os-rust/tests/rustdoc-js/non-english-identifier.js

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

178 lines
5.1 KiB
JavaScript
Raw Permalink Normal View History

2024-06-07 05:49:46 +08:00
const PARSED = [
{
query: '中文',
elems: [{
name: "中文",
fullPath: ["中文"],
pathWithoutLast: [],
pathLast: "中文",
generics: [],
typeFilter: -1,
}],
returned: [],
foundElems: 1,
userQuery: "中文",
error: null,
},
{
query: '_0Mixed中英文',
elems: [{
name: "_0Mixed中英文",
2024-06-07 05:49:46 +08:00
fullPath: ["_0mixed中英文"],
pathWithoutLast: [],
pathLast: "_0mixed中英文",
normalizedPathLast: "0mixed中英文",
2024-06-07 05:49:46 +08:00
generics: [],
typeFilter: -1,
}],
foundElems: 1,
userQuery: "_0Mixed中英文",
2024-06-07 05:49:46 +08:00
returned: [],
error: null,
},
{
query: 'my_crate::中文API',
elems: [{
name: "my_crate::中文API",
2024-06-07 05:49:46 +08:00
fullPath: ["my_crate", "中文api"],
pathWithoutLast: ["my_crate"],
pathLast: "中文api",
generics: [],
typeFilter: -1,
}],
foundElems: 1,
userQuery: "my_crate::中文API",
2024-06-07 05:49:46 +08:00
returned: [],
error: null,
},
{
query: '类型A,类型B<约束C>->返回类型<关联类型=路径::约束D>',
elems: [{
name: "类型A",
2024-06-07 05:49:46 +08:00
fullPath: ["类型a"],
pathWithoutLast: [],
pathLast: "类型a",
generics: [],
}, {
name: "类型B",
2024-06-07 05:49:46 +08:00
fullPath: ["类型b"],
pathWithoutLast: [],
pathLast: "类型b",
generics: [{
name: "约束C",
2024-06-07 05:49:46 +08:00
fullPath: ["约束c"],
pathWithoutLast: [],
pathLast: "约束c",
generics: [],
}],
}],
foundElems: 3,
totalElems: 5,
literalSearch: true,
userQuery: "类型A,类型B<约束C>->返回类型<关联类型=路径::约束D>",
2024-06-07 05:49:46 +08:00
returned: [{
name: "返回类型",
fullPath: ["返回类型"],
pathWithoutLast: [],
pathLast: "返回类型",
generics: [],
bindings: [["关联类型", [{
name: "路径::约束D",
fullPath: ["路径", "约束d"],
pathWithoutLast: ["路径"],
pathLast: "约束d",
generics: [],
}]]],
2024-06-07 05:49:46 +08:00
}],
error: null,
},
{
query: 'my_crate 中文宏!',
elems: [{
name: "my_crate 中文宏",
fullPath: ["my_crate", "中文宏"],
pathWithoutLast: ["my_crate"],
pathLast: "中文宏",
generics: [],
typeFilter: 16,
}],
foundElems: 1,
userQuery: "my_crate 中文宏!",
returned: [],
2024-06-07 05:49:46 +08:00
error: null,
},
{
query: '非法符号——',
elems: [],
foundElems: 0,
userQuery: "非法符号——",
returned: [],
2024-06-07 11:42:17 +08:00
error: "Unexpected `—` after `号` (not a valid identifier)",
2024-06-07 05:49:46 +08:00
}
]
const EXPECTED = [
{
query: '加法',
others: [
{
name: "add",
path: "non_english_identifier",
is_alias: true,
alias: "加法",
href: "../non_english_identifier/macro.add.html"
},
{
name: "add",
path: "non_english_identifier",
is_alias: true,
alias: "加法",
href: "../non_english_identifier/fn.add.html"
},
{
name: "加法",
path: "non_english_identifier",
href: "../non_english_identifier/trait.加法.html",
desc: "Add"
rustdoc: use a trie for name-based search Preview and profiler results ---------------------------- Here's some quick profiling in Firefox done on the rust compiler docs: - Before: https://share.firefox.dev/3UPm3M8 - After: https://share.firefox.dev/40LXvYb Here's the results for the node.js profiler: - https://notriddle.com/rustdoc-html-demo-15/trie-perf/index.html Here's a copy that you can use to try it out. Compare it with [the nightly]. Try typing `typecheckercontext` one character at a time, slowly. - https://notriddle.com/rustdoc-html-demo-15/compiler-doc-trie/index.html [the nightly]: https://doc.rust-lang.org/nightly/nightly-rustc/ The fuzzy match algo is based on [Fast String Correction with Levenshtein-Automata] and the corresponding implementation code in [moman] and [Lucene]; the bit-packing representation comes from Lucene, but the actual matcher is more based on `fsc.py`. As suggested in the paper, a trie is used to represent the FSA dictionary. The same trie is used for prefix matching. Substring matching is done with a side table of three-character[^1] windows that point into the trie. [Fast String Correction with Levenshtein-Automata]: https://github.com/tpn/pdfs/blob/master/Fast%20String%20Correction%20with%20Levenshtein-Automata%20(2002)%20(10.1.1.16.652).pdf [Lucene]: https://fossies.org/linux/lucene/lucene/core/src/java/org/apache/lucene/util/automaton/Lev1TParametricDescription.java [moman]: https://gitlab.com/notriddle/moman-rustdoc User-visible changes -------------------- I don't expect anybody to notice anything, but it does cause two changes: - Substring matches, in the middle of a name, only apply if there's three or more characters in the search query. - Levenshtein distance limit now maxes out at two. In the old version, the limit was w/3, so you could get looser matches for queries with 9 or more characters[^1] in them. [^1]: technically utf-16 code units
2024-11-13 10:46:27 -07:00
}],
in_args: [{
name: "加上",
path: "non_english_identifier::加法",
href: "../non_english_identifier/trait.加法.html#tymethod.加上",
