Inside a fresh git repo, i.e. immediately after a `git init`, usually no
commit template exists yet. In this case, git renders a different
status message than "Initial commit on". We should consider this message
when attempting to parse out the branch name.
Fixes#6301
Deals with app error page, saving true error instead.
Upon app failure, Heroku returns HTML "Application Error" page.
Finding HTML page in .gitignore is confusing, so I replaced
`-s` with `-f` in curl calls, which cuts such output.
Replace instead of addition as no progress meter outputs either.
It is practically impossible to teach good programming style to students that have had prior exposure to BASIC. As potential programmers, they are mentally mutilated beyond hope of regeneration.
-- E. W. Dijkstra
This will work only on files and directories in a DrvFs mount, i.e.
that can be translated to a Windows drive path.
For example: /mnt/c/Users/user.
Files and folders inside the LXSS directory can't be handled in
Windows, they must be ONLY used by the WSL subsystem. That's why
you won't be able to open your $HOME directory, for instance.
See https://blogs.msdn.microsoft.com/commandline/2016/11/17/do-not-change-linux-files-using-windows-apps-and-tools/
* Make steef much faster by not printing all history each time
* Use whence -c to expand shell functions as well
* Use $2 (expanded command about to be ran)
NOTE: The scrapper was previously included, but due to the website it scrapped
being dead it's no longer useful. See #3642 for the code.
Co-authored-by: Marc Cornellà <marc.cornella@live.com>
This fixes the old behavior which made it so all duplicates would be
deleted if the command history filled up with unique events.
> You should be sure to set the value of HISTSIZE to a larger number
> than SAVEHIST in order to give you some room for the duplicated
> events, otherwise this option will behave just like HIST_IGNORE_ALL_DUPS
> once the history fills up with unique events.
This commit changes the history alias into a function which puts
the passed arguments before `-l 1`. It also provides a temporary
workaround to the lack of a `history -c` command in zsh.
For more information see issues 739 and 789.
`source ~/.zshrc` is a bad practice, it doesn't make sure that the full
zsh session is reloaded and it may have side effects.
Use `$SHELL` as the path to zsh if it exists (this will fix edge cases
where the zsh used is not the first on $PATH). Otherwise, use `zsh`.
* Always try and start gpg-agent, with --use-standard-socket it will try and use a standard socket directory. It won't start multiple agents if agent is already running. In addition, XDG_RUNTIME_DIR isn't always set
* ssh socket if broken if --daemon is run again, so onky start if we don't have a socket
* Removed unnecessary allocation of GPG_SSH_AUTH_SOCK
* [rkj-repos] Make `hg prompt` check less strict
Move the `hg prompt` check inside the hg_prompt_info function so that
it returns an empty string if hg-prompt isn't installed.
Fixes#6743.
* [rkj-repos] Check for hg in `hg prompt` function
The git:// transport is completely unauthenticated. An attacker on the local or upstream network can easily man-in-the-middle an oh-my-zsh update and get remote code execution on your system. Only the https:// git transport should be used.
