Format (rewrap) Markdown document to adhere to given line width - format

I have a long Markdown document with lines longer than 80 characters. I would like it to adhere to a style guide prescribing lines no longer than 80 characters (with exceptions of code snippets and tables). Is there a tool an editor or an editor extension reformat (rewrap) the document while respecting Markdown syntax?

Prettier (https://prettier.io), which is available as a plug-in for many editors, seems like a good option.

Open up the file in Emacs, select all content (bound to C-x h), call fill-paragraph (bound to M-q). The Markdown syntax is respected as long as markdown-mode is installed.
You can have it wrap at any given column by setting the variable fill-column, which can be done by evaluating (setq fill-column 80), or calling M-x set-fill-column RET 80, or typing C-x f 80 RET.
Save the file by calling save-buffer, bound to C-x C-s.
Full incantation: C-x f 80 RET C-x h M-q C-x C-s.

Related

Im using vim to check man page, and there is some weird character,how to make it normal?

I'm trying using
man ascii | vi -
to check ascii manual with vim,
then there are some weird character like '^H'
weird character in vim
how to make it normal?
EDIT:
i think this is due to some wrong config of my vim, or terminal,
because i tried some other machine,and it works fine,any body help?
What you're seeing here is the man page convention for typesetting bold and underlined characters.
The ^H character is the "backspace" or BS character, which you can find in ascii(7) as ASCII 8.
Displaying as ^H is a convention to indicate "Control-H", the caret being a symbol for the "Control" key, since that's a key combination that generate this control character (note that "H" is the eighth letter, also it's on the same row as BS in the ASCII chart at the man page.)
Also note that ^H here is a single character and not the separate ^ and H characters. If you move along that line with l motion repeatedly, you'll see Vim skips the Hs. Vim might also display them in a different color to make it clear they're special characters.
The effect you're seeing here is called overstruck characters and it comes from the days when computers would be connected to line printers rather than screens. Man pages were created early in the Unix days and are typeset with a program named nroff (which has roots that predate even Unix) that uses these sequences to encode its text decorations, even to this day.
So one way to make text bold (or, print it stronger) was to print a letter, back out one space, then print it again. In your example, N^HNA^HAM^HME^HE is being used to print the word NAME in bold.
Typically, you'll use a man pager program that knows how to handle those sequences. For instance, less has specific support for overstruck text and is able to recognize those sequences using backspace (or carriage return, also possible to use it for that) and convert into bold or underline text using terminal escape sequences which actually achieve those effects in a terminal.
If you're using a different pager, often you'll use col -b to strip your text from those sequences. (See col(1) for details.) Note that col -b doesn't produce terminal escape sequences around overstruck text, it simply removes all overstriking, leaves the text plain, so any text reader can display it.
(In that sense, man ascii | col -b | vi - is one possible way around your issue, but not the best one.)
Vim doesn't support overstruck text or terminal escape sequences in text it receives as input. So the approach used to use Vim as a man pager is to strip all overstriking (same as col -b does) and then use Vim's syntax highlighting feature to mark man page sections etc. (So instead of seeing the original markings typeset into the man pages, those are all being stripped and you're actually seeing Vim syntax configuration for man pages.)
This is accomplished by the manpager.vim script, which is shipped by default with Vim and which registers a :MANPAGER command that you can access as vim +MANPAGER -.
You'll see that part of what that does is to remove overstriking, same as col -b does. It also sets the file type correctly and enables syntax highlighting, setting up Vim as a suitable viewer for man pages.
I think this is due to some wrong config of my vim, or terminal, because I tried it in some other machine and it works fine there.
I'd say this is most likely due to the man pages being typeset differently between the two systems.
Man pages themselves and the pipelines used to generate them from the source code are not very uniform and you'll find differences across operating systems or even Linux distributions.
It's possible that on your other system all man pages get stripped using col -b at creation time, or that man will strip them when piping them to something that's not what's set in $MANPAGER, or even that they don't have anything marked as bold or underline in their sources.
Hard to tell exactly which (your reference to working fine on another machine lacks details to tell), but this would be my bet.
In short: To use Vim as a pager for man pages, instead of piping them to Vim, set it up so that man will run vim as a pager itself.
In Vim, see :help manpager.vim which has instructions on how to set this up.
In short, all you need is to add this line to your ~/.bash_profile:
export MANPAGER="vim -M +MANPAGER -"
After setting that up and logging in again, all you need is:
$ man ascii
And man will launch vim in the right mode for you.

