Vim-snippet provides a lot of useful Ultisnips snippets for various filetypes. For example, for Markdown, to insert level 1 header, the snippet trigger is sec, and for level 2 and 3 header, the triggers are ssec and sssec respectively. These triggers are non-intuitive and hard to remember and type.

Today, I want to share how to create just one Markdown header snippet using regex keyword and Python interpolation1 with Ultisnips. The final effect of the snippet is that you can type h{NUM} (where NUM is a header level) and press Tab will create the corresponding headers.

# Creating a header snippet

## The regex keyword

In order to create a single snippet with chaning trigger keyword, we need to use regex in the trigger word. So the first line of the snippet looks like this:

snippet "h([1-6])" "Markdown header" br


Note the br flag at the end of the line:

• b: This snippet only triggers at the beginning of a line.
• r: This snippet contains regex and is not a normal snippet.

The idea is that we will capture the number after h in the begining of a line. This number must be in the range [1-6] to trigger the snippet since there are only six header levels in Markdown.

Also note that, unlike other snippet trigger, the regex trigger word needed to be quoted with double quotes.

## A simple snippet

Combining regex keyword and Python interpolation, we can create a simple header snippet like the following one:

snippet "h([1-6])" "Markdown header" br
!p snip.rv = int(match.group(1))*'#' ${1: Section Title} !p snip.rv = int(match.group(1))*'#'$0
endsnippet


In the above snippet, we use two same Python interopolation blocks to create the left and right header level marker, and a regular snippet text is between them.

Inside Python interpolation block, Ultisnips provides the match object, which is the captured result from regex keyword. match.group(1) will be the content of the first capture group, and in this case, it will be the number in string format. The rest of the code is self-explaining so I will not elaborate.

## A complex one using post_jump

A more advanced way is to use post_jump actions, which can perform various complicated logics inside a Python function.

With post_jump, the header snippet now looks like this:

global !p
placeholders_string = snip.buffer[snip.line].strip()
level = int(placeholders_string[0])

# erase current line
snip.buffer[snip.line] = ""
line_content = "#"*level + " ${1:Section Name} " + "#"*level line_content += '\n\n$0'

snip.expand_anon(line_content)
endglobal

!p snip.rv = match.group(1)

In the post jump code, snip.buffer is the object representing the current buffer and is indexable. snip.line is the line in which the snippet expand.
snip.expand_anon() will accept the snippet content and expand it, just like you have typed it in the snippet body.