@frappy/react-datasets v1.4.0

React Data Sets

Data Set Management and Retrieval Components and Functions.

1. Data Set Management
2. Data Set Handler
3. Display Components
   1. TimeSeriesChart
   2. ImageView
   3. BinaryDownloadLink
4. API Endpoints

The following data types are currently supported:

• Time-series (uploaded as CSV file)
• Images (uploaded as single image)
• Generic data (uploaded as JSON)
• Binary (uploaded as single file)

This package is a counterpart to the following server-side packages:

• @frappy/node-datasets (NodeJS)
• frappyflaskdatasets (Python)

Data Set Management

Usage (the wrapping in a PermissionCheck component is optional, but recommended). The API will additionally check for required permissions:

import React from "react"
import { Switch, Route } from "react-router"
import { PermissionCheck } from "@frappy/react-authentication"
import { DataSetManager } from "@frappy/react-datasets"

// example within a router
const MyApp = props => (
    <Switch>
        <Route path="/admin/data" exact component={() =>
            <PermissionCheck currentUser={props.currentUser} requiredPermissions="data" showError>
                <DataSetManager 
                    assignments={{ 
                        group1: {
                            label: "Group 1",
                            dataTypes: ["forecastInput", "decisionInput"],
                        },       
                    }} 
                />
            </PermissionCheck>
        }/>
    </Switch>
)  

This will bind a component for this route that allows to list data sets, create new data sets, inspect them, update their meta data or delete them.

Properties

• apiPrefix - default /api/data-sets - the prefix for all API calls
• assignments - default {} - a JSON object that allows data assignments, which can later be used to fetch limited data. The structure of this is as follows:

• thumbnailSize - default 150 - the maximum width/height of the thumbnail generated of an IMAGE data set

{
    "firstLevelKey": {
        "label": "The First Key",
        "dataTypes": ["type1", "type2", "type3"]
    },
    "secondLevelKey": {
        "label": "Another Topic",
        "dataTypes": ["type4", "type5"]
    }
}

Data Set Handling

In your code, wherever you need access to data sets, you can simply use the DataSetHandler, which is a utility class independent from React that facilitates the communication with the backend:

import { DataSetHandler } from "@frappy/react-datasets"

const dataHandler = new DataSetHandler()

// retrieve only meta data (save bandwidth) - no payload
dataHandler.getDataSetMeta("5e846d104e61db060094ed14").then(dataSet => {
    // only meta data retrieved, not the payload
    console.log(dataSet.label, dataSet.type, dataSet.userId)
})

Constructor Arguments

• apiPrefix - default /api/data-sets - the API prefix to use for all operations

Methods

All methods return a Promise, which resolves the requested data.

• getDataSetList(assignment, assignmentType, dataType) - retrieves a list of all data sets (meta information only), all 3 parameters are optional.
  • assignment queries for available keys of the dataSet.assignments (assignment group)
  • assignmentType queries for the type inside the dataSet.assignments.<someAssignment> array. Note that if assignmentType is provided, you also have to provide assignment
  • dataType the data type of the data set, e.g. "TIME_SERIES" or "IMAGE" - if you only want to filter for this, just null the first 2 parameter (e.g. handler.getDataSetList(null, null, "IMAGE") will give you all images).
• getDataSetMeta(dataSetId) - retrieve only meta information about the data set
• getDataSet(dataSetId) - retrieve the full data set including the payload
• getThumbnail(dataSetId) - if the data set represents an image, this will retrieve a thumbnail of the image as base64 encoded string, which can be directly embedded into an <img src={base64Content} />
• getFullImage(dataSetId) - if the data set represents an image, this will retrieve the full image and return it as object URL, which can be directly embedded into an <img src={objectUrl} />
• getBinary(dataSetId) - if the data set represents any binary format (including images), this allows you to return an object URL, which can be included in a hyperlink <a href={objectUrl}>Text</a>.

