Finding Markdown image tags with regular expressions (E486)

Question

I have a document that contains lines like

![My title](/path/to/image.jpg)

This is the format for images in Markdown.

I have constructed a regular expression which I believe should be able to find the image tags:

!\[(.*)\]\((.+)\)

But when I try the command /!\[(.*)\]\((.+)\), I get E486: Pattern not found: !\[(.*)\]\((.+)\).

---

Motivation

I am trying to do this to change how images are presented in my Hugo static site. I want to be able to search and replace everywhere in a file with

%s/!\[(.*)\]\((.+)\)/{{< figure alt="\1" src="\2" >}}/

If it matters, I'm using Vim version 8.0.1283.

Answer 1

I use "very magic" setting (\v) so that I can use normal regex patterns and not have to convert it into vim's awkward regex format.

First, I would do search as the original regex search term, but prepend with \v to enable "very magic" regex setting:

/\v!\[(.*)\]\((.+)\)

After testing it to make sure it is working correctly, I'd finally do search/replace operation by:

:%s/<ctrl-r>//{{< figure alt="\1" src="\2" >}}/

• <ctrl-r>/ will paste the original search regex, so you don't have to retype it.

Result will be:

:%s/\v!\[(.*)\]\((.+)\)/{{< figure alt="\1" src="\2" >}}/ 

Answer 2

Vim doesn't require escaping of parentheses by default.

As Christian says:

[I]f you want to capture a group, you need to use \(\) instead of () and for capturing at least one of the previous items, you need to use \+ instead of +.

Using this modification, my regular expression becomes

!\[\(.*\)\](\(.\+\))

The parentheses which were part of capture groups have been escaped, as has the +, and the parentheses which were literals have been unescaped. This new expression works as intended.

My full search and replace is now

:%s/!\[\(.*\)\](\(.\+\))/{{< figure alt="\1" src="\2" >}}/