Switching between files without closing vim editor [duplicate]

I've started using Vim to develop Perl scripts and am starting to find it very powerful.
One thing I like is to be able to open multiple files at once with:
vi main.pl maintenance.pl
and then hop between them with:
:n
:prev
and see which file are open with:
:args
And to add a file, I can say:
:n test.pl
which I expect would then be added to my list of files, but instead it wipes out my current file list and when I type :args I only have test.pl open.
So how can I add and remove files in my args list?
Why not use tabs (introduced in Vim 7)?
You can switch between tabs with :tabn and :tabp,
With :tabe <filepath> you can add a new tab; and with a regular :q or :wq you close a tab.
If you map :tabn and :tabp to your F7/F8 keys you can easily switch between files.
If there are not that many files or you don't have Vim 7 you can also split your screen in multiple files: :sp <filepath>. Then you can switch between splitscreens with Ctrl+W and then an arrow key in the direction you want to move (or instead of arrow keys, w for next and W for previous splitscreen)
Listing
To see a list of current buffers, I use:
:ls
Opening
To open a new file, I use
:e ../myFile.pl
with enhanced tab completion (put set wildmenu in your .vimrc).
Note: you can also use :find which will search a set of paths for you, but you need to customize those paths first.
Switching
To switch between all open files, I use
:b myfile
with enhanced tab completion (still set wildmenu).
Note: :b# chooses the last visited file, so you can use it to switch quickly between two files.
Using windows
Ctrl-W s and Ctrl-W v to split the current window horizontally and vertically. You can also use :split and :vertical split (:sp and :vs)
Ctrl-W w to switch between open windows, and Ctrl-W h (or j or k or l) to navigate through open windows.
Ctrl-W c to close the current window, and Ctrl-W o to close all windows except the current one.
Starting vim with a -o or -O flag opens each file in its own split.
With all these I don't need tabs in Vim, and my fingers find my buffers, not my eyes.
Note: if you want all files to go to the same instance of Vim, start Vim with the --remote-silent option.
:ls
for list of open buffers
:bp previous buffer
:bn next buffer
:bn (n a number) move to n'th buffer
:b <filename-part> with tab-key providing auto-completion (awesome !!)
In some versions of vim, bn and bp are actually bnext and bprevious respectively. Tab auto-complete is helpful in this case.
Or when you are in normal mode, use ^ to switch to the last file you were working on.
Plus, you can save sessions of vim
:mksession! ~/today.ses
The above command saves the current open file buffers and settings to ~/today.ses. You can load that session by using
vim -S ~/today.ses
No hassle remembering where you left off yesterday. ;)
To add to the args list:
:argadd
To delete from the args list:
:argdelete
In your example, you could use :argedit test.pl to add test.pl to the args list and edit the file in one step.
:help args gives much more detail and advanced usage
I use buffer commands - :bn (next buffer), :bp (previous buffer) :buffers (list open buffers) :b<n> (open buffer n) :bd (delete buffer). :e <filename> will just open into a new buffer.
I think you may be using the wrong command for looking at the list of files that you have open.
Try doing an :ls to see the list of files that you have open and you'll see:
1 %a "./checkin.pl" line 1
2 # "./grabakamailogs.pl" line 1
3 "./grabwmlogs.pl" line 0
etc.
You can then bounce through the files by referring to them by the numbers listed, e.g.
:3b
or you can split your screen by entering the number but using sb instead of just b.
As an aside % refers to the file currently visible and # refers to the alternate file.
You can easily toggle between these two files by pressing Ctrl Shift 6
Edit: like :ls you can use :reg to see the current contents of your registers including the 0-9 registers that contain what you've deleted. This is especially useful if you want to reuse some text that you've previously deleted.
Vim (but not the original Vi!) has tabs which I find (in many contexts) superior to buffers. You can say :tabe [filename] to open a file in a new tab. Cycling between tabs is done by clicking on the tab or by the key combinations [n]gt and gT. Graphical Vim even has graphical tabs.
Things like :e and :badd will only accept ONE argument, therefore the following will fail
:e foo.txt bar.txt
:e /foo/bar/*.txt
:badd /foo/bar/*
If you want to add multiple files from within vim, use arga[dd]
:arga foo.txt bar.txt
:arga /foo/bar/*.txt
:argadd /foo/bar/*
Many answers here! What I use without reinventing the wheel - the most famous plugins (that are not going to die any time soon and are used by many people) to be ultra fast and geeky.
