nordic-dev.net/rust
nordic-dev.net/rust
2
0
Fork 0
mirror of https://github.com/rust-lang/rust.git synced 2024-11-26 00:34:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

rustdoc: tweak the search index format

Browse Source 
This essentially switches search-index.js from a "array of struct"
to a "struct of array" format, like this:

    {
    "doc": "Crate documentation",
    "t": [ 1, 1, 2, 3, ... ],
    "n": [ "Something", "SomethingElse", "whatever", "do_stuff", ... ],
    "q": [ "a::b", "", "", "", ... ],
    "d": [ "A Struct That Does Something", "Another Struct", "a function", "another function", ... ],
    "i": [ 0, 0, 1, 1, ... ],
    "f": [ null, null, [], [], ... ],
    "p": ...,
    "a": ...
    }

So `{ty: 1, name: "Something", path: "a::b", desc: "A Struct That Does Something", parent_idx: 0, search_type: null}` is the first item.

This makes the uncompressed version smaller, but it really shows on the
compressed version:

    notriddle:rust$ wc -c new-search-index1.52.0.js
    2622427 new-search-index1.52.0.js
    notriddle:rust$ wc -c old-search-index1.52.0.js
    2725046 old-search-index1.52.0.js
    notriddle:rust$ gzip new-search-index1.52.0.js
    notriddle:rust$ gzip old-search-index1.52.0.js
    notriddle:rust$ wc -c new-search-index1.52.0.js.gz
    239385 new-search-index1.52.0.js.gz
    notriddle:rust$ wc -c old-search-index1.52.0.js.gz
    296328 old-search-index1.52.0.js.gz
    notriddle:rust$

That's a 4% improvement on the uncompressed version (fewer `[]`),
and 20% improvement after gzipping it, thanks to putting like-typed
data next to each other. Any compression algorithm based on a sliding
window will probably show this kind of improvement.
This commit is contained in:
Michael Howell 2021-03-10 19:31:49 -07:00
parent f98721f886
commit 3934dd1b3e
3 changed files with 74 additions and 42 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

60
src/librustdoc/html/render/cache.rs
View File

@ -4,7 +4,7 @@ use std::path::Path;
use rustc_data_structures::fx::{FxHashMap, FxHashSet};
use rustc_middle::ty::TyCtxt;
use rustc_span::symbol::{sym, Symbol};
use serde::Serialize;
use serde::ser::{Serialize, SerializeStruct, Serializer};
use crate::clean::types::{
FnDecl, FnRetTy, GenericBound, Generics, GetDefId, Type, TypeKind, WherePredicate,
@ -133,21 +133,69 @@ crate fn build_index<'tcx>(krate: &clean::Crate, cache: &mut Cache, tcx: TyCtxt<
.map(|module| module.doc_value().map_or_else(String::new, |s| short_markdown_summary(&s)))
.unwrap_or_default();
#[derive(Serialize)]
struct CrateData<'a> {
doc: String,
#[serde(rename = "i")]
items: Vec<&'a IndexItem>,
#[serde(rename = "p")]
paths: Vec<(ItemType, String)>,
// The String is alias name and the vec is the list of the elements with this alias.
//
// To be noted: the `usize` elements are indexes to `items`.
#[serde(rename = "a")]
#[serde(skip_serializing_if = "BTreeMap::is_empty")]
aliases: &'a BTreeMap<String, Vec<usize>>,
}
impl<'a> Serialize for CrateData<'a> {
fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
where
S: Serializer,
{
let has_aliases = !self.aliases.is_empty();
let mut crate_data =
serializer.serialize_struct("CrateData", if has_aliases { 9 } else { 8 })?;
crate_data.serialize_field("doc", &self.doc)?;
crate_data.serialize_field(
"t",
&self.items.iter().map(|item| &item.ty).collect::<Vec<_>>(),
)?;
crate_data.serialize_field(
"n",
&self.items.iter().map(|item| &item.name).collect::<Vec<_>>(),
)?;
crate_data.serialize_field(
"q",
&self.items.iter().map(|item| &item.path).collect::<Vec<_>>(),
)?;
crate_data.serialize_field(
"d",
&self.items.iter().map(|item| &item.desc).collect::<Vec<_>>(),
)?;
crate_data.serialize_field(
"i",
&self
.items
.iter()
.map(|item| {
assert_eq!(
item.parent.is_some(),
item.parent_idx.is_some(),
"`{}` is missing idx",
item.name
);
item.parent_idx.map(|x| x + 1).unwrap_or(0)
})
.collect::<Vec<_>>(),
)?;
crate_data.serialize_field(
"f",
&self.items.iter().map(|item| &item.search_type).collect::<Vec<_>>(),
)?;
crate_data.serialize_field("p", &self.paths)?;
if has_aliases {
crate_data.serialize_field("a", &self.aliases)?;
}
crate_data.end()
}
}
// Collect the index into a string
format!(
r#""{}":{}"#,

