Core actions

Core actions in Rewst are your gateway to the platform's vast array of intrinsic functionalities. These actions, usable right out of the box, offer features ranging from ad-hoc HTTP requests to document parsing.

Core actions are used in the same way as regular actions within workflows. They are selected from the list of available actions, configured based on their parameters, and then added to the workflow at the appropriate place.

Expand each of the action types below to see their individual information. Actions are listed alphabetically in both this doc and the Rewst platform.

Await Webhook Request action

Waits for a request to a created one-off webhook. Once a request is received, the workflow continues.

• Parameters: This action only requires the ID of the webhook created from the Create Webhook action.

• Output: The output includes the HTTP method, query params, headers, JSON or form/multipart data in the body of the request, and the timestamp when the request was received.

Compare Strings Set Ratio action

This action calculates the ratio of how many words from the smaller string are found in the larger string. It's designed for measuring string similarity based on word overlap rather than character-by-character comparison. This action is particularly useful when you need to determine how similar two text strings are based on shared words, making it ideal for scenarios where exact string matching is too strict but you still need to measure content similarity.

Use Cases:

• Fuzzy matching between text strings

• Duplicate detection in datasets

• Content similarity analysis

• Partial string matching scenarios

Parameters:

• original_string (required): The original string to compare

• comparison_string (required): The comparison string that is being compared to

Output: Returns an integer representing the ratio of word overlap between the two strings. Higher values indicate greater similarity.

Confirmation Email action

Send a confirmation email with reply options to a specified recipient. This action pauses the workflow and places it in an Awaiting-User-Input state. The workflow will not proceed until the confirmation email is interacted with via buttons, or the task times out. You can configure task timeout on the Advanced tab of the action. Task time out means that the action fails. Note that this setup means that buttons are required for the workflow to proceed.

• Parameters: This action requires the recipient's email address (to), the subject of the email (subject), the title of the email (title), and the message body (message). It also offers user interaction buttons (buttons) and has the option to render markdown as HTML (render_markdown).

• Output: If the action is correctly executed, a confirmation email will be sent. Output variable inquiry_result is an output of the task and can be used to route the workflow in a specific path. Inquiry_result 's value is that of the button clicked, and is configured as a string value on the action itself in the workflow builder.

• Confirmation email class and confirmation examples

  • Button classes:

    • Primary; primary

    • Default; default

    • Danger; danger

  • Title class:

    • Title; title

  • Customization options

    • Custom Button CSS - control the CSS styling for the buttons.

    • Custom footer - overwrite the default Rewst footer or remove the footer entirely.

    • Custom Title CSS - customize the title CSS.

    • Logo Link - customize the hyperlink for the logo image.

    • Logo URI:

      • This will take the link to the image itself.

      • TIP: This is essentially providing a value for the src param of a img tag. You could also provide a base64 encoded value.

• Example custom button CSS

a,

a:visited,

a:hover,

a:active {

color: inherit !important;

}

table, td, div, h1, p {

font-family: "Times New Roman", Times, serif;

}

.primary {

background: #4d7c0f;

}

.default {

background: #365314;

}

.danger {

background: #dc2626;

}

• Example custom footer

  <tr>
  
  <td
  
  style="text-align:center;padding:30px 0 30px 0;background-color:#1a2e05;color:#ffffff;">
  
  <p style="display:inline-block;margin-right:10px;">&copy; The Dougnut {% now 'local', '%Y' %}</p>
  
  <p style="display:inline-block;margin-right:10px;"><a
  
  href="https://rewst.io/terms-of-service/">Order</a></p>
  
  <p style="display:inline-block;margin-right:10px;"><a href="https://rewst.io/privacy-policy/">Contact Us</a></p>
  
    
  
  <p style="display:inline-block;margin-right:10px;color:#84cc16"><a href="mailto:[email protected]">Need
  
  Delivery?</a></p>
  
    
  
  <p style="display:inline-block;margin:0px 4px 0 4px;vertical-align:middle;"><a href="https://www.linkedin.com/company/rewst"><img
  
  src="https://rewst-splash.s3.us-east-2.amazonaws.com/linkedin.png" width="24" height="24"
  
  alt="linkedin"></a></p>
  
    
  
