Table of Contents
- 🏄 Get Started
- 🦑 Environment variables
- 🦀 Data Sources
- 👩🎤 Storybook
- 🤖 Testing
- ✨ Code Style
- 🛳 Production
- ⬆️ Deployment
- 💖 Contributing
- 🍴 Forking
- 💰 Pricing Options
- ✅ GDPR Compliance
- 🏛 License
The app is a React app built with Next.js + TypeScript + CSS modules and will connect to Ocean remote components by default.
Prerequisites:
- Node.js (required). Check the .nvmrc file to make sure you are using the correct version of Node.js.
- nvm (recommended). This is the recommended way to manage Node.js versions.
- Git is required to follow the instructions below.
To start local development:
git clone [email protected]:oceanprotocol/market.git
cd market
# when using nvm to manage Node.js versions
nvm use
npm install
# in case of dependency errors, rather use:
# npm install --legacy-peer-deps
npm startThis will start the development server under
http://localhost:8000.
Using the ocean-market with barge components is recommended for advanced users, if you are new we advice you to use the ocean-market first with remote networks. If you prefer to connect to locally running components instead of remote connections, you can spin up barge and use a local Ganache network in another terminal before running npm start. To fully test all The Graph integrations, you have to start barge with the local Graph node:
git clone [email protected]:oceanprotocol/barge.git
cd barge
# startup with local Ganache and Graph nodes
./start_ocean.sh --with-thegraphBarge will deploy contracts to the local Ganache node which will take some time. At the end the compiled artifacts need to imported over to this project as environment variables. The set-barge-env script will do that for you and set the env variables to use this local connection in .env in the app. You also need to append the chainIdsSupported array with the barge's ganache chainId (8996) in the app.config.js file.
If you are using macOS operating system you should also make same changes to the provider url since the default barge ip can not be accessed due to some network constraints on macOs. So we should be using the 127.0.0.1:8030 (if you have changed the provider port please use that here as well) for each direct call from the market to provider, but we should keep the internal barge url http://172.15.0.4:8030/ (this is the default ip:port for provider in barge, if changed please use the according url). So on inside src/@utils/provider.ts if on macOS you can hardcode this env variable NEXT_PUBLIC_PROVIDER_URL or set
127.0.0.1:8030 as providerUrl on all the methods that call ProviderInstance methods. (eg: getEncryptedFiles, getFileDidInfo, downloadFile etc). You should use the same provider url for src/@utils/nft.ts inside setNFTMetadataAndTokenURI and setNftMetadata and src/components/Publish/index.tsx inisde encrypt method (if you set the env variable there's no need to do this). You also need to use local ip's for the subgraph (127.0.0.1 instead of 172.15.0.15) and the metadatacache (127.0.0.1 instead of 172.15.0.5).
Once you want to switch back to using the market against remote networks you need to comment or remove the env vars that are set by set-barge-env script.
cd market
npm run set-barge-env
npm startTo use the app together with MetaMask, importing one of the accounts auto-generated by the Ganache container is the easiest way to have test ETH available. All of them have 100 ETH by default. Upon start, the ocean_ganache_1 container will print out the private keys of multiple accounts in its logs. Pick one of them and import into MetaMask. Barge private key example : 0xc594c6e5def4bab63ac29eed19a134c130388f74f019bc74b8f4389df2837a58
Cleaning all Docker images so they are fetched freshly is often a good idea to make sure no issues are caused by old or stale images:
docker system prune --all --volumes
The app.config.js file is setup to prioritize environment variables for setting each Ocean component endpoint. By setting environment variables, you can easily switch between Ocean networks the app connects to, without directly modifying app.config.js.
For local development, you can use a .env file:
# modify env variables, Goerli is enabled by default when using those files
cp .env.example .envAll displayed data in the app is presented around the concept of one asset, which is a combination of:
- metadata about an asset
- the actual asset file
- the NFT which represents the asset
- the datatokens representing access rights to the asset file
- financial data connected to these datatokens, either a fixed rate exchange contract or a dispenser for free assets
- calculations and conversions based on financial data
- metadata about publisher accounts
All this data then comes from multiple sources:
All initial assets and their metadata (DDO) is retrieved client-side on run-time from the Aquarius instance, defined in app.config.js. All app calls to Aquarius are done with 2 internal methods which mimic the same methods in ocean.js, but allow us:
- to cancel requests when components get unmounted in combination with axios
- hit Aquarius as early as possible without relying on any ocean.js initialization
Aquarius runs Elasticsearch under the hood so its stored metadata can be queried with Elasticsearch queries like so:
import { QueryResult } from '@oceanprotocol/lib/dist/node/metadatacache/MetadataCache'
import { queryMetadata } from '@utils/aquarius'
const queryLatest = {
query: {
// https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-query-string-query.html
query_string: { query: `-isInPurgatory:true` }
},
sort: { created: 'desc' }
}
function Component() {
const { appConfig } = useMarketMetadata()
const [result, setResult] = useState<QueryResult>()
useEffect(() => {
if (!appConfig.metadataCacheUri) return
const source = axios.CancelToken.source()
async function init() {
const result = await queryMetadata(query, source.token)
setResult(result)
}
init()
return () => {
source.cancel()
}
}, [appConfig.metadataCacheUri, query])
return <div>{result}</div>
}For components within a single asset view the useAsset() hook can be used, which in the background gets the respective metadata from Aquarius.
