Condenser

Condenser is the react.js web interface to the blockchain-based social media platform, Hive.blog. It uses a Hive compatible blockchain, powered by DPoS Governance and ChainBase DB to store JSON-based content for a plethora of web applications.

Why would I want to use Condenser?

• Learning how to build blockchain-based web applications using Hive compatible blockchain as a content storage mechanism in react.js
• Reviewing the inner workings of the Hive.blog social media platform
• Assisting with software development for Hive.blog and Hive.io

Installation

Docker

If you would like to modify, build, and run condenser using docker, it's as simple as pulling in the github repo and issuing one command to build it, like this:

git clone https://gitlab.syncad.com/hive/condenser
cd condenser
./run.sh start dev

To build developer image manually, you can use:

docker build --target development --tag condenser:dev .

Docker Compose

If you like to run and build condenser and additionally a reverse-proxy using an Nginx Docker image, with companion Letsencrypt (SSL) support, you can simple launch the Docker-compose files via the included run.sh-scripts.

git clone https://gitlab.syncad.com/hive/condenser
cd condenser
./run.sh start proxy            # to start the nginx reverse proxy (with ssl support)
./run.sh start (prod|dev|stg)   # to build and start the condensor image
./run.sh logs (prod|dev|stg)    # (optionally) to attach to the condensor image and inspect its logs

Building from source without docker (the 'traditional' way):

(better if you're planning to do condenser development)

Clone the repository and make a tmp folder

git clone https://gitlab.syncad.com/hive/condenser
cd condenser
mkdir tmp

Install dependencies

Install at least Node.js v12 if you don't already have it.

Condenser is known to successfully build using node 12.6, npm 6.13.4, and yarn 1.22.4.

We use the yarn package manager instead of the default npm. There are multiple reasons for this, one being that we have hive-js built from source pulling the gitlab repo as part of the build process and yarn supports this. This way the library that handles keys can be loaded by commit hash instead of a version name and cryptographically verified to be exactly what we expect it to be. Yarn can be installed with npm, but afterwards you will not need to use npm further.

npm install -g yarn
yarn add babel-cli
yarn install --frozen-lockfile --ignore-optional
yarn run build

To run condenser in production mode, run:

yarn run production

When launching condenser in production mode it will automatically use 1 process per available core. You will be able to access the front-end at http://localhost:8080 by default.

To run condenser in development mode, run:

yarn run start

It will take quite a bit longer to start in this mode (~60s) as it needs to build and start the webpack-dev-server.

By default you will be connected to community public api node at https://api.hive.blog. This is actually on the real blockchain and you would use your regular account name and credentials to login - there is not an official separate testnet at this time. If you intend to run a full-fledged site relying on your own, we recommend looking into running a copy of hive (steemd) locally instead https://gitlab.syncad.com/hive/hive.

Debugging SSR code

yarn debug will build a development version of the codebase and then start the local server with --inspect-brk so that you can connect a debugging client. You can use Chromium to connect by finding the remote client at chrome://inspect/#devices.

Configuration

The intention is to configure condenser using environment variables. You can see the names of all of the available configuration environment variables in config/custom-environment-variables.json. Default values are stored in config/defaults.json.

Environment variables using an example like this:

export SDC_CLIENT_STEEMD_URL="https://api.hive.blog"
export SDC_SERVER_STEEMD_URL="https://api.hive.blog"

Keep in mind environment variables only exist in your active session, so if you wish to save them for later use you can put them all in a file and source them in.

If you'd like to statically configure condenser without variables you can edit the settings directly in config/production.json. If you're running in development mode, copy config/production.json to config/dev.json with cp config/production.json config/dev.json and adjust settings in dev.json.

If you're intending to run condenser in a production environment one configuration option that you will definitely want to edit is server_session_secret which can be set by the environment variable SDC_SESSION_SECRETKEY. To generate a new value for this setting, you can do this:

node
> crypto.randomBytes(32).toString('base64')
> .exit

Style Guides For Submitting Pull Requests

File naming and location

• Prefer CamelCase js and jsx file names
• Prefer lower case one word directory names
• Keep stylesheet files close to components
• Component's stylesheet file name should match component name

Js & Jsx

We use prettier to autoformat the code, with this configuration. Run yarn run fmt to format everything in src/, or yarn exec -- prettier --config .prettierrc --write src/whatever/file.js for a specific file.

CSS & SCSS

If a component requires a css rule, please use its uppercase name for the class, e.g. "Header" class for the header's root div. We adhere to BEM methodology with exception for Foundation classes, here is an example for the Header component:

<!-- Block -->
<ul class="Header">
  ...
  <!-- Element -->
  <li class="Header__menu-item">Menu Item 1</li>
  <!-- Element with modifier -->
  <li class="Header__menu-item--selected">Element with modifier</li>
</ul>

Storybook

yarn run storybook

Testing

Run test suite

yarn test

will run jest

Test endpoints offline

If you want to test a server-side rendered page without using the network, do this:

yarn build
OFFLINE_SSR_TEST=true NODE_ENV=production node --prof lib/server/index.js

This will read data from the blobs in api_mockdata directory. If you want to use another set of mock data, create a similar directory to that one and add an argument OFFLINE_SSR_TEST_DATA_DIR pointing to your new directory.

Issues

To report a non-critical issue, please file an issue on this GitHub project.

If you find a security issue please report details to trusted community members.

We will evaluate the risk and make a patch available before filing the issue.

