mirror of
https://github.com/zulip/zulip.git
synced 2025-11-06 06:53:25 +00:00
afba77300a
So far, when ordering typeahead suggestions, any query matches that did not occur at the start of the target string were considered equally. So for example, for the query "ok", "squared_ok" and "smoking" were allotted equal priority, which does not make sense. Now, matches from a word boundary (space, hyphen, underscore or slash) are given priority (after exact matches and beginning matches), so that in the above example, "squared_ok" is regarded as a better match than "smoking". Since recipients' sorting is complex, and needs word boundary matches separately for decoupled sorting post triage, `triage_raw` is extracted from `triage` to return the raw matches. Fixes: #24127.
275 lines
8.5 KiB
JavaScript
275 lines
8.5 KiB
JavaScript
"use strict";
|
|
|
|
const {strict: assert} = require("assert");
|
|
|
|
const {zrequire} = require("./lib/namespace");
|
|
const {run_test} = require("./lib/test");
|
|
|
|
const typeahead = zrequire("../shared/src/typeahead");
|
|
|
|
const unicode_emojis = [
|
|
["1f43c", "panda_face"],
|
|
["1f642", "smile"],
|
|
["1f604", "big_smile"],
|
|
["1f368", "ice_cream"],
|
|
["1f366", "soft_ice_cream"],
|
|
["1f6a5", "horizontal_traffic_light"],
|
|
["1f6a6", "traffic_light"],
|
|
["1f537", "large_blue_diamond"],
|
|
["1f539", "small_blue_diamond"],
|
|
];
|
|
|
|
const emojis = [
|
|
{emoji_name: "japanese_post_office", reaction_type: "realm_emoji", url: "TBD"},
|
|
{emoji_name: "tada", reaction_type: "realm_emoji", random_field: "whatever"},
|
|
...unicode_emojis.map(([emoji_code, emoji_name]) => ({
|
|
emoji_name,
|
|
emoji_code,
|
|
reaction_type: "unicode_emoji",
|
|
})),
|
|
];
|
|
|
|
function emoji_matches(query) {
|
|
const matcher = typeahead.get_emoji_matcher(query);
|
|
return emojis.filter((emoji) => matcher(emoji));
|
|
}
|
|
|
|
function assert_emoji_matches(query, expected) {
|
|
const names = emoji_matches(query).map((emoji) => emoji.emoji_name);
|
|
assert.deepEqual(names.sort(), expected);
|
|
}
|
|
|
|
run_test("get_emoji_matcher: nonmatches", () => {
|
|
assert_emoji_matches("notaemoji", []);
|
|
assert_emoji_matches("da_", []);
|
|
});
|
|
|
|
run_test("get_emoji_matcher: misc matches", () => {
|
|
assert_emoji_matches("da", ["panda_face", "tada"]);
|
|
assert_emoji_matches("smil", ["big_smile", "smile"]);
|
|
assert_emoji_matches("mile", ["big_smile", "smile"]);
|
|
assert_emoji_matches("japanese_post_", ["japanese_post_office"]);
|
|
});
|
|
|
|
run_test("matches starting at non-first word, too", () => {
|
|
assert_emoji_matches("ice_cream", ["ice_cream", "soft_ice_cream"]);
|
|
assert_emoji_matches("blue_dia", ["large_blue_diamond", "small_blue_diamond"]);
|
|
assert_emoji_matches("traffic_", ["horizontal_traffic_light", "traffic_light"]);
|
|
});
|
|
|
|
run_test("matches literal unicode emoji", () => {
|
|
assert_emoji_matches("🐼", ["panda_face"]);
|
|
});
|
|
|
|
run_test("get_emoji_matcher: spaces equivalent to underscores", () => {
|
|
function assert_equivalent(query) {
|
|
assert.deepEqual(emoji_matches(query), emoji_matches(query.replace(" ", "_")));
