Skip to content

This obsolete project was replaced by https://github.com/BloomBooks/BloomLibrary2.... Bloom Library Single Page App, using AngularJS & Bootstrap, parse server backend.

License

Notifications You must be signed in to change notification settings

BloomBooks/LegacyBloomLibrary

Repository files navigation

A web site for sharing literacy materials, especially templates for translation into minority languages.


Quick Start

Make sure you have yarn. Look in packages.json under the "engines" key. Get that version of Nodejs.

git clone  https://github.com/BloomBooks/BloomLibrary.git
cd BloomLibrary
yarn

In one terminal (e.g. in vscode), run

yarn watch

In another terminal (e.g. to its side, in vscode), run

yarn serve

That should open a web browser page at localhost:3000.

Important Notes about S3 Setup

The router we use makes all pages have a hash before the final part of the url. e.g. bloomlibrary.org/#/browse.

When navigating within the site internally, the router happily deals with this. When navigating to a particular page directly via a url, we would prefer to not include the hash. e.g. bloomlibrary.org/browse.

To date, we have accomplished this by two different methods on S3.

  1. Routing Rules
  • In AWS, navigate to the bucket (bloomlibrary.org or dev.bloomlibrary.org)
  • Properties
  • Static website hosting
  • Redirection rules
  • something like
<RoutingRules>
<RoutingRule>
  <Condition>
    <KeyPrefixEquals>about</KeyPrefixEquals>
  </Condition>
  <Redirect>
    <ReplaceKeyPrefixWith>#/about</ReplaceKeyPrefixWith>
  </Redirect>
</RoutingRule>
<RoutingRules>
  • We recently discovered these don't work on Safari (or iOS Chrome)
  1. An empty file with redirect rules
  • In AWS, navigate to the bucket (bloomlibrary.org or dev.bloomlibrary.org)
  • Upload an empty file for your page. e.g. browse
  • Click on that file
  • Properties
  • Metadata
  • Add key=Website-Redirect-Location, value=/#/page-name, e.g. /#/browse

Careful! Method 2 won't work if you have implemented method 1.

Note: Method 2 doesn't work with urls containing information after the would-be file name, e.g. https://bloomlibrary.org/readBook/jnG2YFeIIG

About the grunt files

We built these using ngBoilerplate. Read the build.config.js file to see most of what is going on. See the readme of ngBoilerplate for more information. Here are key things:

Grunt merges many vendor and app-specific less files together into the bloom_xyz.css file. The list of these files is main.less. If you have a css file to load separately, list that in build.config.js.

Normally we have tests run via the offscreen browser phantomjs. If you want to debug using a normal browser, you'll find that setting in karma/X:\dev\BloomLibrary\karma\karma-unit.tpl.js.

Search Engine Optimization

A problem with the current approach is that Google and other web crawlers don't see anything more than our home page. This is discussed at length at https://docs.google.com/document/d/1XAJRmQoJHYkwkf4CVW_OWiBkRT1jqufKF2rJajcnxIw/edit?usp=sharing.

SIL Build Agent Setup

AWS Credentials

The credentials needed to upload the content to the S3 bucket must be supplied in the AWS credential store. On current agents, this can be found at c:\users\bob\.aws\credentials. When running the build, pass the profile name as the ProfileName parameter of the Upload target.

Install on machine

7zip must be accessible in the PATH.

Copyright and License

Copyright 2013-2018 SIL International MIT/X11

About

This obsolete project was replaced by https://github.com/BloomBooks/BloomLibrary2.... Bloom Library Single Page App, using AngularJS & Bootstrap, parse server backend.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Sponsor this project

Packages

No packages published