I have just started with Pelican. It's awesome, I just can't figure out how to use macros in my articles (and pages). I know I can use Jinja when making my own theme, but I can't seem to be able to use it in articles. I'd like to be able to define a macro/function/template/whatever that I'd put directly in the markdown of the article, possibly with parameters, and it would get expanded when the pages are generated. For example a function to generate an image with a caption of given size and css class that would also be a link directly to the image. I'd like to be able to access these macros from all articles, to be able to reuse them everywhere. Something I'd normally do with PhP.
I could probably use JS to do this, but if I'd like to avoid it and keep everything static if possible. Can this be done?
UPDATE:
I found a pelican plugin that solves this - jinja2content.
OLD SOLUTION:
I found a solution here. You can implement a filter in Python to process all texts in articles/pages like this:
Create a python file filters.py in which you write the filter function process_text
to expand my macros (or generally do anything with the article/page text), for example to test the function write something like:
def process_text(input_text):
return "TEST " + input_text
In the Pelican config file (pelicanconfig.py) register this function as a possible filter to be used with Jinja:
import sys
sys.path.append('.')
import filters
JINJA_FILTERS = {'process_text':filters.process_text}
Now you have to edit the templates to apply this filter to article/page texts before adding them to output. In my case I edited two files: themes/themename/templates/article.html and themes/themename/templates/post.html and changed {{ article.content }}
to {{ article.content|process_text }}
and {{ page.content }}
to {{ page.content|process_text }}
in them to apply the filter.
Now all texts in articles and pages should be prefixed with "TEST".
The not so convenient thing about this is I have to write my own macro expander, which shouldn't be extremely hard with regular expression in Python, but if there is a nicer way to do this, feel free to post here.