Data

Tools

Indexes

Applications

Experiments

Awesome

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

https://github.com/movableink/cropduster

Support tools for Movable Ink Web Crops
https://github.com/movableink/cropduster

Last synced: 6 months ago
JSON representation

Support tools for Movable Ink Web Crops

Awesome Lists containing this project

README 

        
# Cropduster



Cropduster is a collection of tools for building pages to be web cropped by

Movable Ink.



[![Build Status](https://travis-ci.org/movableink/cropduster.svg?branch=master)](https://travis-ci.org/movableink/cropduster)



## Installation



Cropduster is on npm. Install it with:



```bash

npm install --save-dev cropduster

```



The use it by importing it from your es6-enabled javascript file:



```javascript

import CD from 'cropduster';

```



## Legacy Installation



To use Cropduster directly in the browser, loaded from a script tag, you can

find legacy versions at:



http://projects.movableink.com/production/libs/cropduster.[version].js



Where you replace `[version]` with the version you want to use.



Browser-only cropduster versions are no longer being published for new versions.



## API



### Asynchronous actions

Sometimes, Capturama needs to be explicitly told to hold off on finishing its

screen capture in order for some asynchronous actions to fire and complete. This

can be accomplished by calling `CD.pause` when an asynchronous action is about

to start, and `CD.resume` when it has completed. `pause` takes a `maxSuspension`

Number argument specifying the maximum amount of time in milliseconds that your

asynchronous action is allowed to take, and both functions take an optional

final argument of a String message explaining the suspension, purely used for

debugging the Capturama log.



If you are using the Cropduster API for common asynchronous actions like

fetching a URL or an image with `CD.get` and `CD.getImage`, or if you are using

Promises directly, you do not need to manually call pause and resume. These

methods are used internally in the necessary places. However, if you are

doing something unusual that requires work to be delayed manually, each call to

`CD.pause` must have a corresponding call to `CD.resume`, or the request to

Capturama will eventually time out.



Example:

```javascript

const target = document.getElementById('text-box');

const customerQuality = CD.param('mi_customer_rating');

const tenSeconds = 10 * 1000;



if (customerQuality === 'very-good') {

  target.innerText = 'good customers get images quickly';

} else if (customerQuality === 'very-bad') {

  CD.pause(tenSeconds, 'making bad customers wait for their email to load...');



  setTimeout(() => {

    CD.resume();

    target.innerText = 'bad customers have to wait for their images';

  }, 1000);

}

```



It is generally recommended to call CD.pause as the last thing before starting

an asynchronous action, and CD.resume as the very first thing once that action

has finished. If your asynchronous action completes successfully, but your

callback runs some code before calling CD.resume again, then you run the risk of

triggering a JavaScript error that stops execution of the current script. If

CD.resume is not called after that first CD.pause, your image will eventually

time out, as opposed to failing immediately.



*NOTE:* Cropduster previously offered `CD.suspend` and `CD.capture` functions

that achieved a similar goal. These functions have been replaced with `pause`

and `resume`, to support a better synchronisation of state with Capturama, and

to give a clearer sense of how these functions affect Capturama's workflow.



### Selecting elements



`CD.$` is useful for getting an array of DOM elements via a CSS selector. It always returns an array.



Example:



```javascript

const elements = CD.$('div.items');

for (let i = 0; i < elements.length; i++) {

  const element = elements[i];

  element.style.display = 'none';

}

```



or:



```javascript

const elements = CD.$('div.items');

for (const element of elements) {

  element.style.display = 'none';

}

```



### Query params



`CD.param` retrieves query parameters from the current document's URL. It does not support nested params or arrays of values.



Example:



```javascript

// document.location is 'http://example.com/?fname=john

const fname = CD.param('fname');

console.log(fname); // logs 'john'

```



### Fetching third-party resources



Movable Ink's capture engine traditionally captures the web page as soon as the page's `ready` event fires. This can cause

issues when the user tries to fetch pages via ajax, as the capture engine does not wait for the ajax load to complete

before rendering. In order to delay capture until the request has finished, use `CD.get`. It temporarily suspends capture

until the request completes. Note: the URL has to be CORS-accessible, see `CD.getCORS` if it isn't. When in doubt, use `CD.getCORS`.



