21norm is a python script which tries to automatically resolve many norm errors issued by norminette
When you are studying at 42 you know: once you finished a project you have to correct all norm errors before submitting. This can be quite tedious if you have many files and is it not kind of a stupid thing to go through all the files, replace spaces by tabs or vice versa, remove consecutive newlines and so on? 21norm will try to easen that burden by loading the files, checking for norm errors, and correcting them (if it can).
It cannot correct the most beautiful norm errors:
- Line too long
- too many lines in function
- too many functions in file
21norm is experimental. It will overwrite your files. There is a chance that it makes changes that will break your code or might even lead to data loss.
I firmly recommend to make a git add ., git commit -m and git push before you run it, so you can return to your old files in case of data loss.
Type 21norm to norm all *.c and *.h files in the current directory. Type 21norm specific_folder/specific_file.c to norm just that file.
norminette must be installed and in the path.
Clone the git repository 21norm to a folder, e.g. /home/<user>/programs/21norm. In the downloaded folder, edit config.py to change USER and EMAIL to your username and email, which will be shown in the 42 header.
Goto /home/<user>/bin, create a symbolic link to the file 21norm.py in the download folder and make it executable:
cd /home/<user>/bin
ln -s /home/<user>/programs/21norm/21norm.py ./21norm
chmod 0100 ./21norm
You can now execute 21norm in your shell. If not, or if the directory /home/<user>/bin should not exist, see the section below.
If there is no folder /home/<user>/bin create it:
mkdir ~/bin`
Check if it is in the $PATH variable:
echo $PATH
If not, you can add export PATH=/home/<user>/bin to your ~/.zshrc (or ~/.bashrc if you are using bash) or to ~/.profile, regarding your system/needs.
Open the file where you can edit the keyboard shortcuts via File -> Preferences -> Keyboard Shortcuts and click on the file icon in the upper right corner:
Add the following code at the top. You can also modify the shortcut to your needs (here I chose CTRL + N):
{
"key": "ctrl+shift+n",
"command": "workbench.action.terminal.sendSequence",
"args": {
"text": "VSCODE_WD=$(sed 's/\\\\/\\//g' <<< '${fileDirname}' | sed 's/C:/\\/mnt\\/c/g') && cd \"$VSCODE_WD\" && echo \"cd $VSCODE_WD\"\n21norm \"${fileBasename}\"\n"
}
},
After pressing the shorcut, it should activate the latest open terminal, change to the directory where currently opened file resides and run 42check on it.
If you use VS Code on Windows with WSL (Windows Subsystem for Linux), try this little hack:
{
"key": "ctrl+n",
"command": "workbench.action.terminal.sendSequence",
"args": {
"text": "VSCODE_WD=`sed 's/\\\\\\\\/\\//g' <<< '${fileDirname}' | sed 's/C:/\\/mnt\\/c/g'` cd \"$VSCODE_WD\" && echo \"cd $VSCODE_WD\"\n42checker \"${fileBasename}\"\n"
}
},
21norm will recognize Doxygen Comments in the following format and will try to wrap lines longer than 80 spaces:
/**
* @brief sort the items on stack_a using stack_b
*
* The idea of double sort is to separate all items into two groups in stack_a
* and stack_b (push_half_to_b) which are then put in order concurrently
* (stack_a sorted increasingly, stack_b sorted decreasingly) and finally
* joined, in order, on stack_a (push_back_to_a). There are to options to
* proceed: either push the smallest elements of stack_a to stack_b or the
* largest elements. Both variants are tested (op1 and op2 respectively) and are
* compared. The variant with the smallest number of oper- ations is returend.
*
* @param stack_a stack configuration to bring in order
* @return t_ring* list of operations to sort stack_a using stack_b
*/
If you like 21norm, you might also like 42check, which will reformat the output of norminette to make links with line and column numbers work in VS Code.