ctrlpvim/ctrlp.vim - to find file by name fuzzy search by its location or just its name
jlanzarotta/bufexplorer - to browse opened buffers (when you do not remember how many files you opened and modified recently and you do not remember where they are, probably because you searched for them with Ag)
rking/ag.vim to search the files with respect to gitignore
scrooloose/nerdtree to see the directory structure, lookaround, add/delete/modify files
EDIT: Recently I have been using dyng/ctrlsf.vim to search with contextual view (like Sublime search) and I switched the engine from ag to ripgrep. The performance is outstanding.
EDIT2: Along with CtrlSF you can use mg979/vim-visual-multi, make changes to multiple files at once and then at the end save them in one go.
Some answers in this thread suggest using tabs and others suggest using buffer to accomplish the same thing. Tabs and Buffers are different. I strongly suggest you read this article "Vim Tab madness - Buffers vs Tabs".
Here's a nice summary I pulled from the article:
Summary:
A buffer is the in-memory text of a file.
A window is a viewport on a buffer.
A tab page is a collection of windows.
To change all buffers to tab view.
:tab sball
will open all the buffers to tab view. Then we can use any tab related commands
gt or :tabn " go to next tab
gT or :tabp or :tabN " go to previous tab
details at :help tab-page-commands.
We can instruct vim to open ,as tab view, multiple files by vim -p file1 file2.
alias vim='vim -p' will be useful.
The same thing can also be achieved by having following autocommand in ~/.vimrc
au VimEnter * if !&diff | tab all | tabfirst | endif
Anyway to answer the question:
To add to arg list: arga file,
To delete from arg list: argd pattern
More at :help arglist
When using multiple files in vim, I use these commands mostly (with ~350 files open):
:b <partial filename><tab> (jump to a buffer)
:bw (buffer wipe, remove a buffer)
:e <file path> (edit, open a new buffer>
pltags - enable jumping to subroutine/method definitions
You may want to use Vim global marks.
This way you can quickly bounce between files, and even to the marked location in the file. Also, the key commands are short:
'C takes me to the code I'm working with,
'T takes me to the unit test I'm working with.
When you change places, resetting the marks is quick too:
mC marks the new code spot,
mT marks the new test spot.
If using only vim built-in commands, the best one that I ever saw to switch among multiple buffers is this:
nnoremap <Leader>f :set nomore<Bar>:ls<Bar>:set more<CR>:b<Space>
It perfectly combines both :ls and :b commands -- listing all opened buffers and waiting for you to input the command to switch buffer.
Given above mapping in vimrc, once you type <Leader>f,
All opened buffers are displayed
You can:
Type 23 to go to buffer 23,
Type # to go to the alternative/MRU buffer,
Type partial name of file, then type <Tab>, or <C-i> to autocomplete,
Or just <CR> or <Esc> to stay on current buffer
A snapshot of output for the above key mapping is:
:set nomore|:ls|:set more
1 h "script.py" line 1
2 #h + "file1.txt" line 6 -- '#' for alternative buffer
3 %a "README.md" line 17 -- '%' for current buffer
4 "file3.txt" line 0 -- line 0 for hasn't switched to
5 + "/etc/passwd" line 42 -- '+' for modified
:b '<Cursor> here'
In the above snapshot:
Second column: %a for current, h for hidden, # for previous, empty for hasn't been switched to.
Third column: + for modified.
Also, I strongly suggest set hidden. See :help 'hidden'.
I use the same .vimrc file for gVim and the command line Vim. I tend to use tabs in gVim and buffers in the command line Vim, so I have my .vimrc set up to make working with both of them easier:
" Movement between tabs OR buffers
nnoremap L :call MyNext()<CR>
nnoremap H :call MyPrev()<CR>
" MyNext() and MyPrev(): Movement between tabs OR buffers
function! MyNext()
if exists( '*tabpagenr' ) && tabpagenr('$') != 1
" Tab support && tabs open
normal gt
else
" No tab support, or no tabs open
execute ":bnext"
endif
endfunction
function! MyPrev()
if exists( '*tabpagenr' ) && tabpagenr('$') != '1'
" Tab support && tabs open
normal gT
else
" No tab support, or no tabs open
execute ":bprev"
endif
endfunction
This clobbers the existing mappings for H and L, but it makes switching between files extremely fast and easy. Just hit H for next and L for previous; whether you're using tabs or buffers, you'll get the intended results.
If you are going to use multiple buffers, I think the most important thing is to
set hidden
so that it will let you switch buffers even if you have unsaved changes in the one you are leaving.
I use the following, this gives you lots of features that you'd expect to have in other editors such as Sublime Text / Textmate
Use buffers not 'tab pages'. Buffers are the same concept as tabs in almost all other editors.
