bjoernager/rust
bjoernager/rust
1
Fork 0
Code 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
Download patch file Download diff file Expand all files Collapse all files

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") {