  <p style="display:inline-block;margin:0px 4px 0 4px;vertical-align:middle;"><a href="https://twitter.com/rewst_dot_io"><img
  
  src="https://rewst-splash.s3.us-east-2.amazonaws.com/twitter.png" width="24" height="24" alt="twitter"></a>
  
  </p>
  
    
  
  </td>
  
  </tr>

• Example custom title CSS

.title {

font-size: 20px;

line-height: 16px;

font-weight: 700;

font-style: normal;

color: #0c0a09;

text-decoration: none;

letter-spacing: 0px;

padding: 25px 30px 0 30px;

}

• Example logo link

https://google.com

• Example logo URI

https://www.svgrepo.com/download/533811/donuts-cake.svg

Create Pending Task action

This action creates a pending task that pauses workflow execution and requires manual intervention through the Rewst UI. It's designed for scenarios where human approval, confirmation, or decision-making is needed within an automated workflow.

Use Cases:

• Manual approval workflows

• User confirmation before critical actions

• Decision points requiring human judgment

• Interactive workflow processes

• Quality control checkpoints

Parameters:

• message (required): The message to display in the pending task that explains what action is needed from the user

• buttons (required): An array of button objects that users can click to respond. Each button has:

  • label: The text displayed on the button

  • style: Visual style (default, primary, danger)

  • value: The value returned when the button is clicked

Output: The workflow pauses at this task until a user interacts with it via the Rewst UI. When a button is clicked, the action returns the corresponding button value, allowing the workflow to continue based on the user's choice.

This action is essential for creating interactive workflows that require human oversight or decision-making at specific points in the automation process.

Create Webhook action

Allows for the creation of a one-off webhook for which can then be used by the Await Webhook action.

Parameters: This action requires the methods allowed to access the webhook, the response status, response headers, response body, and an expiration timeout.

Output: The output of this action is the webhook ID and the full URL of the webhook.

Debug action

The Debug action is a utility feature in our workflow system, specifically designed to assist with debugging and logging purposes. This action can help in understanding the flow of data within your workflows, troubleshoot problems, and generally help you understand what's happening at a certain point in the workflow execution. It logs the input parameters it receives and returns the same as its output.

Use cases

You might want to use the Debug action in the following scenarios:

• When developing workflows, to see how data is flowing between tasks and actions.

• If you're troubleshooting an issue, to inspect the data that's being passed around.

• When you want to log specific information for auditing or reporting purposes.

Input Parameters

The Debug action accepts the following parameters:

• text: This is a general-purpose text field that will be logged and returned by the Debug action.

• template: This field takes a reference to a template in your environment that will be rendered and used as part of the action's input. The system will replace any variables in the template with its actual value at the time of template rendering.

Example Usage

Let's say we have a template named "Greeting Message" with content # Hey there {{ CTX.name }}.

We can use this template in the Debug action with the following parameters:

text: Testing Debug Action
template: Greeting Message

Assuming CTX.name is set to Rewsty, the rendered template would be # Hey there Rewsty.

The Debug action will log these parameters and also return them as its output. The results of the action on the workflow results page would look like this:

Output:

{
  "template": "# Hey there Rewsty",
  "text": "Testing Debug Action"
}

This tells us that CTX.name was set toRewsty, and the text provided with this action was Testing Debug Action. Using this, you can better understand the state of your workflow at the point this Debug action was executed.

DNS Query action

Queries a nameserver for DNS records associated with a given URL.

• Parameters: The URL to query the Nameserver for, the field to query from the nameserver, timeout for the DNS Query (optional, default 60), and the nameserver to use for the query (several options available including Google, Cloudflare, OpenDNS).

• Output: The specified DNS records associated with the given URL from the queried nameserver.

Generate Password V2 action

An upgrade from the deprecated password generation action. It crafts a cryptographically secure password with user-specified values. (This is recommended for use over the deprecated Password Action due to its upgraded structure.)

• Parameters: length, minimum counts of numeric and capital letter characters, and optional punctuation characters.

• Output: the generated password is presented under the "password" key in the output.