If you want the same look of having tabs you can use the vim-airline plugin with the following setting in your .vimrc: let g:airline#extensions#tabline#enabled = 1. This automatically displays all the buffers as tab headers when you have no tab pages opened
Use Tim Pope's vim-unimpaired which gives [b and ]b for moving to previous/next buffers respectively (plus a whole host of other goodies)
Have set wildmenu in your .vimrc then when you type :b <file part> + Tab for a buffer you will get a list of possible buffers that you can use left/right arrows to scroll through
Use Tim Pope's vim-obsession plugin to store sessions that play nicely with airline (I had lots of pain with sessions and plugins)
Use Tim Pope's vim-vinegar plugin. This works with the native :Explore but makes it much easier to work with. You just type - to open the explorer, which is the same key as to go up a directory in the explorer. Makes navigating faster (however with fzf I rarely use this)
fzf (which can be installed as a vim plugin) is also a really powerful fuzzy finder that you can use for searching for files (and buffers too). fzf also plays very nicely with fd (a faster version of find)
Use Ripgrep with vim-ripgrep to search through your code base and then you can use :cdo on the results to do search and replace
My way to effectively work with multiple files is to use tmux.
It allows you to split windows vertically and horizontally, as in:
I have it working this way on both my mac and linux machines and I find it better than the native window pane switching mechanism that's provided (on Macs). I find the switching easier and only with tmux have I been able to get the 'new page at the same current directory' working on my mac (despite the fact that there seems to be options to open new panes in the same directory) which is a surprisingly critical piece. An instant new pane at the current location is amazingly useful. A method that does new panes with the same key combos for both OS's is critical for me and a bonus for all for future personal compatibility.
Aside from multiple tmux panes, I've also tried using multiple tabs, e.g. and multiple new windows, e.g. and ultimately I've found that multiple tmux panes to be the most useful for me. I am very 'visual' and like to keep my various contexts right in front of me, connected together as panes.
tmux also support horizontal and vertical panes which the older screen didn't (though mac's iterm2 seems to support it, but again, the current directory setting didn't work for me). tmux 1.8
In my and other many vim users, the best option is to,
Open the file using,
:e file_name.extension
And then just Ctrl + 6 to change to the last buffer. Or, you can always press
:ls to list the buffer and then change the buffer using b followed by the buffer number.
We make a vertical or horizontal split using
:vsp for vertical split
:sp for horizantal split
And then <C-W><C-H/K/L/j> to change the working split.
You can ofcourse edit any file in any number of splits.
I use the command line and git a lot, so I have this alias in my bashrc:
alias gvim="gvim --servername \$(git rev-parse --show-toplevel || echo 'default') --remote-tab"
This will open each new file in a new tab on an existing window and will create one window for each git repository.
So if you open two files from repo A, and 3 files from repo B, you will end up with two windows, one for repo A with two tabs and one for repo B with three tabs.
If the file you are opening is not contained in a git repo it will go to a default window.
To jump between tabs I use these mappings:
nmap <C-p> :tabprevious<CR>
nmap <C-n> :tabnext<CR>
To open multiple files at once you should combine this with one of the other solutions.
I use multiple buffers that are set hidden in my ~/.vimrc file.
The mini-buffer explorer script is nice too to get a nice compact listing of your buffers. Then :b1 or :b2... to go to the appropriate buffer or use the mini-buffer explorer and tab through the buffers.
have a try following maps for convenience editing multiple files
" split windows
nmap <leader>sh :leftabove vnew<CR>
nmap <leader>sl :rightbelow vnew<CR>
nmap <leader>sk :leftabove new<CR>
nmap <leader>sj :rightbelow new<CR>
" moving around
nmap <C-j> <C-w>j
nmap <C-k> <C-w>k
nmap <C-l> <C-w>l
nmap <C-h> <C-w>h
I made a very simple video showing the workflow that I use. Basically I use the Ctrl-P Vim plugin, and I mapped the buffer navigation to the Enter key.
In this way I can press Enter in normal mode, look at the list of open files (that shows up in a small new window at the bottom of the screen), select the file I want to edit and press Enter again. To quickly search through multiple open files, just type part of the file name, select the file and press Enter.
I don't have many files open in the video, but it becomes incredibly helpful when you start having a lot of them.
