Feeds are a powerful feature of markata that allow you to build feeds of
posts from the same arguments that get passed into Markata.map, all within
markata.toml. You can create feeds for things like your published articles,
articles that contain a certain tag, articles that are stored in a certain
folder, articles posted before or after a certain date (likely today).
Load up ipython
or python or a script that you run, whatever your fancy
Now create an instance of markata.
from markata import Markta
m = Markata()
Let's poke through our articles
Now you should be able to list out the articles, for this site it will be something like.
m.articles
# [
# <frontmatter.Post object at 0x7f609831b0a0>,
# <frontmatter.Post object at 0x7f6098318b80>,
# <frontmatter.Post object at 0x7f609831a890>,
# <frontmatter.Post object at 0x7f609831ab30>,
# <frontmatter.Post object at 0x7f609831aa70>,
# <frontmatter.Post object at 0x7f609831b760>
# ]
Now let's say you want the title for all the articles, you could do this with a list comp.
[post['title'] for post in markata.articles]
# ['Building the Site', 'Home Page', 'Loading Markata into Ipython', 'Jinja Variables', 'Your Own Style', '404']
Now if you wanted to filter for only published posts, you can filter by the boolean variable published.
[post['title'] for post in markata.articles if published]
# ['Building the Site', 'Home Page', 'Loading Markata into Ipython', 'Jinja Variables', 'Your Own Style']
And a little more arbitrary, and slightly more complex filter.
[post['title'] for post in markata.articles if "y" in post['path']]
# ['Loading Markata into Ipython', 'Your Own Style']
Using map
The same results can be achieved with the Markata.map function. Again let's
start by getting the title.
[post['title'] for post in markata.articles if published]
markata.map('title')
# ['Building the Site', 'Home Page', 'Loading Markata into Ipython', 'Jinja Variables', 'Your Own Style', '404']
Then we can add the filter for published posts back in.
markata.map('title', filter='published')
# ['Building the Site', 'Home Page', 'Loading Markata into Ipython', 'Jinja Variables', 'Your Own Style', '404']
And the more arbitrary filter looking for "y" in the article's path.
markata.map('title', filter='"y" in post["path"]')
# ['Loading Markata into Ipython', 'Your Own Style']
Since each article is unpacked into the map function, path is also directly
available to the filter, so we don't necessarily need to reach for the post.
markata.map('title', filter='"y" in path')
# ['Loading Markata into Ipython', 'Your Own Style']
Lastely we can also sort on date, by adding a sort argument.
markata.map('title', filter='published', sort='date')
# ['Building the Site', 'Home Page', 'Loading Markata into Ipython', 'Jinja Variables', 'Your Own Style', '404']
What does this have to do with feeds.
Markata can make a feed page, displaying each post in the returned list inside of an html feed. It does this by unpacking the arguments from the feeds config into the map function.
[markata.feeds.published]
# creates a feed at /published
filter="date<=today and post.get('published', False)"
sort="date"