Files
2026-07-13 12:25:07 +08:00

144 lines
7.2 KiB
JavaScript

// ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
// ┃ ██████ ██████ ██████ █ █ █ █ █ █▄ ▀███ █ ┃
// ┃ ▄▄▄▄▄█ █▄▄▄▄▄ ▄▄▄▄▄█ ▀▀▀▀▀█▀▀▀▀▀ █ ▀▀▀▀▀█ ████████▌▐███ ███▄ ▀█ █ ▀▀▀▀▀ ┃
// ┃ █▀▀▀▀▀ █▀▀▀▀▀ █▀██▀▀ ▄▄▄▄▄ █ ▄▄▄▄▄█ ▄▄▄▄▄█ ████████▌▐███ █████▄ █ ▄▄▄▄▄ ┃
// ┃ █ ██████ █ ▀█▄ █ ██████ █ ███▌▐███ ███████▄ █ ┃
// ┣━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┫
// ┃ Copyright (c) 2017, the Perspective Authors. ┃
// ┃ ╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌ ┃
// ┃ This file is part of the Perspective library, distributed under the terms ┃
// ┃ of the [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0). ┃
// ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┛
import { test, expect } from "@perspective-dev/test";
import perspective from "../perspective_client";
// Concrete use cases from issue #1527 ("Better string functions in
// expressions"). The original issue body sketched these in a hypothetical
// dialect; the tests below port them to the dialect Perspective actually
// implements. Differences from the issue's pseudocode:
//
// - `find(str, substr) -> int` does not exist. The closest function is
// `indexof(col, regex, out_vec) -> bool`, which performs a *regex* search,
// writes [start, end] of the first capturing group into `out_vec`, and
// requires the regex to have at least one capturing group (else it
// returns STATUS_CLEAR). The tests therefore wrap the literal char in a
// capturing group: `' '` -> `'( )'`, `','` -> `'(,)'`, `$` -> `'([$])'`.
// - `null()` is not a function call; `null` is a literal.
// - `strlen(s)` -> `length(s)`.
// - `substring(s, start, count)` takes a *count*, not an end index, and
// returns null if `start + count > length(s)`.
// - String literals pass through ExprTK's `cleanup_escapes`, which drops
// unrecognized escape characters (`\s` -> `s`, `\.` -> `.`).
((perspective) => {
test.describe("Issue 1527 use cases", function () {
test("contains literal substring", async function () {
const table = await perspective.table({
a: ["abcdef", "xyz", "abXabY", null, "abc"],
});
const view = await table.view({
expressions: {
has_ab: "contains(\"a\", 'ab')",
},
});
const result = await view.to_columns();
const schema = await view.expression_schema();
expect(schema["has_ab"]).toEqual("boolean");
expect(result["has_ab"]).toEqual([true, false, true, null, true]);
view.delete();
table.delete();
});
// Parse "USD $1000"-style strings into Currency (string) and Value
// (float) columns, tolerant of malformed rows.
test("split currency/value string column", async function () {
const table = await perspective.table({
"Bad Column": [
"USD $1000",
"EUR $250",
"malformed",
null,
"GBP $42",
],
});
const view = await table.view({
expressions: {
Currency: `var v[2];
if (indexof("Bad Column", '( )', v)) { substring("Bad Column", 0, v[0]) } else { null }`,
Value: `var v[2];
if (indexof("Bad Column", '([$])', v)) { float(substring("Bad Column", v[0] + 1)) } else { null }`,
},
});
const result = await view.to_columns();
const schema = await view.expression_schema();
expect(schema["Currency"]).toEqual("string");
expect(schema["Value"]).toEqual("float");
expect(result["Currency"]).toEqual([
"USD",
"EUR",
null,
null,
"GBP",
]);
expect(result["Value"]).toEqual([1000, 250, null, null, 42]);
view.delete();
table.delete();
});
// Parse "(123, 456)"-style strings into Longitude and Latitude
// float columns.
test("split longitude/latitude string column", async function () {
const table = await perspective.table({
Coords: [
"(123, 456)",
"(1.5, -2.25)",
"broken",
null,
"(0, 0)",
],
});
const view = await table.view({
expressions: {
Longitude: `var v[2];
if (indexof("Coords", '(,)', v)) { float(substring("Coords", 1, v[0] - 1)) } else { null }`,
Latitude: `var v[2];
if (indexof("Coords", '(,)', v)) { float(substring("Coords", v[0] + 1, length("Coords") - v[0] - 2)) } else { null }`,
},
});
const result = await view.to_columns();
const schema = await view.expression_schema();
expect(schema["Longitude"]).toEqual("float");
expect(schema["Latitude"]).toEqual("float");
expect(result["Longitude"]).toEqual([123, 1.5, null, null, 0]);
expect(result["Latitude"]).toEqual([456, -2.25, null, null, 0]);
view.delete();
table.delete();
});
// Normalize spelling variants by stripping dots and whitespace.
test("replace_all regex strips dots/whitespace", async function () {
const table = await perspective.table({
State: ["NC", "N.C.", "N. C.", "N .C.", "VA"],
});
const view = await table.view({
expressions: {
Normalized: `replace_all("State", '[. ]', '')`,
},
});
const result = await view.to_columns();
expect(result["Normalized"]).toEqual([
"NC",
"NC",
"NC",
"NC",
"VA",
]);
view.delete();
table.delete();
});
});
})(perspective);