Since the plugin sorts the buffers using a MRU ordering, you can just press Enter twice and jump to the most recent file you were editing.
After the plugin is installed, the only configuration you need is:
nmap <CR> :CtrlPBuffer<CR>
Of course you can map it to a different key, but I find the mapping to enter to be very handy.
I would suggest using the plugin
NERDtree
Here is the github link with instructions.
Nerdtree
I use vim-plug as a plugin manager, but you can use Vundle as well.
vim-plug
Vundle
When I started using VIM I didn't realize that tabs were supposed to be used as different window layouts, and buffer serves the role for multiple file editing / switching between each other. Actually in the beginning tabs are not even there before v7.0 and I just opened one VIM inside a terminal tab (I was using gnome-terminal at the moment), and switch between tabs using alt+numbers, since I thought using commands like :buffers, :bn and :bp were too much for me. When VIM 7.0 was released I find it's easier to manager a lot of files and switched to it, but recently I just realized that buffers should always be the way to go, unless one thing: you need to configure it to make it works right.
So I tried vim-airline and enabled the visual on-top tab-like buffer bar, but graphic was having problem with my iTerm2, so I tried a couple of others and it seems that MBE works the best for me. I also set shift+h/l as shortcuts, since the original ones (moving to the head/tail of the current page) is not very useful to me.
map <S-h> :bprev<Return>
map <S-l> :bnext<Return>
It seems to be even easier than gt and gT, and :e is easier than :tabnew too. I find :bd is not as convenient as :q though (MBE is having some problem with it) but I can live with all files in buffer I think.
Most of the answers in this thread are using plain vim commands which is of course fine but I thought I would provide an extensive answer using a combination of plugins and functions that I find particularly useful (at least some of these tips came from Gary Bernhardt's file navigation tips):
To toggle between the last two file just press <leader> twice. I recommend assigning <leader> to the spacebar:
nnoremap <leader><leader> <c-^>
For quickly moving around a project the answer is a fuzzy matching solution such as CtrlP. I bind it to <leader>a for quick access.
In the case I want to see a visual representation of the currently open buffers I use the BufExplorer plugin. Simple but effective.
If I want to browse around the file system I would use the command line or an external utility (Quicklsilver, Afred etc.) but to look at the current project structure NERD Tree is a classic. Do not use this though in the place of 2 as your main file finding method. It will really slow you down. I use the binding <leader>ff.
These should be enough for finding and opening files. From there of course use horizontal and vertical splits. Concerning splits I find these functions particularly useful:
Open new splits in smaller areas when there is not enough room and expand them on navigation. Refer here for comments on what these do exactly:
set winwidth=84
set winheight=5
set winminheight=5
set winheight=999
nnoremap <C-w>v :111vs<CR>
nnoremap <C-w>s :rightbelow split<CR>
set splitright
Move from split to split easily:
nnoremap <C-J> <C-W><C-J>
nnoremap <C-K> <C-W><C-K>
nnoremap <C-L> <C-W><C-L>
nnoremap <C-H> <C-W><C-H>
if you're on osx and want to be able to click on your tabs, use MouseTerm and SIMBL (taken from here). Also, check out this related discussion.
You can be an absolute madman and alias vim to vim -p by adding in your .bashrc:
alias vim="vim -p"
This will result in opening multiple files from the shell in tabs, without having to invoke :tab ball from within vim afterwards.
To open 2 or more files with vim type: vim -p file1 file2
After that command to go threw that files you can use CTRL+Shift+↑ or ↓ , it will change your files in vim.
If u want to add one more file vim and work on it use: :tabnew file3
Also u can use which will not create a new tab and will open file on screen slicing your screen: :new file3
If u want to use a plugin that will help u work with directories
and files i suggest u NERDTree.
To download it u need to have vim-plug so to download other plugins also NERDTree to type this commands in your ~/.vimrc.
let data_dir = has('nvim') ? stdpath('data') . '/site' : '~/.vim'
if empty(glob(data_dir . '/autoload/plug.vim'))
silent execute '!curl -fLo '.data_dir.'/autoload/plug.vim --create-dirs
https://raw.githubusercontent.com/junegunn/vim-plug/master/plug.vim'
autocmd VimEnter * PlugInstall --sync | source $MYVIMRC
endif
call plug#begin('~/.vim/plugged')
Plug 'scrooloose/nerdtree'
call plug#end()
Then save .vimrc via command :wq , get back to it and type: :PlugInstall
After that the plugins will be installed and u could use your NERDTree with other plugins.

