Files
Lovell Fuller 7e9b350e22 Modernise all unit tests
- Ensure all tests have an expected assertion count
- Ensure all tests have at least one assertion
- Use async/await unless test is callback-specific
2026-05-20 21:57:40 +01:00

254 lines
5.7 KiB
JavaScript

/*!
Copyright 2013 Lovell Fuller and others.
SPDX-License-Identifier: Apache-2.0
*/
const { suite, test } = require('node:test');
const exifReader = require('exif-reader');
const sharp = require('../../');
const fixtures = require('../fixtures');
suite('Gain maps', () => {
test('Metadata contains gainMap', async (t) => {
t.plan(4);
const { format, gainMap } = await sharp(
fixtures.inputJpgWithGainMap,
).metadata();
t.assert.strictEqual(format, 'jpeg');
t.assert.strictEqual(typeof gainMap, 'object');
t.assert.ok(Buffer.isBuffer(gainMap.image));
t.assert.strictEqual(gainMap.image.length, 31738);
});
test('Can be regenerated', async (t) => {
t.plan(4);
const data = await sharp(fixtures.inputJpgWithGainMap)
.withGainMap()
.toBuffer();
const metadata = await sharp(data).metadata();
t.assert.strictEqual(metadata.format, 'jpeg');
t.assert.strictEqual(typeof metadata.gainMap, 'object');
t.assert.ok(Buffer.isBuffer(metadata.gainMap.image));
const {
format,
width,
height,
channels,
depth,
space,
hasProfile,
chromaSubsampling,
} = await sharp(metadata.gainMap.image).metadata();
t.assert.deepEqual(
{
format,
width,
height,
channels,
depth,
space,
hasProfile,
chromaSubsampling,
},
{
format: 'jpeg',
width: 1920,
height: 1080,
channels: 1,
depth: 'uchar',
space: 'b-w',
hasProfile: true,
chromaSubsampling: '4:4:4',
},
);
});
test('Buffer can be detached and reattached', async (t) => {
t.plan(4);
const data = await sharp(fixtures.inputJpgWithGainMap)
.keepGainMap()
.toBuffer();
const metadata = await sharp(data).metadata();
t.assert.strictEqual(metadata.format, 'jpeg');
t.assert.strictEqual(typeof metadata.gainMap, 'object');
t.assert.ok(Buffer.isBuffer(metadata.gainMap.image));
const {
format,
width,
height,
channels,
depth,
space,
hasProfile,
chromaSubsampling,
} = await sharp(metadata.gainMap.image).metadata();
t.assert.deepEqual(
{
format,
width,
height,
channels,
depth,
space,
hasProfile,
chromaSubsampling,
},
{
format: 'jpeg',
width: 960,
height: 540,
channels: 1,
depth: 'uchar',
space: 'b-w',
hasProfile: false,
chromaSubsampling: '4:4:4',
},
);
});
test('File can be detached and reattached', async (t) => {
t.plan(4);
const outputPath = fixtures.path('output-with-gain-map.jpg');
await sharp(fixtures.inputJpgWithGainMap)
.keepGainMap()
.toFile(outputPath);
const metadata = await sharp(outputPath).metadata();
t.assert.strictEqual(metadata.format, 'jpeg');
t.assert.strictEqual(typeof metadata.gainMap, 'object');
t.assert.ok(Buffer.isBuffer(metadata.gainMap.image));
const {
format,
width,
height,
channels,
depth,
space,
hasProfile,
chromaSubsampling,
} = await sharp(metadata.gainMap.image).metadata();
t.assert.deepEqual(
{
format,
width,
height,
channels,
depth,
space,
hasProfile,
chromaSubsampling,
},
{
format: 'jpeg',
width: 960,
height: 540,
channels: 1,
depth: 'uchar',
space: 'b-w',
hasProfile: false,
chromaSubsampling: '4:4:4',
},
);
});
test('Can be detached, resized and reattached', async (t) => {
t.plan(4);
const data = await sharp(fixtures.inputJpgWithGainMap)
.keepGainMap()
.resize(32)
.toBuffer();
const metadata = await sharp(data).metadata();
t.assert.strictEqual(metadata.format, 'jpeg');
t.assert.strictEqual(typeof metadata.gainMap, 'object');
t.assert.ok(Buffer.isBuffer(metadata.gainMap.image));
const {
format,
width,
height,
channels,
depth,
space,
hasProfile,
chromaSubsampling,
} = await sharp(metadata.gainMap.image).metadata();
t.assert.deepEqual(
{
format,
width,
height,
channels,
depth,
space,
hasProfile,
chromaSubsampling,
},
{
format: 'jpeg',
width: 8,
height: 5,
channels: 1,
depth: 'uchar',
space: 'b-w',
hasProfile: false,
chromaSubsampling: '4:4:4',
},
);
});
test('Cannot keep existing gain map with certain operations', async (t) => {
t.plan(2);
await t.assert.rejects(
sharp(fixtures.inputJpgWithGainMap)
.keepGainMap()
.modulate({ hue: 180 })
.toBuffer(),
/Modulate is not supported when keeping gain maps/
);
await t.assert.rejects(
sharp(fixtures.inputJpgWithGainMap)
.keepGainMap()
.convolve({
width: 3,
height: 3,
kernel: [0, -1, 0, -1, 5, -1, 0, -1, 0]
})
.toBuffer(),
/Convolve is not supported when keeping gain maps/
);
});
test('other metadata can be retained when keeping gain map', async (t) => {
t.plan(2);
const input = fixtures.inputJpgWithGainMap;
const data = await sharp(input)
.keepGainMap()
.keepMetadata()
.resize(32)
.toBuffer();
const { exif, gainMap } = await sharp(data).metadata();
const exifData = exifReader(exif);
t.assert.strictEqual(exifData.Photo.PixelXDimension, 32);
t.assert.ok(Buffer.isBuffer(gainMap.image));
});
});