|
|
}
|
|
assert_equivalent("da ");
|
|
assert_equivalent("panda ");
|
|
assert_equivalent("japanese post ");
|
|
assert_equivalent("ice ");
|
|
assert_equivalent("ice cream");
|
|
assert_equivalent("blue dia");
|
|
assert_equivalent("traffic ");
|
|
assert_equivalent("traffic l");
|
|
});
|
|
|
|
run_test("triage", () => {
|
|
const alice = {name: "alice"};
|
|
const alicia = {name: "Alicia"};
|
|
const joan = {name: "Joan"};
|
|
const jo = {name: "Jo"};
|
|
const steve = {name: "steve"};
|
|
const stephanie = {name: "Stephanie"};
|
|
|
|
const names = [alice, alicia, joan, jo, steve, stephanie];
|
|
|
|
assert.deepEqual(
|
|
typeahead.triage("a", names, (r) => r.name),
|
|
{
|
|
matches: [alice, alicia],
|
|
rest: [joan, jo, steve, stephanie],
|
|
},
|
|
);
|
|
|
|
assert.deepEqual(
|
|
typeahead.triage("A", names, (r) => r.name),
|
|
{
|
|
matches: [alicia, alice],
|
|
rest: [joan, jo, steve, stephanie],
|
|
},
|
|
);
|
|
|
|
assert.deepEqual(
|
|
typeahead.triage("S", names, (r) => r.name),
|
|
{
|
|
matches: [stephanie, steve],
|
|
rest: [alice, alicia, joan, jo],
|
|
},
|
|
);
|
|
|
|
assert.deepEqual(
|
|
typeahead.triage("fred", names, (r) => r.name),
|
|
{
|
|
matches: [],
|
|
rest: [alice, alicia, joan, jo, steve, stephanie],
|
|
},
|
|
);
|
|
|
|
assert.deepEqual(
|
|
typeahead.triage("Jo", names, (r) => r.name),
|
|
{
|
|
matches: [jo, joan],
|
|
rest: [alice, alicia, steve, stephanie],
|
|
},
|
|
);
|
|
|
|
assert.deepEqual(
|
|
typeahead.triage("jo", names, (r) => r.name),
|
|
{
|
|
matches: [jo, joan],
|
|
rest: [alice, alicia, steve, stephanie],
|
|
},
|
|
);
|
|
|
|
assert.deepEqual(
|
|
typeahead.triage(" ", names, (r) => r.name),
|
|
{
|
|
matches: [],
|
|
rest: [alice, alicia, joan, jo, steve, stephanie],
|
|
},
|
|
);
|
|
|
|
assert.deepEqual(
|
|
typeahead.triage(";", names, (r) => r.name),
|
|
{
|
|
matches: [],
|
|
rest: [alice, alicia, joan, jo, steve, stephanie],
|
|
},
|
|
);
|
|
});
|
|
|
|
run_test("triage: prioritise word boundary matches to arbitrary substring matches", () => {
|
|
const book = {name: "book"};
|
|
const hyphen_ok = {name: "hyphen_ok"};
|
|
const space_ok = {name: "space ok"};
|
|
const no_space_ok = {name: "nospaceok"};
|
|
const number_ok = {name: "number1ok"};
|
|
const okay = {name: "okay"};
|
|
const ok = {name: "ok"};
|
|
|
|
const emojis = [book, hyphen_ok, space_ok, no_space_ok, number_ok, okay, ok];
|
|
|
|
assert.deepEqual(
|
|
typeahead.triage("ok", emojis, (r) => r.name),
|
|
{
|
|
matches: [ok, okay, hyphen_ok, space_ok],
|
|
rest: [book, no_space_ok, number_ok],
|
|
},
|
|
);
|
|
});
|
|
|
|
function sort_emojis(emojis, query) {
|
|
return typeahead.sort_emojis(emojis, query).map((emoji) => emoji.emoji_name);
|
|
}
|
|
|
|
run_test("sort_emojis: th", () => {
|
|
const emoji_list = [
|
|
{emoji_name: "mother_nature", reaction_type: "realm_emoji"},
|
|
{emoji_name: "thermometer", reaction_type: "realm_emoji"},
|
|
{emoji_name: "thumbs_down", reaction_type: "realm_emoji"},
|
|
{emoji_name: "thumbs_up", reaction_type: "unicode_emoji", emoji_code: "1f44d"},
|
|
];
|
|