Condenser's Oauth Server

We implemented a simple Oauth 2.0 server for authorizing Hive users in our forked Rocket Chat. This Oauth server is disabled by default. To enable it, set environment variable SDC_OAUTH_SERVER_ENABLE to yes, when starting Condenser. Server authorizes users logged in via passing username and password or logged in via hivesigner. Only these login methods are available at login page, when we are in Oauth flow. When user is already logged in via other method, e.g. hiveauth or keychain, Condenser displays a page telling that user should logout and try again.

Test login to Rocket Chat via Condenser's Oauth Server on local development host

Run our forked Rocket Chat instance on http://localhost:3000. Do this in branch from PR1. Setup a new Oauth client using page http://localhost:3000/admin/settings/OAuth. You can see how my client is configured below:

syncad@dev-66:~/src/condenser$ curl -H "X-Auth-Token: ${ROCKET_CHAT_LOCAL_TOKEN}" -H "X-User-Id: ${ROCKET_CHAT_LOCAL_USER_ID}" http://localhost:3000/api/v1/settings.oauth
{
  "services": [
    {
      "_id": "8JFqgmogbfbiFKtcA",
      "service": "hiveblog",
      "accessTokenParam": "access_token",
      "authorizePath": "/oauth/authorize",
      "avatarField": "picture",
      "buttonColor": "#1d74f5",
      "buttonLabelColor": "#FFFFFF",
      "buttonLabelText": "Login with Hiveblog",
      "channelsAdmin": "rocket.cat",
      "channelsMap": "{\n\t\"rocket-admin\": \"admin\",\n\t\"tech-support\": \"support\"\n}",
      "clientId": "openhive_chat",
      "custom": true,
      "emailField": "",
      "groupsClaim": "groups",
      "identityPath": "/oauth/userinfo",
      "identityTokenSentVia": "default",
      "keyField": "username",
      "loginStyle": "redirect",
      "mapChannels": false,
      "mergeRoles": false,
      "mergeUsers": true,
      "nameField": "",
      "rolesClaim": "roles",
      "rolesToSync": "",
      "scope": "openid profile",
      "serverURL": "http://localhost:8080",
      "showButton": true,
      "tokenPath": "/oauth/token",
      "tokenSentVia": "header",
      "usernameField": "username"
    }
  ],
  "success": true
(syncad) syncad@dev-66:~/src/condenser$

Set environment variables for Oauth Server in another terminal:

export SDC_OPENHIVE_CHAT_API_URI="http://localhost:3000" ;
export SDC_OPENHIVE_CHAT_URI="http://localhost:3000" ;
export SDC_OAUTH_SERVER_ENABLE="yes" ;
export SDC_OAUTH_SERVER_CLIENTS="`cat config/oauth-server-clients-development.json`" ;

Start Condenser in the same terminal:

yarn start

Login to Rocket Chat via Condenser's Oauth Server on production host

Do something similar as in section above, but edit file config/oauth-server-clients-development.json according to your needs. It's very important to set hard to guess secret for Oauth client in Condenser. Set the same secret in Rocket Chat Oauth client's config.

Integration with Rocket Chat via iframe

Create following environment variable:

export CREATE_TOKENS_FOR_USERS="true" ;

Start Rocket Chat.

Activate the Iframe Integration in Rocket Chat instance – follow guide "Part I: Activate the Iframe Integration". On development machine set http://localhost:8080/chat/parking for "Iframe URL", and http://localhost:8080/chat/sso for "API URL". Go to Administration > Settings > General > Iframe Integration and check "Enable Send" and "Enable Receive".

Create Personal Access Token for an admin account in Rocket Chat (http://localhost:3000/account/tokens) for use in step below. Hit "Ignore Two Factor Authentication" upon creation.

TODO We should give the lowest possible privileges to that token. Maybe create a new, custom role in Rocket Chat.

Create following environment variables:

export SDC_OPENHIVE_CHAT_API_URI="http://localhost:3000" ;
export SDC_OPENHIVE_CHAT_URI="http://localhost:3000" ;
export SDC_OPENHIVE_CHAT_IFRAME_INTEGRATION_ENABLE="yes" ;
export SDC_OPENHIVE_CHAT_IFRAME_VISIBLE="yes" ;
export SDC_OPENHIVE_CHAT_IFRAME_CREATE_USERS="no" ;
export SDC_OPENHIVE_CHAT_ADMIN_USER_ID="<your-admin-user-id>" ;
export SDC_OPENHIVE_CHAT_ADMIN_USER_TOKEN="<your-admin-user-token>" ;

Variable SDC_OPENHIVE_CHAT_IFRAME_INTEGRATION_ENABLE enables or disables iframe integration. Variable SDC_OPENHIVE_CHAT_IFRAME_VISIBLE shows or hides chat button and iframe (hidden iframe still operates in background).

Start Condenser.

Logger

We implemented Logger in src/app/utils/Logger.js file, for logging to console or nowhere (suppress logging), used as a replacement for standard console messaging. Use it this way:

import { logger } from './Logger';
logger.info('my info');
logger.error('my error');

Then you should see message in console in development environment, but not in production environment. This behavior depends on following environment variables (showing default values):

export SDC_LOGGER_OUTPUT="noop"
export SDC_LOGGER_LOG_LEVEL="all"
export SDC_LOGGER_ADMINS=""

Setting SDC_LOGGER_OUTPUT="console" enables logging to console.

You can set SDC_LOGGER_LOG_LEVEL to following log levels"

off
fatal
error
warn
info
debug
trace
all

When user is logged in and he is listed in SDC_LOGGER_ADMINS (should be string with Hive usernames delimited with space), the application allows him to see all Logger messages on all log levels, regardless of anything else, so also on production.