I'm a newbie on Manjaro Linux with zsh and using ranger as a file manager. It seems like the script was working before an update but not sure if it's at all related.
The script (shortcuts.sh) reads a file (.key_directories) that contains a list of paths to specific folders. Each line contains a few letters and a path to a directory, i.e. 2 columns:
examples:
md /run/media/
mov /run/media/movies
docs /run/media/docs
then adds a little bit of text which depends on which file the list will end up in. Either a text file (.shortcuts) which will be used as a list of aliases for bashrc and zshrc or a text file (shortcuts.conf) which will be used by ranger file manager as a list of path to different folders.
The script worked perfectly until it didn't. I'm not sure where the problem is.
Here is the full script shorcuts.sh:
#!/bin/bash
# Shell rc file (i.e. bash vs. zsh, etc.)
shellrc="$HOME/.zshrc"
bshellrc="$HOME/.bashrc"
# Config locations
folders="$HOME/.config/ranger/.key_directories"
configs="$HOME/.config/ranger/.key_files"
# Output locations
shell_shortcuts="$HOME/.shortcuts"
ranger_shortcuts="$HOME/.config/ranger/shortcuts.conf"
# Remove
rm -f "$ranger_shortcuts" 2>/dev/null
echo "alias \\" > "$shell_shortcuts"
# Ensure text of argument 1 exists in the file argument 2
ensure() { (grep "$1" "$2")>/dev/null 2>&1 || echo "$1" >> "$2" ;}
ensure "source $shell_shortcuts" "$shellrc"
ensure "source $shell_shortcuts" "$bshellrc"
ensure "source $HOME/.config/ranger/shortcuts.conf" "$HOME/.config/ranger/rc.conf"
# Format the `folders` file in the correct syntax and sent it to all three configs.
sed "s/#.*$//;/^$/d" "$folders" | tee >(awk '{print $1"=\"cd "$2" && ls -a\" \\"}' >> "$shell_shortcuts") \
| awk '{print "map g"$1" cd "$2"\nmap t"$1" tab_new "$2"\nmap m"$1" shell mv -v %s "$2"\nmap Y"$1" shell cp -rv %s "$2}' >> "$ranger_shortcuts"
# Format the `configs` file in the correct syntax and sent it to both configs.
sed "s/#.*$//;/^$/d" "$configs" | tee >(awk '{print $1"=\"$EDITOR "$2"\" \\"}' >> "$shell_shortcuts") \
| awk '{print "map "$1" shell $EDITOR "$2}' >> "$ranger_shortcuts"
Here is the .key_directories content:
# add here the path to your directories
md /run/media/
mov /run/media/movies
docs /run/media/docs
The error I get is:
shortcuts.sh: line 28: syntax error unexpected token `('
shortcuts.sh: line 28: sed "s/#.*$//;/^$/d" "$folders" | tee >(awk '{print $1"=\"cd "$2" && ls -a\" \\"}' >> "$shell_shortcuts") \
The script is supposed to use the text file .key_directories, ignore lines starting with # and empty lines. then take each line add the necessary text and create a new file with the result.
example: .key_directory
md /run/media
The script creates a text file .shortcuts with this content
alias \
md = cd /run/media
and then the script creates a text file shortcuts.conf with this content:
map gmd cd /run/media
map tmd tab_new /run/media
map mmd shell mv -v %s /run/media
map Ymd shell cp -rc %s /run/media
So far I've double checked if there was additional or missing spaces. Also tried swapping single quotes with double quotes and also removing them. But nothing really works and I spent already a few hours trying to understand how sed, tee and awk works seperatly and together with a bunch of examples but I still can't figure out why the script stopped working and how to fix.
If anybody could help, that would be awesome. Thank you in advance.
EDIT: bash version 5.0.7 / ZSH version 5.7.1
Looks like your script is being called with /bin/sh.
Note: The #! line is used only if the script is called with absolute_or_relative_path/script_name.sh.
If you call it as sh absolute_or_relative_path/script_name.sh, it will use /bin/sh as interpreter.