Nested Collections

[zachleat]

I believe some other tools call this taxonomies

This could enable some other neat things like nested levels of pagination. An example of a workaround for two levels of pagination can be found here #332

Sold best by this tweet: https://twitter.com/sirinath/status/1230855052547514369

Tags can be hierarchically bundled together by different criteria. E.g. tag 'rust' in category 'programming' can be kept sperate from tag 'rust' in 'metallurgy'. 'Star' in 'music' can be separated from 'star' in 'cosmology' in a 'series' of posts on 'raggae'.

See also

• Hugo https://gohugo.io/content-management/taxonomies/
• WordPress https://developer.wordpress.org/themes/basics/categories-tags-custom-taxonomies/

[zachleat]

This repository is now using lodash style issue management for enhancements. This means enhancement issues will now be closed instead of leaving them open.

View the enhancement backlog here. Don’t forget to upvote the top comment with 👍!

[sirinath]

Best writeup on this is:

• 5 steps to organizing your blog’s content—taxonomy simplified - scroll down for the content to appear.

Also might be interesting:

• What Is Taxonomy, and Why Does It Matter?

Other tools which have it:

• Hugo Taxonomies
• Zola Taxonomies
• WordPress - Categories, Tags, & Custom Taxonomies

[paulshryock]

I feel like custom taxonomies are already built into eleventy, even if it's not documented as such.

For example:

---
title: 'Hello World'
contentType: 'article'
category: 'news'
tags: [ 'greeting', 'friendly' ]
---

In .eleventy.js, create an articles collection:

module.exports = function (eleventyConfig) {
  eleventyConfig.addCollection('articles', collection => {
    return collection.getAll().filter(post => post.data.contentType === 'article'))
  })
}

Then in a Javascript data file, query and filter data as needed.

data.articles = collections.articles.filter((article) => {
  return article.category === 'news' && article.tags.includes('greeting')
})

Something like that?

Although I'm not sure if it's possible to access the Eleventy collections object from JavaScript data files.

[sielay]

I don't want here to promote my plugin, but you can find here examples how to create taxonomies and paginate on them. I'm still to make few improvements. First thing I'd like do anyhow is to allow disabling native support for tags, as it makes such approach a bit more problematic.

https://github.com/sielay/eleventy-plugin-blog/blob/master/index.js#L156

All is mutation of idea promoted by @jeromecoupe https://www.webstoemp.com/blog/basic-custom-taxonomies-with-eleventy/

[hidegh]

@zachleat I'd also consider nested collections, which HEXO has out of the box e.g.: /collections/technical/testing/1 or /collections/technical/testing/unit-tests/1

Again here (beside nesting) we can have 2 things to solve:

1. do we want to display items (and counts) to a category (we might have just a pure nested category listing)
2. if so, do we want to display immediate items only (or do we want to include items that belong to the subcategories of the currently selected category)

Paging can be also interesting, as we could page over:

1. the merged sub-category + item list
2. only over items

[m4rrc0]

Hey everyone. After doing some digging, I think I can safely say that this feature request's importance is being overlooked. I have found many Github issues pointing to the same kind of request. Of course, some are duplicates but you can look at it any way you want, many people have this problem and different solutions are used while not exactly knowing what the impact is on build performance, maintainability, flexibility of the data structure, ...

In the enhancement backlog alone, nr 7 and 8 (in order of number of votes) point to the same request.

Then there are requests like Programmatically create pages from data files #136 which in fact point to the same issue of lacking control over how pages are built and/or dissociating the relation between 1 file => 1 flat collection.

I totally get that not everyone needs this feature and that there are workarounds. It just seems to me that there are more people needing this than meets the eye.

To try and make my case, and in an effort of compiling what has been discussed on the topic, here are the issues and discussions I could find about this topic (sorry this isn't sorted in any relevant manner):

• Double-layered pagination? #332
• Allow pagination over multiple data #1412
• add support for multi pagination #1559
• Core pagination support for double layered pagination #461
• Pagination on nested data #294
• Is it possible to use multiple paginations? #308
• Multi-tier pagination for 'dynamic page' output #427
• Collections for directory indexes #502
• Generating a post tag based on the post date #316
• Let me define my own pagination function using JS #772
• Organising content into a set of pages with a calendar type structure #964
• What's the correct way to get data to a layout so I can do a "groupby" on it? #1357
• How do you list all posts grouped by year? #1284
• I just converted a site from jekyll to 11ty. How do you do blog categories #776
• How do we paginate categories. #793
• Nested Collections #955
• Navigate through specific tag in double layered pagination #1120
• How to split by pages paginated data? #2141
• Pagination: let data key be a function #1711
• Pagination: bind this to before callback #1697
• Let me transform pagination data using JS before it is paginated #699

Non core projects trying to solve this:

• https://github.com/jeromecoupe/11ty-paginate-taxonomies with this article covering the method I believe
• https://github.com/sielay/eleventy-plugin-blog
• https://github.com/solution-loisir/eleventy-deep-pagination with this demo website
• https://gist.github.com/MWDelaney/6dd60be5e4c223217d597983d20bde20
• 📝 Blog MWDelaney/ZeroPoint#31

I challenge everyone to find more (non-trivial) issues on any other topic! 🤣

I think all those requests have found workarounds to make their use cases work but there is still no canonical way to handle this in Eleventy I believe (with all the obvious short and long-term consequences of hacking around an issue).

Isn't it time to have a proper discussion about this? 😄

(And in the meantime I'll find my own workaround 😅)