Giters

nuxeo-sandbox / nuxeo-website-preview

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

nuxeo-website-preview

This plug-in allows to display a website stored in Nuxeo either as:

  • A Folderish Document containing the site, with relative paths. So for example, an index.html file with an img tag whose src attribute is "img/logo.png", and at first level of the Document, there is an img folder containing the logo.png Picture document

  • Or a non Folderish document whose file:content field contains a blob that can be unzipped as a website (typically a .zip file). Please read below, "WARNING About the Zip Format"

  • Notice: It will also display the minisite, if any, when the document is a Folderish whose content does not have any html file to display at first level and if this Folderish also have zip file in it's file schema.

  • Also, this is about previewing The preview is displayed in the context of the Nuxeo application. This means, some JavaScript or access to other website, if any, may fail (typically because of CORS)

Important

Please read all the WARNINGS below:

  • Security Warning
  • About the Note document type
  • About the Zip Format
  • Relative Paths and Max Length of name

How to Use

The plug-in creates a WebEngine module allowing to access the embedded website via a URL. The name of the module (in the URLs) is WSP (WebSitePreview).

  • In order to be as easy possible to use, the main url must end with index.html, whatever the real name of your main html file.

  • So, to access the preview(*), the URL to use is: {server:port}/nuxeo/site/WSP/main-parent-doc-id/index.html

    This URL will typically be used in the UI using an iframe

  • For example, say you have a Folderish document, named "My Site", whose id is 1234-5678-9ABC-DEF0, and you are testing on your localhost, you can display the preview using this URL: http://localhost:8080/nuxeo/site/WSP/1234-5678-9ABC-DEF0/index.html.

    The exact same URL is used if instead of a Folderish containing all the website as Nuxeo documents, it is a File whose file:content holds a zip containing the website

  • The plugin contributes the Website facet:

    • Because not every Folderish and not every .zip host a website
    • So you can dynamically add/remove (using the Document.AddFacet and Document.RemoveFacet operations) this facet from your documents for quick test to display a "preview" button in the UI for example
    • IMPORTANT: The plugin does not add/remove this facet at anytime, it is for you to use.
      • A typical example would be a button in the UI, like "This is a website". The user clicks it and you run the Document.Addfacet operation to add the Website facet.
      • This could also be done automatically in a listener depending on some metadata and rules, and if the Document.HasWebsite operation returns true (see below).

(*) Assuming current user is logged in and has enough rights to at least read the blobs.

The "Main" HTML File

Whatever the source (a Folderish or a .zip), the plugin searches for a main html file using this algorithm:

  • Whatever the name, it must be at first level, the plugin does not search in nested folders
  • If, at first level, one file is index.html, the document is considered being a website and this file will be returned at the first preview
  • If there is no index.html file, the plugin searches for any other .html and returns the first it finds, to be used at first display
  • If there is no .html file at all at first level, the plug-in returns a 404 error.

WARNING

This means that if the source document contains at least one html document at first level, will be seen as a mini website. The preview will display the HTML, but the result is unpredictable if it is not website with relative paths for sub-elements.

This is why you should use the Website facet and add your preview only when it is relevant

Utilities - Operation(s)

The plugin provides the following operation(s):

####    Document.HasWebsite

  • input is a document
  • output is the document, unchanged
  • The operation sets the WSP_hasWebsite boolean Context Variable with the result:
  • If the Document has the Website facet, WSP_hasWebsite is set to true
  • Else, if it is a Folderish document with an html child at first level, WSP_hasWebsite is set to true
  • Else, it the document has the file schema and file:content can be unzipped and contains an html file at first level, WSP_hasWebsite is set to true
  • Else, WSP_hasWebsite is set to false.

Usage in WebUI

  • First, create an element (in Studio Designer > Resources) with an iframe and set the src of the iframe to the correct url. For example:
<dom-module id="my-website-preview">
  <template>
    <style>
      *[role=widget] {
        padding: 5px;
      }

      iframe {
        position: absolute;
        width: 95%;
        height: 90%;
        border: 2 2 2 2;
      }

    </style>
    <!--
    It is important to filter with the facet: If this element is displayed as a tab, calls to the server
    will always be done for evey document _before_ applying WebUI filter. So, you display Root, Domain, ...
    => The plugin is called and tries to get an index.html file.
    This is not efficient => you should show/hide the iframe here to avoid these useless calls
    -->
    <iframe src="/nuxeo/site/WSP/[[document.uid]]/index.html"></iframe>