}],
returned: [],
},
{ // levensthein and substring checking only kick in at three characters
query: '加法宏',
others: [
2024-06-07 05:49:46 +08:00
{
name: "中文名称的加法宏",
path: "non_english_identifier",
href: "../non_english_identifier/macro.中文名称的加法宏.html",
rustdoc: use a trie for name-based search Preview and profiler results ---------------------------- Here's some quick profiling in Firefox done on the rust compiler docs: - Before: https://share.firefox.dev/3UPm3M8 - After: https://share.firefox.dev/40LXvYb Here's the results for the node.js profiler: - https://notriddle.com/rustdoc-html-demo-15/trie-perf/index.html Here's a copy that you can use to try it out. Compare it with [the nightly]. Try typing `typecheckercontext` one character at a time, slowly. - https://notriddle.com/rustdoc-html-demo-15/compiler-doc-trie/index.html [the nightly]: https://doc.rust-lang.org/nightly/nightly-rustc/ The fuzzy match algo is based on [Fast String Correction with Levenshtein-Automata] and the corresponding implementation code in [moman] and [Lucene]; the bit-packing representation comes from Lucene, but the actual matcher is more based on `fsc.py`. As suggested in the paper, a trie is used to represent the FSA dictionary. The same trie is used for prefix matching. Substring matching is done with a side table of three-character[^1] windows that point into the trie. [Fast String Correction with Levenshtein-Automata]: https://github.com/tpn/pdfs/blob/master/Fast%20String%20Correction%20with%20Levenshtein-Automata%20(2002)%20(10.1.1.16.652).pdf [Lucene]: https://fossies.org/linux/lucene/lucene/core/src/java/org/apache/lucene/util/automaton/Lev1TParametricDescription.java [moman]: https://gitlab.com/notriddle/moman-rustdoc User-visible changes -------------------- I don't expect anybody to notice anything, but it does cause two changes: - Substring matches, in the middle of a name, only apply if there's three or more characters in the search query. - Levenshtein distance limit now maxes out at two. In the old version, the limit was w/3, so you could get looser matches for queries with 9 or more characters[^1] in them. [^1]: technically utf-16 code units
2024-11-13 10:46:27 -07:00
}],
in_args: [],
returned: [],
},
{ // levensthein and substring checking only kick in at three characters
query: '加法A',
others: [
2024-06-07 05:49:46 +08:00
{
name: "中文名称的加法API",
path: "non_english_identifier",
href: "../non_english_identifier/fn.中文名称的加法API.html",
}],
rustdoc: use a trie for name-based search Preview and profiler results ---------------------------- Here's some quick profiling in Firefox done on the rust compiler docs: - Before: https://share.firefox.dev/3UPm3M8 - After: https://share.firefox.dev/40LXvYb Here's the results for the node.js profiler: - https://notriddle.com/rustdoc-html-demo-15/trie-perf/index.html Here's a copy that you can use to try it out. Compare it with [the nightly]. Try typing `typecheckercontext` one character at a time, slowly. - https://notriddle.com/rustdoc-html-demo-15/compiler-doc-trie/index.html [the nightly]: https://doc.rust-lang.org/nightly/nightly-rustc/ The fuzzy match algo is based on [Fast String Correction with Levenshtein-Automata] and the corresponding implementation code in [moman] and [Lucene]; the bit-packing representation comes from Lucene, but the actual matcher is more based on `fsc.py`. As suggested in the paper, a trie is used to represent the FSA dictionary. The same trie is used for prefix matching. Substring matching is done with a side table of three-character[^1] windows that point into the trie. [Fast String Correction with Levenshtein-Automata]: https://github.com/tpn/pdfs/blob/master/Fast%20String%20Correction%20with%20Levenshtein-Automata%20(2002)%20(10.1.1.16.652).pdf [Lucene]: https://fossies.org/linux/lucene/lucene/core/src/java/org/apache/lucene/util/automaton/Lev1TParametricDescription.java [moman]: https://gitlab.com/notriddle/moman-rustdoc User-visible changes -------------------- I don't expect anybody to notice anything, but it does cause two changes: - Substring matches, in the middle of a name, only apply if there's three or more characters in the search query. - Levenshtein distance limit now maxes out at two. In the old version, the limit was w/3, so you could get looser matches for queries with 9 or more characters[^1] in them. [^1]: technically utf-16 code units
2024-11-13 10:46:27 -07:00
in_args: [],
2024-06-07 05:49:46 +08:00
returned: [],
},
{ // Extensive type-based search is still buggy, experimental & work-in-progress.
query: '可迭代->可选',
others: [{
name: "总计",
path: "non_english_identifier",
href: "../non_english_identifier/fn.总计.html",
desc: "“sum”"
}],
in_args: [],
returned: [],
},
2024-06-07 06:09:30 +08:00
];