Which editor is used to open man pages in macOS? Is it even an editor?

I am using macOS Sierra(unfortunately), and for example when I typed man curl to see what the -LSso flag does, it opens the man page in a editor which resembles vim slightly. I know its not Vim since it doesn't have :syntax on and you can't use :wq and so on.
The heading of the page says man(less).
I can't find what editor this might be and I'm learning vim and emacs configurations. Or maybe it's just a man page and no editor is used?
Less is a program similar to more (1), but which allows backward movement in the file as well as forward movement. Also, less does not have to read the entire input file before starting, so with large
input files it starts up faster than text editors like vi (1). Less uses termcap (or terminfo on some systems), so it can run on a variety of terminals. There is even limited support for hardcopy
terminals. (On a hardcopy terminal, lines which should be printed at the top of the screen are prefixed with a caret.)
Commands are based on both more and vi. Commands may be preceded by a decimal number, called N in the descriptions below. The number is used by some commands, as indicated.
-- Copied from less man page

iTerm2: delete line?

I'm trying to map ⌘+Delete (backspace) to delete to the beginning of the line (like it works in browsers and text editors) in iTerm2 and I'm unable to find a working escape code for it. I tried 1K (^[1K) based on what I read in Wikipedia. It just prints a "K".
Edit: I found Ctrl+U. Now to find out how to map it. Maybe Hex code 21 (U being 21st letter), so 0x15?
I got it. I have no idea why Hex Code mappings in iTerm2 produce the associated Ctrl+key mappings, but they do. No idea what 0x00 means, either, as it's not assigned to A as might be expected. (though I do believe Unix has its own conventions relating to treating null bytes -- we have e.g. xargs accepting a null byte delimiting format from find for example -- It would be neat if we can bind this to a hotkey with iTerm2)
I was able to find that Ctrl+U does nearly the exact task I want (it deletes the entire line rather than deleting only what is before cursor, but whatever... Ctrl+Y as a bonus can bring it all back). Then I curiously saw that I had hex codes 0x1 and 0x5 mapped to ^A and ^E respectively, for my Cmd+Left and Cmd+Right... so 0x15 is for ^U!
For Mac OS, most editor share the common shortcut ⌘ + Delete: delete to start of the line, in iTerm2 we can switch to this key configuration
Mapping hex code 0x15 to ⌘ + ←Delete in most shells deletes the entire line (content to the left and right of the cursor). While sometimes not as compatible, I find that mapping:
⌘+←Delete to Send Hex Codes:
0x18 0x7f
performs the desired functionality. If you're running ZSH, you'll likely also need to add this to your .zshrc file:
$ echo 'bindkey "^X\\x7f" backward-kill-line' >> ~/.zshrc
as by default ZSH doesn't map backward-kill-line to anything.
Furthermore, you can also delete everything to the right of your cursor by mapping:
⌘+fn+←Delete or ⌘+Delete→ to Send Hex Codes:
0x0b
