GuideAutomator

Automated User Guide Generation with Markdown

GuideAutomator allows developers and technical writers to create user manuals with screenshots that are always in sync with the latest version of the application being documented. For that end, it takes as input a Markdown file containing the user manual text with short JavaScript chunks that determine how to capture each screencast.

Sample GuideAutomator input:

## How to find GuideAutomator

Open the Bing website and type `GuideAutomator aside-ufba`:

```js
get('https://bing.com/');
fillIn('#sb_form_q', 'GuideAutomator aside-ufba');
takeScreenshotOf('#sb_form>div', true);
```

Press *Enter*. The first result should be what you are looking for:

```js
submit('#sb_form');
takeScreenshotOf('#b_results>li:nth-child(1)', true);
```

Corresponding output:

GuideAutomator was proposed by professor Rodrigo Souza and originally implemented by Allan Oliveira as part of his Bachelor's thesis. The project is now being maintained by Welbert Serra. The project is being developed in the context of aSide, a research group at the Federal University of Bahia specialized in software design and evolution.

Related publications:

• GuideAutomator: Continuous Delivery of End User Documentation -- paper accepted at ICSE 2017, NIER Track (New Ideas and Emerging Results)
• GuideAutomator: Automated User Manual Generation with Markdown -- Allan Oliveira's bachelor thesis
• GuideAutomator: Automated User Manual Generation with Markdown -- paper presented at CBSoft 2016 (Tools Track)

Summary

• Installation
  • Prerequisites
  • Linux
  • Windows
• Getting started
  • Description
  • Example
  • Css selector
  • Firefox extension
• Usage
• Api commands
• Using GD.driver
• Author
  • Contributors

Installation

Prerequisites

• Nodejs && npm (Compiler)
• ImageMagick (Needed for cropping images)
• Wkhtmltopdf (Needed for converting html to pdf)
• Chrome and Chrome WebDriver (Selenium Web Browser automation)

back to top

Linux

• (Option 1)You can install with script

  sudo wget -qO- https://raw.githubusercontent.com/aside-ufba/guide-automator/master/install_linux.sh | bash -

• (Option 2) Manual installation. You need install some binary dependencies.

  1.Node and npm:

    https://nodejs.org/en/download/package-manager/#debian-and-ubuntu-based-linux-distributions

  2.ImageMagick (Usually is already installed):

    https://www.imagemagick.org/script/binary-releases.php#unix

  3.Wkhtmltopdf

    http://wkhtmltopdf.org/downloads.html

  4.Chrome/Chromium Browser

  -Chrome:
    https://www.google.com/chrome/browser/desktop/index.html
  -Chromium:
    sudo apt-get install chromium-browser

  5.Chrome WebDriver

    https://chromedriver.storage.googleapis.com/index.html?path=2.25/

  6.Guide-Automator

    sudo npm install -g guide-automator

Node, wkhtmltopdf and Chrome WebDriver need to be add in path.

back to top

Windows

1.Node and npm:

  https://nodejs.org/en/download/

2.ImageMagick: (Check to install legacy utilities)

  https://www.imagemagick.org/script/binary-releases.php#windows

3.Wkhtmltopdf

  http://wkhtmltopdf.org/downloads.html

4.Chrome

-Chrome:
  https://www.google.com/chrome/browser/desktop/index.html

5.Chrome WebDriver

  https://chromedriver.storage.googleapis.com/index.html?path=2.25/

6.Guide-Automator

  npm install -g guide-automator

Node, ImageMagick(convert.exe), Wkhtmltopdf and Chrome WebDriver need to be add in path.

back to top

Getting started

Description

Guide-automator extract javascript or js tags (```javascript guide-automator commands ``` or ```js guide-automator commands ```) from markdown file and generate manual from them. You need use our API commands in markdown file.

back to top

Example:

# This is my github

```js
  get('https://github.com/welbert');
  takeScreenshot();
  takeScreenshotOf('.avatar',false,true);

```(<- three back-ticks)

More examples with comments here or see raw, if you want to look the output pdf (This pdf was created using command '$ guide-automator -i examples/example.md -o output/ -s lightBlue')

back to top

Css selector

You will need know how to extract css selector of Web elements, you can use the browser to do that. If you dont know how to do this, see below.

