Jekyll extracting filename from url with Ruby

Question

I'm trying to extract the file name from a url on Jekyll using this ruby snippet:

{% assign filename = page.url.split('/')[-1] | replace: '.html', '.md' %}

If I just use:

{% assign filename = page.url | replace: '.html', '.md' %}

I get back the url with the replaced file type but my .split('/')[-1] doesn't seem to work.

I tried running the following in standalone ruby to ensure my syntax was right, and it returned bird as expected:

"cat/dog/b­ird".split­('/')[-1]

Why doesn't the same syntax work in my Jekyll instance? Is it that page.url isn't a string, or something else?

Answer 1

The problem is mixing ruby code with Liquid tags.

To extract the filename from a url in Jekyll you can use just pure Liquid template filters, using the equivalents of what you tried:

• .split­('/') -> | split: '/'
• [-1] -> | last

As an example with a custom URL:

{% assign url_example = "cat/dog/bird.html" %}
{% assign filename = url_example | split: '/' | last | replace: '.html', '.md' %}
{{filename}}

outputs:

bird.md