to write output, you would usually just use the print builtin
since commanders are just lua custom commands but this does not
consider the fact of pipes or other shell operators being used
to redirect or whatever.
this adds readable/writable "sinks" which is a type for input
or output and is currently only used for commanders but can be
used for other hilbish things in the future
fixes an issue with expanded aliases being added
to history with a recent commit (6th time now with
this issue?) and makes behavior with other runners
consistent
this can technically be a breaking change to people
overriding the sh runner function
* refactor: use custom event emitter
* fix: sigint hook emit on windows
* fix: restore correct hilbish conf file
* fix: call recoverer for go listeners
* refactor(golibs/bait): use 1 map for listeners
* feat: add once listeners, ability to remove listeners and remove listener on error
* perf: reslice listener slice instead of trying to do ordered move with append
* feat(bait): add release function to remove event listener
* perf: remove listener directly from once emit instead of using off function
* refactor: use bait event emitter on commander
* docs(golibs/bait): add doc strings for functions
* docs: set changelog
* docs(golibs/bait): add docs for lua release function
* refactor: put file history handler in line reader instance instead of global
* feat: add lua history handler in go
* feat: use lua to retrieve readline history
* refactor: handle history in lua
this also introduces a new opt: history
if it is false, command history won't get added
* fix: remove nature.history require
* docs: add changes in changelog
* fix: add comma after history opt
* refactor!: make runners require returning a table
allows for more options for runners in the future,
and makes it so that you can avoid passing
certain args more easily.
* feat: allow runners to specify continue in return to prompt for more input
* docs: update changelog
* refactor: reorder returns of handleSh function
* refactor: move out reprompting and runner handling to functions
makes codefactor happy hopefully. this commit includes
a fix to check if after reprompt the user hits ctrl d
and just exits cleanly
* feat: add hilbish.job.add function
this is mainly to accomodate for the employer
handler (#152)
* feat!: add start function to jobs
the commit itself adds a few things but the
main purpose is to facilitate a lua side start
function that can restart the job
there is a breaking change in the hilbish.job.add
function; it is now required to provide an extra
table for arguments, since the first cmd
table isnt really what's actually ran
* fix: reuse standard files for jobs
* fix: deadlock in lua job add function and not taking proper amount of args
* fix: assign binary path to job
* feat: emit job.add hook when job is added
* chore: update modules
* fix: use setpgid on cmd procattr for background jobs
* fix: use right procattr on correct os
* fix: set bg proc attr in build tagged file
* feat: add disown function
* fix: stop jobs on exit
* feat: add disown command
* feat: add jobs.last function to get last job
* feat: make disown command get last job if id isnt suppied as arg
* chore: remove unused code
* feat: add job output
* chore: fix comments
* fix!: make exec path in job add explicit in lua side
* docs: add docs and changelogs relating to jobs
if an alias is something which isn't valid syntax,
specifically if hilbish cant split up the input
properly to execute, it will report the error to
the user. the previous behaviour was a panic since
on error the args slice will be of length 0
this is basically an edge case and fixes a bug
which shouldnt happen normally
this introduces a breaking change to runner functions.
they are now required to return 3 values, the first
being the user's input, and the 2 others that it was
before. the `hilbish.runner` functions respectively
have been updated, so if you just return from those
there will be no difference
major rewrite which changes the library hilbish uses for it's lua vm
this one implements lua 5.4, and since that's a major version bump,
it's a breaking change. introduced here also is a fix for `hilbish.login`
not being the right value
* refactor: start work on lua 5.4
lots of commented out code
ive found a go lua library which implements lua 5.4
and found an opportunity to start working on it.
this commit basically removes everything and just leaves
enough for the shell to be "usable" and able to start.
there are no builtins or libraries (besides the `hilbish` global)
* fix: call cont next in prompt function
this continues execution of lua, very obvious
fixes an issue with code stopping at the prompt function
* fix: handle errors in user config
* fix: handle panic in lua input if it is incorrect
* feat: implement bait
* refactor: use util funcs to run lua where possible
* refactor: move arg handle function to util
* feat: implement commander
* feat: implement fs
* feat: add hilbish module functions used by prelude
* chore: use custom fork of golua
* fix: make sure args to setenv are strings in prelude
* feat: implement completions
* chore: remove comment
* feat: implement terminal
* feat: implement hilbish.interval
* chore: update lunacolors
* chore: update golua
* feat: implement aliases
* feat: add input mode
* feat: implement runner mode
* style: use comma separated cases instead of fallthrough
* feat: implement syntax highlight and hints
* chore: add comments to document util functions
* chore: fix dofile comment doc
* refactor: make loader functions for go modules unexported
* feat: implement job management
* feat: add hilbish properties
* feat: implement all hilbish module functions
* feat: implement history interface
* feat: add completion interface
* feat: add module description docs
* feat: implement os interface
* refactor: use hlalias for add function in hilbish.alias interface
* feat: make it so hilbish.run can return command output
* fix: set hilbish.exitCode to last command exit code
* fix(ansikit): flush on io.write
* fix: deregister commander if return isnt number
* feat: run script when provided path
* fix: read file manually in DoFile to avoid shebang
* chore: add comment for reason of unreading byte
* fix: remove prelude error printing
* fix: add names at chunk load for context in errors
* fix: add newline at the beginning of file buffer when there is shebang
this makes the line count in error messages line up properly
* fix: remove extra newline after error
* fix: handle path binaries properly on windows (closes#117, #118)
* refactor: dont return exec name since it isnt needed
* fix: return correct error in find exec function and stat always
* fix: remove filepath import for exec file check on unix
this is a pretty big commit which mainly contains a refactor
and breaking change to how command completions are done.
before that, a hilbish.completion interface has been added
which for now just has 2 functions (`files` and `bins`)
for completions of normal files and executables.
hilbish.complete is now expected to return a table of
"completions groups," which are as the name suggests a group
for a completion. a completion group is a table which has
the fields `type`, which can be either `list` or `grid`,
and `items`, being an array (or string keyed table) of items
if an item is string keyed the item itself is the key name
and the value is a table with the first value in it being the
description for the item. this description is only applied
with the list type.
this is probably the longest commit message ive written