadedTom Ryder1-5/+2
2016-12-18Add ksh version distinguisher to promptTom Ryder1-2/+12
2016-12-18Change tack; force ENV if the file existsTom Ryder1-6/+0
2016-12-18Change syntax of ksh prompt func declarationTom Ryder1-1/+1
Turns out that `typeset` only makes a variable local if the function was declared with this syntax. This actually makes a fair bit of sense if I think about it. Wouldn't do this in Bash though.
2016-12-18Undo strange default for login ksh93 ENVTom Ryder1-0/+6
I haven't yet figured out why it does this