rico-chet
rico-chet
Apparently, tests are run in non-interactive mode, so they will need to be adapted or the business logic for interactive mode moved to a separate script so it can be...
Re https://github.com/Bash-it/bash-it/pull/1533#issuecomment-606743441 > Just wondering if this is something to be concerned about what with the user delegating so much of their startup routine to bash-it (per design). IMHO, that's...
With the current PR, we have the issues that: 1. Tests fail in CI due to non-interactive shell being used. 2. Users who call from a custom `~/.bash_profile` → `bash_it.sh`...
To make Travis happy, we would need to figure out how to run the tests in CI pretending the shell is interactive, because otherwise they just fail. Shouldn't be hard...
For now, I found that it's complicated on the `bash` side already: [`run_startup_files()`](https://git.savannah.gnu.org/cgit/bash.git/tree/shell.c?h=bash-5.0#n1083) is full of `if`s, `ifdef`s and flags, so no wonder I cannot get the SCP scenario test...
~~I guess the issue here is that `LD_PRELOAD` contains the `stderred.so` _twice_ when you put the `export` directive from the documentation verbatim to your `~/.bashrc` because it is run twice...
@twmccart I wanted to find out first where the leading dash is coming from and whether a non-standard `tmux` config would possibly break the check. I found this: https://github.com/tmux/tmux/blob/c9d482ab489d5d57d481858091608ee1b32e46ab/window.c#L993 and...