  </template>

  <script>
    Polymer({
      is: 'my-website-preview',
      behaviors: [Nuxeo.LayoutBehavior],
      properties: {
        
        document: {
          type: Object
        }
      }

    });
  </script>
</dom-module>
  • Create a Document Page tab (In Studio Designer > UI) displaying this element, with the correct filter (typically, if the document has the Website facet)

WARNING About the Zip Format

When using a zip file, it must not be built with a tool that prefixes the paths of every item. For example, if you zip the mysite folder containing...

img
  logo.png
index.html
otherpage.html

... the zip file must not have a TOC prefixing every path with mysite. The list of files in the zip must be...

img/
img/logo.png
index.html
otherpage.html

...and not:

mysite/img/
mysite/img/logo.png
mysite/index.html
mysite/otherpage.html

About the Note document type

When using the plugin to display the ocntent of a Folderish, we strongly recommand to make sure Nuxeo dose not create a Note document type for your HTML. Nuxeo may automatically strip some HTML content form the HTML, and the preview will display poorly.

So, we recommend to add the following XML extension in your Studio project, to disable the creation of Note when you drag-drop and HTML file. It will instead create a File, which is good.

<extension target="org.nuxeo.ecm.platform.filemanager.service.FileManagerService"
           point="plugins">

  <plugin name="NoteImporter" enabled="false"/>

</extension>

Relative Paths and Max Length of name

This warning applies only when previewing a Folderishand its content.

In this context, make sure the relative path is valid. Nuxeo truncates/modifies the name (what is used in the URL) when they are too long. For example, if your file name is "my-long-file-name-for-a-js-but-like-very-very-very-long-and-could-be-longer.js", when created (or imported) Nuxeo will truncate, by default, to "my-long-file-name-for-a-" (24 chars). So, when requesting it from its full name, it will fail.

You can change this value by using the nuxeo.path.segment.maxsize configuration parameter. Either in nuxeo.conf:

nuxeo.path.segment.maxsize=YOUR_VALUE

Or via XML:

<extension target="org.nuxeo.runtime.ConfigurationService" point="configuration">
  <property name="nuxeo.path.segment.maxsize">YOUR_VALUE</property>
</extension>

Notice the plugin tries to workaround this in a single attempt. In the previous example, if the path is...

/domain/aworkspace/afolder/yoursite/main/js/my-long-file-name-for-a-js-but-like-very-very-very-long-and-could-be-longer.js

...if Nuxeo throws a DocumentNotFoundException, the plugin will search inside /domain//aworkspace/afolder/yoursite/main/js for a Document whose dc:title is "my-long-file-name-for-a-js-but-like-very-very-very-long-and-could-be-longer.js".

Security Warning

The plugin sends the files as they are stored (either as single Nuxeo Document or inside the .zip file). This means no sanitizing is done, the JavaScript is not filtered and is sent as is.

If there is any risk of a user uploading a website that contains some malicious JavaScript (like getting the cookies to steel the session), then we recommend to add configuration to approve/reject the document before making it available.

The plugin could also be forked and sanitizing the JavaScript can be done (Nuxeo has APIs for this purpose). This means the plugin would then send html files that contain no <script> tag at all for example.

Support

These features are not part of the Nuxeo Production platform.

These solutions are provided for inspiration and we encourage customers to use them as code samples and learning resources.

This is a moving project (no API maintenance, no deprecation process, etc.) If any of these solutions are found to be useful for the Nuxeo Platform in general, they will be integrated directly into platform, not maintained in this repository.

License

Apache License, Version 2.0

Contributors: Thibaud Arguillere (https://github.com/ThibArg)

About Nuxeo

Nuxeo, developer of the leading Content Services Platform, is reinventing enterprise content management (ECM) and digital asset management (DAM). Nuxeo is fundamentally changing how people work with data and content to realize new value from digital information. Its cloud-native platform has been deployed by large enterprises, mid-sized businesses and government agencies worldwide. Customers like Verizon, Electronic Arts, ABN Amro, and the Department of Defense have used Nuxeo's technology to transform the way they do business. Founded in 2008, the company is based in New York with offices across the United States, Europe, and Asia. Learn more at www.nuxeo.com.

About

Languages

Language:Java 97.6%Language:HTML 2.4%