Example:



```javascript

CD.get('http://cors-enabled-site.com/page').then((response) => {

  const { data, status, contentType } = response;

  CD.$('h1')[0].innerHTML = data.header;

});

```



Send POST and sending extra headers:



```javascript

CD.get('http://cors-enabled-site.com/page', {

  method: 'POST',

  body: '{"ok": "yes"}',

  headers: {

    'Accept': 'application/json'

  }

}).then((response) => {

  CD.$('h1')[0].innerHTML = response.data.h1;

})

```



### Fetching third-party resources with CORS



Security restrictions prevent web pages from making cross-domain ajax requests. CORS

is a workaround, but requires support from the website, and many websites do not

support it. Instead, use `CD.getCORS` to use Movable Ink's CORS proxy to access the

page.



Example:



```javascript

CD.getCORS('http://example.com/page').then((response) => {

  CD.$('h1')[0].innerHTML = response.data.header;

});

```



### Fetching images



_Deprecated:_ The `CD.getImage()` and `CD.getImages()` calls have been superseded by `CD.waitForAsset()`. They will continue to work for the foreseeable future, but `CD.waitForAsset` is preferred.



`CD.waitForAsset` ensures that in-flight requests for a particular asset URL complete before capturing the page. `CD.waitForAsset` does not fetch the image, that is up to you to do however you may normally do it. If the image is not loading ("in-flight") when capturama tries to capture the page, there will be no impact. It can be called any time before page capture. `CD.waitForAsset` is compatible with `` tags, css `background-image` properties, and any other way an image is loaded into a webpage.



Example:



```javascript

CD.$('.background')[0].style.backgroundImage = 'url(http://example.com/foo.png)';

CD.waitForAsset('http://example.com/foo.png');

```



### Redirecting to another image



Sometimes, it is necessary to redirect to an image rather than rendering dynamic content. For example, a countdown timer may

want to just show an image after the countdown has expired. It is certainly possible to just render the image inside the

webpage and crop that, but a better solution is to use `CD.setImageRedirect` to issue a 302 redirect to the user to send them

to the static content. If the function is called multiple times, the last image URL called is used.



Example:



```javascript

CD.setImageRedirect('http://example.com/foo.png');

console.log('user will be shown image located at http://example.com/foo.png');

```



### Storing extra analytics data



It is possible to store extra data using the `CD.setExtraData` call. This data will be available in the `extra_data` field of

the User-level Reports. Pass a javascript object, and it will be turned into JSON and stored in `extra_data`. Calling multiple

times results in the objects being combined and any duplicate keys overwritten.



Example:



```javascript

CD.setExtraData({userId: 5});

CD.setExtraData({shownCategory: 'shoes'});

console.log('extra_data field now contains {"userId":5,"shownCategory":"shoes"}');

```



### Setting custom per-user clickthrough URLs



When showing dynamic content, it may be useful to be able to associate dynamic content with matching clickthroughs. Using

`CD.setClickthrough`, you can save a clickthrough URL that users will visit when they click on the web crop. In order to work

properly, the query params on both the embed code's image and link must match. If called multiple times, only the last called

clickthrough URL will be used.



Example:



```javascript

CD.setClickthrough('http://example.com');

console.log('If user clicks on the web crop, they will go to http://example.com');

```



## Testing



    yarn install

    yarn run test



## Publishing



 1. Open a pull request with your change; add a new entry to the changelog called "Latest" and list out your changes.

 2. Wait for your pull request to get approved, then merge it into master.

 3. Run the release command for the type of change, following the prompts (this will tag and push to github and release to npm):

   * If you do not have access to publish to npm, ask @mnutt.

   * If your change introduces any breaking changes whatsoever, it needs a major version bump. These should definitely be listed in the changelog.

    - `yarn run release major`

   * If your change introduces new features but does not break any existing workflows, it should bump the minor version.

    - `yarn run release minor`

   * If your change is fixes bugs or updates tests or something, you can just bump the patch version.

