MonkeyLearn Pane

sanity-plugin-monkeylearn-pane

Send the contents of Portable Text to MonkeyLearn's API to Extract insights with the power of Machine Learning!

Notes

1. You will need a MonkeyLearn account (free!) in order to use this plugin
2. By default the plugin will only display extraction results. To perform the below interactions back to your source Document, you will need to drop in your own Component in the Pane options.

Installation

sanity install monkeylearn-pane

This is designed to be used as a Component inside of a View.

// ./src/deskStructure.js
import MonkeyLearnPane from 'sanity-plugin-monkeylearn-pane'

// ...all other list items

S.view
  .component(MonkeyLearnPane)
  .options({
    extractors: [`ex_vqBQ7V9B`],
    field: 'content',
    // Optional but recommended!
    // Use your own Component to display/interact with extractions
    displayComponent: <Report />,
  })
  .title('Analysis')

The .options() configuration works as follows:

• extractors ([string], required) The ID's of the Text Extractors you want to produce reports for.
• field (string, required) A dot-notated string from the document object to a field containing the Portable Text array.
• displayComponent (React Component, optional) A custom Component for handling the report from each text extraction. You will need to BYO a Component that will accept the Document ID and Report as props, so that it can do something fancy like generating tag documents and sending a patch to the currently edited document.

Customising the output

The example shown above is included in this repo, see ReportTagCreator.js – this Component takes the report, generates a tag document if one does not already exists, and can send a patch back to the currently edited document in an array of references named tags.

License

MIT © Simeon Griggs See LICENSE