I wrote a comprehensive guide to adding most of OSX's standard keybinding to your terminal here
In OSX, ⌥+⌫ and ⌘+⌫ are the shortcuts for deleting a word and deleting a line respectively. ⌘+ ← and ⌘+ → are for going to the beginning and end of lines. By default, iTerm2 isn't configured this way, and there are a lot of misleading guides online. The following is what I've found to work on my machine.
Open the preferences (⌘+,) and go to the Keys tab.
Add a global shortcut key, and just type in your shortcut
In the Action dropdown, select Send Hex Code
The hex codes for...
Deleting a word: 0x17.
Deleting a line: 0x15.
Moving to the beginning of the line: 0x01.
Moving to the end of the line: 0x05.
Just open a new tab, and it should work!
Here is a screenshot, for clarity:
I hope this may help you
map ⌥ <- Delete to Send Hex Codes: 0x1B 0x08
I had test for it, and it is correct.
18.09.2013 update
this delete one word, not a line.
As pointed ^U deletes the line. You can easily remap the command by using Better Touch Tool.
It also has cool features for automation, mouse, pad and keyboard mapping. Also includes a window feature for smart borders.
On iTerm2 you can set the exact same shortcuts of your OS. In this case, as the default shortcut to delete a line on Mac OS is ⌘+Delete (backspace), you can do so.
To set the default keybindings of iTerm you have to:
Go to Preferences (or ⌘+,)
Profiles
Keys
Key Mappings
Presets...
Select the "Natural text editing" option
The location has changed. In order to enable natural editing go to:
Preferences -> Profile -> Keys -> Load Presets... -> and select Natural Text Editing
For more info and tweeks go https://blog.arturofm.com/install-iterm-terminal-emulator-on-macos/

Is there a Textmate bundle to replace umlauts?

I often find myself going through text files replacing German umlauts, for example:
ä -> ae
ü -> ue
ö -> oe
ß -> sz
the same applies for their capitalized variants
Going through find/replace for this is tedious, so I wonder if there's a bundle for this or if someone could come up with a small code snippet?
Although I don't know of any bundle that does that specific task, remember that TextMate has the ability to execute any scripting language that recognizes the shebang (#!) sequence as a language indicator, so if you're familiar with PHP, Perl, bash, Ruby or anything else that can use it, you can simply write a command, perhaps adding it to the existing Text bundle. Alternatively, you could record a macro (Bundles>Macros>Start Recording) of the steps you take to accomplish the find/replace.
If you are editing an HTML or PHP file you just need to hit ⌘+& to replace all "special characters" in your selection with their HTML Entities counterpart.
If it's not an HTML document you can still use the cog menu at the bottom of the window and navigate to HTML -> Entities -> Convert Selection to Entities.
Edit:
You can do this by recording and saving a macro. Here is how I'd do it:
Start recording your macro
Bundle -> Macros -> Start Recording
Do consecutive searches & replaces for all the special characters you want to convert.
Stop recording.
Save the macro and give it a keyboard shortcut.

Resources