    - `yarn run release` or `yarn run release patch`



## Changelog



### 7.0.0



 * Skip proxying requests for internal domains that support CORS



### 6.0.0

 * Remove use of `fetch` (major bump as exposure to `response` object is a breaking change)



### 5.5.0

 * Support non `utf-8` response encoding for `CD.get`



### 5.4.0

 * Expose `response` in resolved CD.get Promise



### 5.3.0

 * Bundling change: our package.json `main` points to webpacked version, `module` points to es6 version. Prefer es6 version unless you know you need the webpacked version.



### 5.2.0

 * CD.get uses the `fetch` API instead of XMLHttpRequest



### 5.1.0

  * Add `withoutCredentials` to `CD.get()` options to disable sending `withCredentials` in requests.



### 5.0.0

  * Compiles the app with Webpack and Babel down to ES5 with sourcemaps.

  * Returns Promises from CD.get, CD.getCORS, CD.getImage and CD.getImages

  * NPM publishes the browser-ready script in dist/cropduster.browser.js



### 4.1.0

  * Add `withCredentials: true` to all `CD.get()` requests.



### 4.0.1

  * npm publishes the alternate es5 scripts in dist/cropduster.es5.js.



### 4.0.0

  * Switch from bower to npm; CD is now exported as es6 module. See the new installation instructions above.



### 3.4.2

  * Remove `bower` from npm postinstall, for npm usage.



### 3.4.1

  * Adds `CD.params()`, returning an object with all parameters.



### 3.4.0

  * Exposes the `content-type` response header within the callback of `CD.get()`.



### 3.3.0

  * Add support for `CD.waitForAsset()`. This API can be used to ensure that we will not render while the passed URL is in-flight.



### 3.2.3

  * fix bower/npm version numbers



### 3.2.2

  * Ensure that `CD.suspend(msg)` always logs the `msg` string



### 3.2.1

  * `CD.get()` and `CD.getCORS()` return the (integer) http status as the second argument of their callbacks. This is currently completely backwards-compatible, but we would like to make backwards-incompatible changes to this API in the future.



### 3.2.0

  * Support `CD.cancelRequest(msg)` for when crop wants to show fallback but not trigger error condition

  * Support `CD.throwError(msg)` for when crop fails (such as XHR failure) and trigger error

  * `CD.begin(msg)` and `CD.end(msg)` support an argument that will be logged for easier debugging



### 3.1.1

  * Support for `CD.getCORS()` with non-standard ports



### 3.1.0

  * `CD.getCORS()` sends `x-mi-cbe` request header to ensure requests consistently go to the same backend



### 3.0.0

  * No more reference counting or maxSuspensions

  * `CD.getImages()` accepts two callbacks, one for when all images resolve and once for after each image resolves



### 2.7.3

  * `CD.capture` forces a redraw on the page



### 2.7.2

  * Ensure that `CD.getImages` actually suspends crop



### 2.7.1

  * Remove window.onerror handler

  * Ensure that failing requests actually call callback with `null`, rather than empty string



### 2.7.0

  * Failing requests call the callback with `null`.

  * Do not pass cors ttl headers on regular `get` request.



### 2.6.0

  * Add `get` to fetch resources without using CORS



### 2.5.0

  * Add `setImageRedirect`, `setClickthrough`, and `setExtraData`

  * Set up TravisCI test suite



### 2.4.0

  * New options for CD.getCORS: `method` for changing HTTP method, `body` for sending request body when `POST` method is used,

    and `headers` object for sending extra request headers.



### 2.3.0

  * Guarantee correct ordering of callbacks in getImages()



## License



Copyright © 2015 Movable, Inc



Permission is hereby granted, free of charge, to any person obtaining a copy of

this software and associated documentation files (the "Software"), to deal in

the Software without restriction, including without limitation the rights to

use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies

of the Software, and to permit persons to whom the Software is furnished to do

so, subject to the following conditions:



The above copyright notice and this permission notice shall be included in all

copies or substantial portions of the Software.



THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR

IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,

FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE

AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER

LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,

OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE

SOFTWARE.