Improve performance #16
Comments
|
Yes, you're probably better off using https://github.com/sindresorhus/ansi-styles for high-performance situations. It will probably be faster than Though I'll keep this open to remind me about doing a perf review of Chalk. There might be some optimizations to be made. |
|
Thanks for the recommendation. I'll take a look. |
sindresorhus
changed the title
|
@CrabDude I made an alternate implementation that might be faster. It should be a drop-in replacement for https://github.com/aceface/crayon Can you try running it on your codebase and see if things run faster? If you don't a ton of chaining, it should be significantly faster (turning 100000 simple strings red was 31ms vs 7s with chalk for me) @sindresorhus take a look at the code. In general, I think the |
|
Definitely. I'll try it out this weekend and report back. On Friday, May 23, 2014, Charlie Cheever notifications@github.com wrote:
|
- Precomputed style function
- Skip arguments to array + join if there's only one argument (the
common case)
- Merge multiple return statements to one
To calculate the performance benefit:
```javascript
var chalk = require('./index.js');
console.time('100000 iterations');
for (var i = 0; i < 100000; i++) {
chalk.red('A string that is about 80 characters long (normal use I think?)');
}
console.timeEnd('100000 iterations');
```
Running this before this commit:
```shell
for i in {1..5}; do node time.js; done
100000 iterations: 19485ms
100000 iterations: 18933ms
100000 iterations: 19365ms
100000 iterations: 19332ms
100000 iterations: 18660ms
```
After:
```shell
100000 iterations: 268ms
100000 iterations: 261ms
100000 iterations: 264ms
100000 iterations: 259ms
100000 iterations: 254ms
```
Performance gain, taking the middle result of both:
```shell
19332 / 261 = 74.~
```
Closes chalk#16
- Precomputed style function
- Skip arguments to array + join if there's only one argument (the
common case)
- Merge multiple return statements to one
To calculate the performance benefit:
```javascript
var chalk = require('./index.js');
console.time('100000 iterations');
for (var i = 0; i < 100000; i++) {
chalk.red('A string that is about 80 characters long (normal use I think?)');
}
console.timeEnd('100000 iterations');
```
Running this before this commit:
```shell
for i in {1..5}; do node time.js; done
100000 iterations: 19485ms
100000 iterations: 18933ms
100000 iterations: 19365ms
100000 iterations: 19332ms
100000 iterations: 18660ms
```
After:
```shell
100000 iterations: 268ms
100000 iterations: 261ms
100000 iterations: 264ms
100000 iterations: 259ms
100000 iterations: 254ms
```
Performance gain, taking the middle result of both:
```shell
19332 / 261 = 74.~
```
Closes chalk#16
- Precomputed style function
- Skip arguments to array + join if there's only one argument (the
common case)
- Merge multiple return statements to one
To calculate the performance benefit:
```javascript
var chalk = require('./index.js');
console.time('100000 iterations');
for (var i = 0; i < 100000; i++) {
chalk.red('A string that is about 80 characters long (normal use I think?)');
}
console.timeEnd('100000 iterations');
```
Running this before this commit:
```shell
for i in {1..5}; do node time.js; done
100000 iterations: 19485ms
100000 iterations: 18933ms
100000 iterations: 19365ms
100000 iterations: 19332ms
100000 iterations: 18660ms
```
After:
```shell
100000 iterations: 268ms
100000 iterations: 261ms
100000 iterations: 264ms
100000 iterations: 259ms
100000 iterations: 254ms
```
Performance gain, taking the middle result of both:
```shell
19332 / 261 = 74.~
```
Closes chalk#16
- Precomputed style function
- Skip arguments to array + join if there's only one argument (the
common case)
- Merge multiple return statements to one
To calculate the performance benefit:
```javascript
var chalk = require('./index.js');
console.time('100000 iterations');
for (var i = 0; i < 100000; i++) {
chalk.red('A string that is about 80 characters long (normal use I think?)');
}
console.timeEnd('100000 iterations');
```
Running this before this commit:
```shell
for i in {1..5}; do node time.js; done
100000 iterations: 19485ms
100000 iterations: 18933ms
100000 iterations: 19365ms
100000 iterations: 19332ms
100000 iterations: 18660ms
```
After:
```shell
100000 iterations: 268ms
100000 iterations: 261ms
100000 iterations: 264ms
100000 iterations: 259ms
100000 iterations: 254ms
```
Performance gain, taking the middle result of both:
```shell
19332 / 261 = 74.~
```
Closes chalk#16
- Precomputed style function
- Skip arguments to array + join if there's only one argument (the
common case)
- Merge multiple return statements to one
To calculate the performance benefit:
```javascript
var chalk = require('./index.js');
console.time('100000 iterations');
for (var i = 0; i < 100000; i++) {
chalk.red('A string that is about 80 characters long (normal use I think?)');
}
console.timeEnd('100000 iterations');
```
Running this before this commit:
```shell
for i in {1..5}; do node time.js; done
100000 iterations: 19485ms
100000 iterations: 18933ms
100000 iterations: 19365ms
100000 iterations: 19332ms
100000 iterations: 18660ms
```
After:
```shell
100000 iterations: 268ms
100000 iterations: 261ms
100000 iterations: 264ms
100000 iterations: 259ms
100000 iterations: 254ms
```
Performance gain, taking the middle result of both:
```shell
19332 / 261 = 74.~
```
Closes chalk#16
I don't have specific other than that I was upgrading my server's trycatch version and it was very slow. Over a couple hours, I tracked the issue down to trycatch's coloring and then to the chalk module itself.
Removing just the line that adds coloring took my 3k unit tests from 30s down to 10s. The colors module doesn't have this problem. Not sure what the difference is. Looked through the source, but it looked like every time a color is accessed you're doing a dynamic getter?
The text was updated successfully, but these errors were encountered: