🚜cypress-harvester

cypress-harvester on npmjs

A life enhancing plug-in for Cypress allowing you to easliy work with html <table>elements and repeating elements, whether it be for test assertions or for web scarping purposes.

Installing

Using npm:

$ npm install cypress-harvester --save-dev

Enable this plugin by adding this line to your project's cypress/support/commands.js:

import 'cypress-harvester'

Example - Testing static

Given a simple html table below:

Created	Account Id	Account Holder	Balance
10-04-2021 13:40:17	UA-11876-3	Terrell E. Evert	$33
10-04-2021 12:00:17	UA-10876-1	James L. Silver	$50.5
10-04-2021 13:00:17	UA-10346-1	Christian A. Lavalle	$-22.98

<table id="example" border="1">
  <tr>
      <td>Created</td>
      <td>Account Id</td>
      <td>Account Holder</td>
      <td>Balance</td>
  </tr>
  <tr>
      <td>10-04-2021 13:40:17</td>
      <td>UA-11876-3</td>
      <td>Terrell E. Evert</td>
      <td>$33</td>
  </tr>    
  <tr>
      <td>10-04-2021 12:00:17</td>
      <td>UA-10876-1</td>
      <td>James L. Silver</td>
      <td>$50.5</td>
  </tr>
  <tr>
      <td>10-04-2021 13:00:17</td>
      <td>UA-10346-1</td>
      <td>Christian A. Lavalle</td>
      <td>$-22.98</td>
  </tr>
</table>

When table is passed through the Cypress harvester, Cypress is able to easily extract data and convert to a json representation of the table:

cy.get('#example')
  .scrapeTable()
  .then((table) => {
    expect(table.getData()).to.deep.eq([
      {
        created: '10-04-2021 13:40:17',
        account_id: 'UA-11876-3',
        account_holder: 'Terrell E. Evert',
        balance: '$33',
      },
      {
        created: '10-04-2021 12:00:17',
        account_id: 'UA-10876-1',
        account_holder: 'James L. Silver',
        balance: '$50.5',
      },
      {
        created: '10-04-2021 13:00:17',
        account_id: 'UA-10346-1',
        account_holder: 'Christian A. Lavalle',
        balance: '$-22.98',
      },
    ]);
  });

Example - Testing repeating elements

Given a simple set of repeating elements below:

<div id="sale-items" style="margin-left:45px;">
  <div class="product">
    <div class="product-name">iPad Pro</div>
    <div class="model">11-inch Liquid Retina Display</div>
    <div class="price">$829.99</div>
  </div>
  <div class="product">
    <div class="product-name">iPad Air</div>
    <div class="model">64 GB  Wi-Fi + Cellular</div>
    <div class="price">$599.99</div>
  </div>
  <div class="product">
    <div class="product-name">iPad mini</div>
    <div class="model">256 GB  Wi-Fi + Cellular</div>
    <div class="price">$399</div>
  </div>
</div>

When the set of repeating elements is passed through to the scrapeElements scraper. It will yeild a nice json represenation of the data on the page. This will allow you to assert the data or save the results.

cy.get('#sale-items .product')
  .scrapeElements({
    elementsToScrape: [
      { label: 'product_name', locator: '.product-name' },
      { label: 'product_model', locator: '.model' },
      { label: 'item_price', locator: '.price' },
    ],
  })
  .then((scrapedData) => {
    expect(scrapedData.data).to.deep.eq([
      {
        product_name: 'iPad Pro',
        product_model: '11-inch Liquid Retina Display',
        item_price: '$829.99',
      },
      {
        product_name: 'iPad Air',
        product_model: '64 GB  Wi-Fi + Cellular',
        item_price: '$599.99',
      },
      {
        product_name: 'iPad mini',
        product_model: '256 GB  Wi-Fi + Cellular',
        item_price: '$399',
      },
    ]);
  });

Example - Web scraping records

This plugin also allows for scraping of data from tables which can be persisted to a json file:

cy.get('#example')
  .scrapeTable({
    exportFileName: 'scrapedData.json',
    exportFilePath: 'cypress/downloads',
  })
  .then((table) => {
    expect(table.exportStatus).to.contain(
      'Data table successfully saved'
    );
  });

A json representation of the html table is then saved to a json file within the cypress/downloads folder:

Other Useful assertions

cy.get('#example')
  .scrapeTable()
  .then((table) => {
    // assert the number of records in the table
    expect(table.rowCount()).to.eq(3);

    // validate a record exists in the table
    expect(
      table.hasItem({
        account_holder: 'Christian A. Lavalle',
      })
    ).to.have.property('account_id', 'UA-10346-1');

    // validate a record does not exist in the table
    expect(
      table.hasItem({
        account_holder: 'John Babs',
      })
    ).to.be.undefined;

    // check the tables' column labels
    expect(table.columnLabels).to.deep.eq([
      'Created',
      'Account Id',
      'Account Holder',
      'Balance',
    ]);

    // test whether column(s) are sorted
    expect(
      table.isPropertySorted(['account_id'], ['desc']),
      'account_id sorted in desc order'
    ).to.be.true;

    // find records matching a search term for a given property
    // the example below return only 2 records where the account holder contains 'ver'
    // it will find [Terrell E. Evert and James L. Silver]
    expect(
      table.containsItem(
        'account_holder', 'ver')
    ).to.deep.eq([
      {
        created: '10-04-2021 13:40:17',
        account_id: 'UA-11876-3',
        account_holder: 'Terrell E. Evert',
        balance: '$33',
      },
      {
        created: '10-04-2021 12:00:17',
        account_id: 'UA-10876-1',
        account_holder: 'James L. Silver',
        balance: '$50.5',
      }
    ]);
  
    // choose to only return certain properties you wish to validate against
    let mappedProperties = table.getData().map((d) => {
      return {
        account_id: d.account_id,
        balance: d.balance,
      };
    });

    expect(mappedProperties).to.deep.eq([
      {
        account_id: 'UA-11876-3',
        balance: '$33',
      },
      {
        account_id: 'UA-10876-1',
        balance: '$50.5',
      },
      {
        account_id: 'UA-10346-1',
        balance: '$-22.98',
      },
    ]);   
  });

Date assertions

In order to assert date columns, you will need to specify the column index for the data to be interpreted as a date data type. The underlying data will be converted to unix epoch format to allow sort assertions.

cy.get('#example')
  .scrapeTable({ dateColumns: [0] })
  .then((table) => {
    // ensure the 'created' AKA 'Created' date field is sorted by ascending
    expect(
      table.isPropertySorted(['created'], ['asc']),
      'created sorted in asc order'
    ).to.be.false;
  });

Use fixture as baseline

Take advantage to the fixtures within Cypress when dealing with large datasets.

cy.get('#example')
  .scrapeTable()
  .then((table) => {
    cy.fixture('expected_table_values').then((expectedTableData) => {
      expect(table.getData()).to.deep.eq(expectedTableData);
    });
  });

Infer data types and aggregate columns

Provide the index of numeric columns (starting at zero), in the example above. The balance column index is 3, when this value is supplied to the plug-in, we are able to validate the total sum of the column against an expected value.

cy.get('#example')
  .scrapeTable({ decimalColumns: [3] })
  .then((table) => {
    expect(table.sumOfColumn('balance', 2)).to.eq(60.52);
  });

Configuration

Property	Default Value	Type	Purpose
exportFileName		string	The name of the exported file. Both the [exportFileName] and the [exportFilePath] properties must be specified in order to export the scrapted data to a file.
exportFilePath		string	Where the exported files are to be saved.
includeTimestamp	false	boolean	Add a unique timestamp to the generated file
propertyNameConvention	snakeCase	string	Controls the naming convention of the resultant data table json representation. The value can either be 'snakeCase' or 'camleCase'
removeAllNewlineCharacters	false	boolean	Instructs the plugin to remove all new line characters from the table cell values
applyDataTypeConversion	false	boolean	Converts the column values to integers for the columns specified in the [decimalColumns] property.
decimalColumns	[]	array	This configuration is applied only when the applyDataTypeConversion flag is set to true. Expects an array of intergers, these numbers map to the column index (starting at zero) of each of columns in the table. The values contained in the columns will be converted to integers.

Best Practice

This plugin doesn't wait for data to appear in your table. This can be achieved by adding guards into your code prior to calling scrapeTable.

Examples:

Use your application's api to sync against

    cy.intercept('api/endpoint/fetches/data').as('myTableData')
    cy.visit('https://myapp.com')
    cy.wait('@myTableData')
    cy.get('#simpleTable')
      .scrapeTable()
      .then((table) => {})

Add a have.length.above assertion after your cy.get call

    cy.get('#simpleTable')
      .should('have.length.above', 0)
      .scrapeTable()
      .then((table) => {})

Contributions

Contributionsare more than welcome. Go nuts.

Credits

Solution in this stackoverflow question @metarmask

License

MIT