pmat

English | 中文

🚀 A analysis tool for performance measurement

Install

npm install d -g pmat

# or use yarn:
# yarn global add pmat

The output

Notice: It loaded .... the period means This test takes time. So -n go up and the period go up. but after you measure TTI, try to keep count below 5, because it will take a lot of time.

PerformanceNavigationTiming

Key	Value
Duration time	duration
Redirect time	redirectEnd - redirectStart
App cache time	domainLookupStart - fetchStart
DNS lookup time	domainLookupEnd - domainLookupStart
TCP connect time	connectEnd - connectStart
Time To First Byte time	responseStart - requestStart
Download time of the page	responseEnd - responseStart
Fetch resource time	responseEnd - fetchStart
White screen time	domInteractive - fetchStart
DOM Ready time	domContentLoadedEventEnd - fetchStart
DOM Content Load time	domContentLoadedEventEnd - domContentLoadedEventStart

https://developer.mozilla.org/en-US/docs/Web/API/PerformanceNavigationTiming

Metric

Key	Value
FP	First Paint
FCP	First Content Paint
LCP	Largest Contentful Paint
CLS	Cumulative Layout Shift
FID	First Input Delay
TBT	Total Blocking Time
TTI	Time to Interactive

Usage

pmat --help

Usage: pmat [options] [url]

🚀 A analysis tool for performance measurement

Options:

   -v, --version                output the version number
   -n, --count <n>              specified loading times (default: 20)
   -u, --useragent <ua>         to set the useragent
   --no-cache                   disable cache (default: false)
   --no-javascript              disable javascript (default: false)
   --no-online                  disable network (defalut: false)
   -h, --help                   output usage information

For instance

 # We can omit the protocol header if has omitted, the protocol header will be `https://`

 # The simplest usage
 pmat taobao.com

 # if the url has any parameter, surround the url with double quotes
 pmat "taobao.com?a=1&b=2"

 #  Load the specified page 100 times
 pmat -n 5 "taobao.com?a=1&b=2"

 #  Load the specified page 100 times without `cache`
 pmat -n 5 "taobao.com?a=1&b=2" --no-cache

 #  Load the specified page 100 times without `javascript`
 pmat -n 5 "taobao.com?a=1&b=2" --no-javascript

 #  Load the specified page 100 times with `headless = false`
 pmat -n 5 "taobao.com?a=1&b=2" -H false

 #  Load the specified page 100 times with set `useragent`
 pmat -n 5 "baidu.com?a=1&b=2" -u "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.181 Safari/537.36"

Pain point

After we have developed a project or optimized the performance of a project,

how do we measure the performance of this project?

A common approach is to look at the data in the performance and network in the Dev Tool, record a few key performance metrics, and refresh them a few times before looking at those performance metrics,

Sometimes we find that due to the small sample size, the current Network/CPU/Memory load is heavily impacted, and sometimes the optimized project is slower than before the optimization.

If there is a tool, request web page many times, and then taking out the various performance indicators averaging, we can very accurately know the optimization is positive or negative.

In addition, you can also make a comparison and get accurate data about how much you have optimized. This tool is designed to solve the pain point.

At the same time, this tool is also a good tool for us to learn about the "browser's process of load and rendering" and "performance optimization", so that we don't get wrong conclusions when there are too few samples

Contributing

1. Fork it
2. Create your feature branch (git checkout -b my-new-feature)
3. Commit your changes (git commit -am 'Add some feature')
4. Push to the branch (git push origin my-new-feature)
5. Create new Pull Request

License

MIT

Welcome Star and PR

Copyright (c) 2018 [e-mail]