import { useAsset } from '@context/Asset'
function Component() {
const { ddo } = useAsset()
return <div>{ddo}</div>
}Most financial data in the market is retrieved with GraphQL from our own subgraph, rendered on top of the initial data coming from Aquarius.
The app has Urql Client setup to query the respective subgraph based on network. In any component this client can be used like so:
import { gql, useQuery } from 'urql'
const query = gql`
query TopSalesQuery {
users(first: 20, orderBy: totalSales, orderDirection: desc) {
id
totalSales
}
}
`
function Component() {
const { data } = useQuery(query, {}, pollInterval: 5000 })
return <div>{data}</div>
}Publishers can fill their account's ENS domain profile and when found, it will be displayed in the app.
For this our own ens-proxy is used, within the app the utility method getEnsProfile() is called as part of the useProfile() hook:
import { useProfile } from '@context/Profile'
function Component() {
const { profile } = useProfile()
return (
<div>
{profile.avatar} {profile.name}
</div>
)
}Based on list-purgatory some assets get additional data. Within most components this can be done with the internal useAsset() hook which fetches data from the market-purgatory endpoint in the background.
For asset purgatory:
import { useAsset } from '@context/Asset'
function Component() {
const { isInPurgatory, purgatoryData } = useAsset()
return isInPurgatory ? <div>{purgatoryData.reason}</div> : null
}For account purgatory:
import { useAccount } from 'wagmi'
import { useAccountPurgatory } from '@hooks/useAccountPurgatory'
function Component() {
const { address } = useAccount()
const { isInPurgatory, purgatoryData } = useAccountPurgatory(address)
return isInPurgatory ? <div>{purgatoryData.reason}</div> : null
}All displayed chain & network metadata is retrieved from https://chainid.network on build time and integrated into NEXT's GraphQL layer. This data source is a community-maintained GitHub repository under ethereum-lists/chains.
Within components this metadata can be queried for under allNetworksMetadataJson. The useNetworkMetadata() hook does this in the background to expose the final networkDisplayName for use in components:
export default function NetworkName(): ReactElement {
const { isTestnet } = useNetworkMetadata()
const { networkData, networkName } = useNetworkMetadata()
return (
<>
{networkName} {isTestnet && `(Test)`}
</>
)
}Storybook helps us build UI components in isolation from our app's business logic, data, and context. That makes it easy to develop hard-to-reach states and save these UI states as stories to revisit during development, testing, or QA.
To start adding stories, create a index.stories.tsx inside the component's folder:
src └─── components │ └─── @shared │ └─── │ │ index.tsx │ │ index.module.css │ │ index.stories.tsx │ │ index.test.tsx
Starting up the Storybook server with this command will make it accessible under http://localhost:6006:
npm run storybookIf you want to build a portable static version under storybook-static/:
npm run storybook:buildTest runs utilize Jest as test runner and Testing Library for writing tests.
All created Storybook stories will automatically run as individual tests by using the StoryShots Addon.
Creating Storybook stories for a component will provide good coverage of a component in many cases. Additional tests for dedicated component functionality which can't be done with Storybook are created as usual Testing Library tests, but you can also import existing Storybook stories into those tests.
Executing linting, type checking, and full test run:
npm testWhich is a combination of multiple scripts which can also be run individually:
npm run lint
npm run type-check
npm run jestA coverage report is automatically shown in console whenever npm run jest is called. Generated reports are sent to CodeClimate during CI runs.
During local development you can continuously get coverage report feedback in your console by running Jest in watch mode:
npm run jest:watchCode style is automatically enforced through ESLint & Prettier rules:
- Git pre-commit hook runs
prettieron staged files, setup with Husky - VS Code suggested extensions and settings for auto-formatting on file save
- CI runs a linting & TypeScript typings check as part of
npm test, and fails if errors are found
For running linting and auto-formatting manually, you can use from the root of the project:
# linting check
npm run lint
# auto format all files in the project with prettier, taking all configs into account
npm run formatTo create a production build, run from the root of the project:
npm run build
# serve production build
npm run serveEvery branch or Pull Request is automatically deployed to multiple hosts for redundancy and emergency reasons:
A link to a preview deployment will appear under each Pull Request.
