knight/datahub
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #641 from datopian/fix/bin/use-example-code

Browse Source 
CLI tool improvements
This commit is contained in:
Rising Odegua 2022-01-31 00:54:25 -08:00 committed by GitHub
commit 3cd6e2aad7
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 53 additions and 52 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

99
bin/README.md
View File

@ -1,22 +1,52 @@
# portal-cli-app
Simple CLI for displaying and publishing datasets
## DataHub CLI
This command line tool helps you to instantly build and deploy data-driven pages. It doesn't require you to know any frontend technologies (HTML, CSS, JS) so that you can focus on your data. Simply run a command (`datahub deploy`) and get a URL for your data page.
You can still customize the look and feel of your data-driven pages. It uses [Portal.js][] under the hood and with some basic knowledge of frontend development users can easily get started.
[Portal.js]: https://portaljs.org/learn
## Install
Install the CLI tool:
```bash
npm install -g 'https://gitpkg.now.sh/datopian/portal.js/bin?main'
# or with yarn
yarn add 'https://gitpkg.now.sh/datopian/portal.js/bin?main'
```
Next, confirm if the `datahub` command is available:
```bash
which datahub
# should print path to the executable
```
> Note: It is recommended you install the CLI tool globally so you can run it from anywhere on your machine.
## Features
### Show
Easily preview data locally.
Easily preview data-driven pages locally.
I have a dataset `my-data` with the following content:
```
README.md
data.csv
## descriptor is optional (we infer if not there)
# datapackage.json
# descriptor is optional (we infer if not there)
datapackage.json
```
I can do the following:
```
cd my-data
portal show
datahub show
```
I get a nice dataset page like:
The single show command gives me access to the following:
* Elegant presentation
@ -25,57 +55,28 @@ The single show command gives me access to the following:
* Show graphs
* Data summary
Show works with:
Show works with:
* README + csv
* Frictionless dataset
* Frictionless resource
* Pure README with frontmatter
### Deploy [Not implemented yet]
Easily publish your dataset online.
If i have a dataset `my_data`:
```
cd my-data
portal deploy
```
Gives me a url like:
`Myusername-my-dataset.datahub.io`
Deploy: what does it do?
Once you are ready to deploy your data:
```
datahub deploy
```
It starts building your production ready data-driven page and deploys it in the cloud. With the single command you get a URL for you page (e.g., `https://my-data-abcd123.datahub.io`). Now, share it with everyone!
Why to deploy my data?
* Deploys a shareable url with the content of show
* Semi-private
* Can integrate access control (?)
* Deploys a data API
* [Other integrations e.g. push to google spreadsheets]
* Dashboard showing your DataHub/Portal projects
## User Guide
Install portal-cli-app globally on your computer
```bash
npm install -g git+https://github.com/datopian/portal-experiment.git
```
>Note: It is recommended you install portal-cli-app globally so you can run the portal show command from anywhere in your computer.
In a folder with your dataset and optionally datapackage.json, run the command:
```bash
portal show
```
You can also specify a folder from directory:
```bash
portal show path=\some\path\dataset
```
Optional args you can pass yo the show command are:
- path: The path to a dataset directory. If blank, defaults to current working directory
- port: The port number to display your dataset in. Defaults to 3000.
- npm: Whether or not to use npm when install packages. Defaults to `false`, uses yarn.
```
portal show [path=/some/path/dataset | port=4000 | npm=true]
```
* Dashboard showing your DataHub/Portal projects

6
bin/index.js
View File

@ -77,10 +77,10 @@ async function run() {
process.exit(1)
}
const portalGithubRepo = "https://github.com/datopian/portal-experiment.git"
const portalLocalRepoDirectory = path.join(datasetPath, 'portal-experiment')
const portalGithubRepo = 'https://github.com/datopian/portal.js/tree/main/examples/dataset-frictionless'
const portalLocalRepoDirectory = path.join(datasetPath, 'portal')
const cloneRepoCmd = `cd ${datasetPath} && git clone ${portalGithubRepo}`
const cloneRepoCmd = `cd ${datasetPath} && npx create-next-app portal -e ${portalGithubRepo}`
// CD into portalLocalRepoDirectory, create an .env file and write PORTAL_DATASET_PATH={datasetPath} to it
const createEnvFileCmd = `cd ${portalLocalRepoDirectory} && touch .env && echo PORTAL_DATASET_PATH=${datasetPath} >> .env`