17
src/librustdoc/html/render/mod.rs
View File

@ -166,23 +166,6 @@ crate struct IndexItem {
crate search_type: Option<IndexItemFunctionType>,
}
impl Serialize for IndexItem {
fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
where
S: Serializer,
{
assert_eq!(
self.parent.is_some(),
self.parent_idx.is_some(),
"`{}` is missing idx",
self.name
);
(self.ty, &self.name, &self.path, &self.desc, self.parent_idx, &self.search_type)
.serialize(serializer)
}
}
/// A type used for the search index.
#[derive(Debug)]
crate struct RenderType {

39
src/librustdoc/html/static/main.js
View File

@ -1847,13 +1847,18 @@ function defocusSearchBar() {
});
currentIndex += 1;
// an array of [(Number) item type,
// (String) name,
// (String) full path or empty string for previous path,
// (String) description,
// (Number | null) the parent path index to `paths`]
// (Object | null) the type of the function (if any)
var items = rawSearchIndex[crate].i;
// an array of (Number) item types
var itemTypes = rawSearchIndex[crate].t;
// an array of (String) item names
var itemNames = rawSearchIndex[crate].n;
// an array of (String) full paths (or empty string for previous path)
var itemPaths = rawSearchIndex[crate].q;
// an array of (String) descriptions
var itemDescs = rawSearchIndex[crate].d;
// an array of (Number) the parent path index + 1 to `paths`, or 0 if none
var itemParentIdxs = rawSearchIndex[crate].i;
// an array of (Object | null) the type of the function, if any
var itemFunctionSearchTypes = rawSearchIndex[crate].f;
// an array of [(Number) item type,
// (String) name]
var paths = rawSearchIndex[crate].p;
@ -1867,28 +1872,24 @@ function defocusSearchBar() {
paths[i] = {ty: paths[i][0], name: paths[i][1]};
}
// convert `items` into an object form, and construct word indices.
// convert `item*` into an object form, and construct word indices.
//
// before any analysis is performed lets gather the search terms to
// search against apart from the rest of the data. This is a quick
// operation that is cached for the life of the page state so that
// all other search operations have access to this cached data for
// faster analysis operations
len = items.length;
len = itemTypes.length;
var lastPath = "";
for (i = 0; i < len; ++i) {
var rawRow = items[i];
if (!rawRow[2]) {
rawRow[2] = lastPath;
}
var row = {
crate: crate,
ty: rawRow[0],
name: rawRow[1],
path: rawRow[2],
desc: rawRow[3],
parent: paths[rawRow[4]],
type: rawRow[5],
ty: itemTypes[i],
name: itemNames[i],
path: itemPaths[i] ? itemPaths[i] : lastPath,
desc: itemDescs[i],
parent: itemParentIdxs[i] > 0 ? paths[itemParentIdxs[i] - 1] : undefined,
type: itemFunctionSearchTypes[i],
};
searchIndex.push(row);
if (typeof row.name === "string") {