• Google Chrome

• Mozilla Firefox

After this steps, the css selector will be copied to your clipboard

back to top

Firefox extension

You can use the firefox addon to make easy the construction of blocks

Guide-automator-addon

back to top

USAGE

$ guide-automator -h

  Usage: guide-automator [options]

  Options:


    -h, --help             Output usage information
    -V, --version          Output the version number
    -i, --input <File.md>  Input .md file
    -o, --output <Folder>  Output destination folder
    -P, --pdf              Export manual to PDF, default is export for all types
    -H, --html             Export manual to HTML, default is export for all types
    -I, --image            Export ONLY manual's image and ignore others types, default is export for all types
    -s, --style <style.css>  Css style to be used in the manual or theme [lightBlue,lightOrange]
    -t, --autosleep <Millisecond>  Time to sleep before screenshot
    -d, --debug            Show progress of code
    -b, --browser <path>   Use Chromium browser at given path
    -l, --headless         Use headless Chrome (does not require a GUI)
    -w, --window <dimensions>  Set browser window's dimensions (e.g., 800x600)

  Examples:

    $ guide-automator -i input.md -o output/
    $ guide-automator -i input.md
    $ guide-automator -i input.md -o output/ -s lightBlue

back to top

API commands

Params: <Required> [Optional]

• get
• click
• clickByLinkText
• highlight
• takeScreenshot
• takeScreenshotOf
• fillIn
• submit
• wait
• sleep
• console.print
• pageContext

get

• Description: Url to access
• Params: <url>
• Example:
  • get('https://github.com/welbert');

click

• Description: Click on element
• Params: <cssSelector>
• Example:
  • click('.unstarred > button:nth-child(2)');

clickByLinkText

• Description: Click on link text(<a>)
• Params: <linkText>
• Example:
  • clickByLinkText('README.md');

highlight

• Description: Highlights an HTML element by adding a red outline
• Params: <cssSelector>
• Example:
  • highlight('.header-search-input');
  • highlight(['.header-search-input', '#frameID']);

takeScreenshot

• Description: Take Screenshot
• Params: [imageWidth]
• Example:
  • takeScreenshot();
  • takeScreenshot('10%');

takeScreenshotOf

• Description: Take Screenshot of specific element
• Params: <cssSelector>, [crop], [outline], [imageWidth]
• Params: Array(<cssSelector>,[cssSelector],...), [crop], [outline], [imageWidth]
• Params: Array(Array(<cssSelector>,<Iframe-CssSelector>),[cssSelector],...), [crop], [outline], [imageWidth]
• Example:
  • takeScreenshotOf('#user-content-guideautomator');
  • takeScreenshotOf('#user-content-guideautomator',false,true);
  • takeScreenshotOf(['#user-content-guideautomator','.avatar'],false,true); //Multiple cssSelector(only for outline)
  • takeScreenshotOf([['#user-content-guideautomator','#frameID'],'.avatar'],false,true) //Multiple cssSelector (only for outline) on different contexts(Iframe)

fillIn

• Description: Fill in field element
• Params: <cssSelector>, <input>
• Example:
  • fillIn('.header-search-input','guide-automator');

submit

• Description: Submit a form
• Params: <cssSelector>
• Example:
  • submit('.js-site-search-form');

wait

• Description: Wait element load on page
• Params: <cssSelector>, [timeOut]
• Example:
  • wait('.commit-author-section');
  • wait('.commit-author-section',2000);

sleep

• Description: Sleep for a while
• Params: <milleseconds>
• Example:
  • sleep(1000);

console.print

• Description: Print text on markdown file using javascript blocks
• Params: <text>
• Example:
  • console.print('# Title');

pageContext

• Description: Need to change context to iframe or context default(Main page)
• Params: [cssSelector]
• Example:
  • pageContext('#Iframe-CssSelector');
  • pageContext('default'); //Equivalent to pageContext();
  • pageContext();

back to top

Using GD.driver

This topic is for more complex cases or personal needs that guide-automator does not support

Example:

GD.driver.actions().
  mouseMove({x:216, y:41}).
  click().
  perform();

See raw

Selenium Documentation

back to top