The latest deployment of the main branch is automatically aliased to market.oceanprotocol.com, where the deployment on Netlify is the current live deployment.
We welcome contributions in form of bug reports, feature requests, code changes, or documentation improvements. Have a look at our contribution documentation for instructions and workflows:
We encourage you to fork this repository and create your own data marketplace. When you publish your forked version of this market there are a few elements that you are required to change for copyright reasons:
- The typeface is copyright protected and needs to be changed unless you purchase a license for it.
- The Ocean Protocol logo is a trademark of the Ocean Protocol Foundation and must be removed from forked versions of the market.
- The name "Ocean Market" is also copyright protected and should be changed to the name of your market.
Additionally, we would also advise that you retain the text saying "Powered by Ocean Protocol" on your forked version of the marketplace in order to give credit for the development work done by the Ocean Protocol team.
Everything else is made open according to the apache2 license. We look forward to seeing your data marketplace!
If you are looking to fork Ocean Market and create your own marketplace, you will find the following guides useful in our docs:
To allow publishers to set pricing as "Fixed" you need to add the following environmental variable to your .env file: NEXT_PUBLIC_ALLOW_FIXED_PRICING="true" (default).
To allow publishers to set pricing as "Free" you need to add the following environmental variable to your .env file: NEXT_PUBLIC_ALLOW_FREE_PRICING="true" (default).
This allocates the datatokens to the dispenser contract which dispenses data tokens to users for free. Publishers in your market will now be able to offer their assets to users for free (excluding gas costs).
Ocean Market comes with prebuilt components for you to customize to cover GDPR requirements. Find additional information on how to use them below.
Feel free to adopt our provided privacy policies to your needs. Per default we cover four different languages: English, German, Spanish and French. Please be advised, that you will need to adjust some paragraphs in the policies depending on your market setup (e.g. the use of cookies). You can easily add or remove policies by providing your own markdown files in the content/pages/privacy directory. For guidelines on how to format your markdown files refer to our provided policies. The pre-linked content tables for these files are automatically generated.
Additionally, Ocean Market provides a privacy preference center for you to use. This feature is disabled per default since we do not use cookies requiring consent on our deployment of the market. However, if you need to add some functionality depending on cookies, you can simply enable this feature by changing the value of the NEXT_PUBLIC_PRIVACY_PREFERENCE_CENTER environmental variable to "true" in your .env file. This will enable a customizable cookie banner stating the use of your individual cookies. The content of this banner can be adjusted within the content/gdpr.json file. If no optionalCookies are provided, the privacy preference center will be set to a simpler version displaying only the title, text and close-button. This can be used to inform the user about the use of essential cookies, where no consent is needed. The privacy preference center supports two different styling options: 'small' and 'default'. Setting the style property to 'small' will display a smaller cookie banner to the user at first, only showing the default styled privacy preference center upon the user's customization request.
Now your market users will be provided with additional options to toggle the use of your configured cookie consent categories. You can always retrieve the current consent status per category with the provided useConsent() hook. See below, how you can set your own custom cookies depending on the market user's consent. Feel free to adjust the provided utility functions for cookie usage provided in the src/utils/cookies.ts file to your needs.
import { CookieConsentStatus, useConsent } from '@context/CookieConsent'
import { deleteCookie, setCookie } from '@utils/cookies'
// ...
const { cookies, cookieConsentStatus } = useConsent()
cookies.map((cookie) => {
const consent = cookieConsentStatus[cookie.cookieName]
switch (consent) {
case CookieConsentStatus.APPROVED:
// example logic
setCookie(`YOUR_COOKIE_NAME`, 'VALUE')
break
case CookieConsentStatus.REJECTED:
case CookieConsentStatus.NOT_AVAILABLE:
default:
// example logic
deleteCookie(`YOUR_COOKIE_NAME`)
break
}
})The privacy preference centre has two styling options default and small. The default view shows all of the customization options on a full-height side banner. When the small setting is used, a much smaller banner is shown which only reveals all of the customization options when the user clicks "Customize".
The style can be changed by altering the style prop in the PrivacyPreferenceCenter component in src/components/App.tsx. For example:
<PrivacyPreferenceCenter style="small" />Copyright 2023 Ocean Protocol Foundation Ltd.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.