assert.deepEqual(sort_emojis(emoji_list, "th"), [
|
|
"thumbs_up",
|
|
"thermometer",
|
|
"thumbs_down",
|
|
"mother_nature",
|
|
]);
|
|
});
|
|
|
|
run_test("sort_emojis: sm", () => {
|
|
const emoji_list = [
|
|
{emoji_name: "big_smile", reaction_type: "realm_emoji"},
|
|
{emoji_name: "slight_smile", reaction_type: "unicode_emoji", emoji_code: "1f642"},
|
|
{emoji_name: "small_airplane", reaction_type: "realm_emoji"},
|
|
];
|
|
assert.deepEqual(sort_emojis(emoji_list, "sm"), [
|
|
"slight_smile",
|
|
"small_airplane",
|
|
"big_smile",
|
|
]);
|
|
});
|
|
|
|
run_test("sort_emojis: SM", () => {
|
|
const emoji_list = [
|
|
{emoji_name: "big_smile", reaction_type: "realm_emoji"},
|
|
{emoji_name: "slight_smile", reaction_type: "unicode_emoji", emoji_code: "1f642"},
|
|
{emoji_name: "small_airplane", reaction_type: "realm_emoji"},
|
|
];
|
|
assert.deepEqual(sort_emojis(emoji_list, "SM"), [
|
|
"slight_smile",
|
|
"small_airplane",
|
|
"big_smile",
|
|
]);
|
|
});
|
|
|
|
run_test("sort_emojis: prefix before midphrase, with underscore (traffic_li)", () => {
|
|
const emoji_list = [
|
|
{emoji_name: "horizontal_traffic_light", reaction_type: "realm_emoji"},
|
|
{emoji_name: "traffic_light", reaction_type: "realm_emoji"},
|
|
];
|
|
assert.deepEqual(sort_emojis(emoji_list, "traffic_li"), [
|
|
"traffic_light",
|
|
"horizontal_traffic_light",
|
|
]);
|
|
});
|
|
|
|
run_test("sort_emojis: prefix before midphrase, with space (traffic li)", () => {
|
|
const emoji_list = [
|
|
{emoji_name: "horizontal_traffic_light", reaction_type: "realm_emoji"},
|
|
{emoji_name: "traffic_light", reaction_type: "realm_emoji"},
|
|
];
|
|
assert.deepEqual(sort_emojis(emoji_list, "traffic li"), [
|
|
"traffic_light",
|
|
"horizontal_traffic_light",
|
|
]);
|
|
});
|
|
|
|
run_test("sort_emojis: remove duplicates", () => {
|
|
// notice the last 2 are aliases of the same emoji (same emoji code)
|
|
const emoji_list = [
|
|
{emoji_name: "laughter_tears", emoji_code: "1f602", reaction_type: "unicode_emoji"},
|
|
{emoji_name: "tear", emoji_code: "1f972", reaction_type: "unicode_emoji"},
|
|
{emoji_name: "smile_with_tear", emoji_code: "1f972", reaction_type: "unicode_emoji"},
|
|
];
|
|
assert.deepEqual(typeahead.sort_emojis(emoji_list, "tear"), [emoji_list[1], emoji_list[0]]);
|
|
});
|
|
|
|
run_test("sort_emojis: prioritise realm emojis", () => {
|
|
const emoji_list = [
|
|
{emoji_name: "thank_you", emoji_code: "1f64f", reaction_type: "unicode_emoji"},
|
|
{
|
|
emoji_name: "thank_you_custom",
|
|
url: "something",
|
|
is_realm_emoji: true,
|
|
reaction_type: "realm_emoji",
|
|
},
|
|
];
|
|
assert.deepEqual(typeahead.sort_emojis(emoji_list, "thank"), [emoji_list[1], emoji_list[0]]);
|
|
});
|
|
|
|
run_test("sort_emojis: prioritise perfect matches", () => {
|
|
const emoji_list = [
|
|
{emoji_name: "thank_you", emoji_code: "1f64f", reaction_type: "unicode_emoji"},
|
|
{
|
|
emoji_name: "thank_you_custom",
|
|
url: "something",
|
|
is_realm_emoji: true,
|
|
reaction_type: "realm_emoji",
|
|
},
|
|
];
|
|
assert.deepEqual(typeahead.sort_emojis(emoji_list, "thank you"), emoji_list);
|
|
});
|