Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/ringabout/timeit

measuring execution times written by Nim.
https://github.com/ringabout/timeit

bench nim nimble timeit

Last synced: about 1 month ago
JSON representation

measuring execution times written by Nim.

Awesome Lists containing this project

README 

        
# timeit [![nimble](https://raw.githubusercontent.com/yglukhov/nimble-tag/master/nimble.png)](https://github.com/yglukhov/nimble-tag)

measuring execution times written in nim.



## Installation

```text

nimble install timeit

```

## Documention

https://xflywind.github.io/timeit/src/main.html



## Usage

### import timeit

In **timeit** module, you can use the **timeGo** to

measure execution times of proc. \

There are three parameters in **timeGo**.\

**myFunc** is the call of proc. **repeatTimes** is

how many times you want to measure, the result is the average of all the times, **loopTimes** is how many loops you want to execute in each time.

If you don't give the value of **loopTimes**, the program will calculate the **loopTimes** according to the execution times in first time.    

```nim

template timeGo*(myFunc: untyped, 

                repeatTimes: int = repeatTimes, 

                loopTimes: int = loopTimes): Timer

```

When you want to pass **myFunc** parameter to **timeGo**, there are some points you should take care of. \

If you want to pass the proc without return value, you should use the call of the proc, namely **myProc(args)**. \

If you want to pass the proc with return value,

you should add the pragma **{.discardable.}** to the

definitions of your proc.Then you can use **timeGo**

to measure the execution times of your proc, namely **myProc(args)**.



### another Way

You can also use timeGo as follow:

```nim

import os



echo timeGo do:

  os.sleep(12)

  var a = 12

  for i in 1 .. 1000:

  a += 12

# output [12ms 883μs 34.58ns] ± [19μs 974.83ns] per loop (mean ± std. dev. of 7 runs, 10 loops each)

```

or you can specify the repeatTimes and loopTimes as follows:

```nim

timeGo(1, 1):

  var b = 12

  b += 12

# output [200.00ns] ± [0.00ns] per loop (mean ± std. dev. of 1 runs, 1 loops each)

```

### use monit function

You can use monit function to measure the times

of code executions. \

First you can specify the **name** of this test,

for example, "first".Then you can place the

**start** function in the begin of the code you want

to measure, and place the **finsih** function in the

end of the code you want to measure.

```nim

import timeit

var m = monit("first")

m.start()

let a = 12

echo a + 3 

m.finish()

# first -> [17μs 0.00ns]

```

You can also monit once as follows:

```nim

timeOnce("test-once"):

  var a = 12

  for i in 1 .. 10000:

    a += i

  echo a

```



### use command-line interface

Firstly, you need to make sure that your **.nimble** directory must be in your path environment.

Then you can use **timeit --name=yourNimFile --def=yourProc**. \

However, string parameters can't appear in yourProc.And If you want

to specify more than one parameter, you should use **"yourProc(parameters)"**.



```nim

# in bash

timeit --name=test --def=hello()

# [46ms 290μs 893.17ns] ± [1ms 164μs 333.97ns] per loop (mean ± std. dev. of 7 runs, 10 loops each)

```

You can also measure the execution time of the whole Nim file.

```nim

# in bash

timeit test.nim

# debug mode

# test-whole -> [208μs 100.00ns]



# or specify the -d flag

timeit test.nim -d:release

# release mode

# test-whole -> [29μs 200.00ns]

```



## Examples

for proc without return value

```nim

import timeit



proc mySleep(num: varargs[int]) = 

  for i in 1 .. 10:

    discard

echo timeGo(mySleep(5, 2, 1))

# [216.50ns] ± [1μs 354.44ns] per loop (mean ± std. dev. of 7 runs, 1000000 loops each)

```

for proc with return value

```nim

import timeit



proc mySleep(num: varargs[int]): int {.discardable.} = 

  for i in 1 .. 10:

    discard

echo timeGo(mySleep(5, 2, 1)) 

# [221.82ns] ± [1μs 837.93ns] per loop (mean ± std. dev. of 7 runs, 1000000 loops each) 

```