Display Components

TimeSeriesChart

After uploading a CSV time series, this component will parse the data set's payload and convert it into what Plotly (charting library) expects as input.

import React from "react"
import { TimeSeriesChart } from "@frappy/react-datasets"

const MyComponent = props => (
    <TimeSeriesChart data={props.dataSet.payload} width={800} height={400} title={"My Chart"} showLegend />
)

Properties

• data - the full data set payload retrieved from the API. It needs to be a JSON object with keys: columns, data and indexColumn, which needs to be one of the columns. The index column values needs to be Unix timestamps.
  • In case you want to use a custom date format, you also need to add the dateFormat, specified either as Python date format or Luxon date format to the main JSON object, that allows the front-end to parse the date.
  • In case you want to provide custom column names, simply provide a columnMapping JSON object that maps from the original name of the column to the label you want it to have.
• width - default 800 - the width of the chart area
• height - default 400 - the height of the chart area
• title - default "Chart" - a custom title for the chart
• showLegend - default false - whether to show a legend for each of the time series value lines.

ImageView

Fetches an image (full size binary) from the API and renders it. This component is only necessary, if you cannot use the default <img> tag due to the endpoints for retrieving data sets being protected by authentication. If your backend for the public data set endpoints is not protected, don't use this component.

import React from "react"
import { ImageView } from "@frappy/react-datasets"

const MyComponent = props => (
    <ImageView src={url} width={200} height={300} />
)

Properties

• src - the URL to the image, e.g. /api/data-sets/5e846d104e61db060094ed14/image
• width - (optional) the width of the image to display
• height - (optional) the width of the image to display
• style - (optional) a custom style for the img tag that this component will render

BinaryDownloadLink

Allows a user to download a binary file from the API. The developer can choose to provide a custom download text, which will be rendered as a link or even provide children to be rendered instead of a static text. If none is provided, a generic "Download File" text will be displayed.

Properties

• apiPrefix - default /api/data-sets - in case you use a different API endpoint for the backend
• fileName - default <label>.<extension> - override the default file name for this. Please provide the file name without extension, the component will figure that out automatically using the payload's mimeType
• dataSet - the data set for which to download the file
• text - optional instead of the default text provide a custom text (if not using children).
• payload - optional the data set payload of the binary data set, if this is not provided, the component will have to run a separate query for the mime type.

Examples

Most basic example showing a simple "Download File" text.

import { BinaryDownloadLink } from "@frappy/react-datasets"

// most basic example show default "Download File" text
const link = dataSet => <BinaryDownloadLink dataSet={dataSet} /> 

// custom text
const link = dataSet => <BinaryDownloadLink dataSet={dataSet} text={"Download this awesome file"} />

// custom content, in this example the whole list will be come the link (intended to be used with icons + text)
const link = dataSet => (
    <BinaryDownloadLink dataSet={dataSet}>
        <ul>   
            <li>Some</li>
            <li>Arbitrary Content</li>
        </ul>
    </BinaryDownloadLink>
)

API Endpoints

The following assumes the default prefix /api/data-sets. URL parameter are prefixed with :

• GET /api/data-sets - retrieves a list of all data sets (without their payload)
• GET /api/data-sets/:dataSetId - retrieves a full data set (including payload)
• GET /api/data-sets/:dataSetId/meta - retrieves a specific data set (without payload)
• GET /api/data-sets/:dataSetId/image - assumes the data set specified is an image and if so, returns the image as binary

Data Management Endpoints

• POST /api/data-sets - creates a new data set. The request body is multi-part/form-data and contains a file (CSV, image or JSON, depending on the data set type) and meta key. The meta value is provided as JSON (stringified) and has attributes such as type, label and other information specific to specific data set types.
• POST /api/data-sets/:dataSetId - updates the meta information of a data set. The body can contain any attribute on the root level of the JSON representation structure and payload elements (only one level depth).
• DELETE /api/data-sets/:dataSetId - deletes a data set. In case of an image, this will also delete the image on disc.