1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
//! Classification: the named **collections** and **taxonomies** that templates
//! (`collection()` / `taxonomy()`) and archives read. It is split across the
//! pipeline so each half runs when its inputs are ready:
//!
//! - [`collections`] runs **before markup**. Collection membership is pure
//! frontmatter metadata (path glob + ordering), so it needs nothing from the
//! render. Computing it early lets the defaults phase fill each collection's
//! members before their bodies render.
//! - [`taxonomies`] runs **after markup**, because the inline `#hashtag` pass
//! mutates `doc.terms` during markup.
//! - [`backlinks`] runs **after markup**, because it inverts `doc.links`, which
//! wikilink resolution populates during markup.
//!
//! Both define onto the live index in place; generated archive pages are added
//! later and are deliberately never classified, so `collection()`/`taxonomy()`
//! only ever list authored docs.
use crateConfig;
use crateDocIndex;
/// Evaluate every configured collection query and cache its membership.
/// Bucket every configured taxonomy's terms.
/// Invert `doc.links` into the cached backlink graph that the `backlinks` and
/// `related` template filters read.