Run the test *files* in parallel, but not the tests *within* the files. This can be reverted after configuration (i.e., `$BASH_IT/enabled` et al) lives *outside* the repo.
There is no reason for this to be in the `plugins` directory, it just needs to have a load priority sufficiently high that it runs after any aliases are defined.
Improve `_bash-it-erase-term()`, `_bash-it-flash-term()`, `_bash-it-rewind()`, `_bash-it-search-result()`, and `_bash-it-search-component()`. Minor tweaks to `_bash-it-is-partial-match()`, and `_bash-it-search()`.
This reverts bash-it/bash-it#99, a metaprogramming adventure in terminal color code escape computation. It was functionally reverted in bash-it/bash-it#699; I'm just finishing the job.
...to `_bash-it()`.
The norm is for the completion function for, e.g., `teh_cmd`. to be named with the same name and a prepended underscore, i.e. `_teh_cmd`. This alsö reduces namespace confusion, which will be relevant in a future patch.
Quote things, SC2268, SC2143, SC2181, SC2162, SC2016, SC2013, &c.
Rewrite globbing per `shellcheck`’s SC2013, and alsö s/typeset/local/g. Eliminate `compgen` where possible.
Alsö: use the existing utility functions `_bash-it-get-component-type-from-path` and `_bash-it-get-component-name-from-path`, which just use parameter substitution anyway. Why was `sed` here?
Alsö, don't add not-existing directories to `$PATH` in `pathmunge()`.
Finally, merge PR #1865 from NoahGorny...and clean it a bit...
Symlink `/test_lib/bats-core/test/fixtures/parallel/suite/helper.bash` references a not-existing file, but it's in a submodule so I can't just fix it. Exclude it from `rsync` in `setup_test_fixture()`.
Test was failing only on Mac OS X for some reason, so refactor a little. It turned out to be due to explicitly setting `$HOME` *after* loading the plugin.
Literally copying a line from the source to be tested is perhaps not the best way to test that code. 😉
That said, we do want to verify that the function was actually loaded.
TODO: actually test the function.