HTTP Request action

Performs an HTTP request to a specified URL, supporting a variety of methods, body content types, and configurations. This is useful for interacting with APIs or other web services within a workflow, or for performing any other tasks that involve HTTP requests.

Parameters

• URL: The URL to which the HTTP request is sent.

• Request Method: The HTTP method to use for the request. You can select from the dropdown options (HEAD, GET, POST, PUT, DELETE, OPTIONS, TRACE, PATCH, PURGE).

• Auth Username: The username for basic HTTP Authentication, if needed.

• Auth Password: The password for basic HTTP Authentication, if needed. (This parameter is secret to ensure security.)

• Allow Redirects: Specifies whether the HTTP request will follow redirects. By default, it's set to true.

• Body: The body to send with the request. This parameter is not required if JSON or Files is provided.

• JSON: The JSON body to send with the request. This field is not required if Body or Files is provided.

• Files: Here, you can add files to be uploaded with the HTTP request using multipart/form-data. Each file requires the following information:

  • Field Name: The name of the form field (not the filename).

  • File Name: The name of the file.

  • File Contents: Contents of the file to upload.

  • File URL: A publicly-accessible URL to the file contents to upload.

  • Content Type: The MIME type of the file to include in the multipart field.

• Cookies: Input the cookies to send with the request. You can add more than one cookie by clicking on the + icon.

• Headers: Specify the custom HTTP headers to be sent with the request. You can add more than one header by clicking on the + icon.

• Params: Enter the query parameters to be used with the HTTP request. You can add more than one parameter by clicking on the + icon.

• Timeout: Enter the timeout for the HTTP request in seconds. The default value is 5.

• Require Success Status: If you check this box, the task will fail if a non-2xx HTTP status code is returned. This is useful for identifying and handling HTTP errors during the task's execution.

Output: The action returns the content returned by the server in response to the HTTP request. This could be a success message, a failure message, a data object, or any other content that the server sends as a response.

Note: If you need more advanced security around calling particular endpoints, you can also use a Custom Integration in addition to this HTTP Request action. This allows for enhanced security and customization when interacting with your external APIs.

Delay Workflow For Period action

Pauses the workflow for a specified duration.

• Parameters: The number of days, hours, minutes or seconds to delay the workflow.

• Output: No specific output, the workflow resumes after the specified delay.

Delay Workflow Until Date/Time action

Pauses the workflow until a specified date and time.

• Parameters: The date and time when the workflow should resume.

• Output: No specific output, the workflow resumes at the specified date and time.

Mock action

The Mock action is designed to provide you with the capability to simulate the result of a not yet implemented action. This is particularly useful during workflow development and testing phases as it allows you to simulate responses from services that are not yet available or are impractical to call during the development process.

Use cases

The Mock action comes in handy in scenarios such as:

• When designing new workflows where certain steps are not fully implemented.

• In testing stages, to simulate conditions without making actual calls to the services.

• To create controlled conditions in your workflow for troubleshooting issues.

Input Parameters

The Mock action accepts the following parameter:

• Mock Result: This parameter should contain the key / value pairs that you want to be returned by this action. You can press the + to add as many objects as necessary.

Example Usage

You are developing a workflow that is expected to interact with a service which is not yet implemented. You know the expected format of the response, and you want to build and test your workflow logic based on that response.

Let's say you know the response will look something like this:

{
  "name": "Rewsty",
  "valid": true,
  "message": "Successfully completed the task."
}

You can simulate this response using the Mock action as follows:

mock_result:
  name: {{ CTX.name }}
  valid: true
  message: Successfully completed the task.

While using the Mock action, the values can be literal Jinja expressions like {{ CTX.name }}above. The action will return this exact input structure wrapped inside a data object, and the Jinja expressions will not be evaluated but returned as is. The result of the action on the workflow results page would look like:

Output:

{
  "data": {
    "name": "{{ CTX.name }}",
    "valid": true,
    "message": "Successfully completed the task."
  }
}

This can be useful for catching issues early in the development phase such as incorrect Jinja expression usage, understanding how the workflow will handle dynamic data, or verifying that your workflow is properly constructed to handle the expected responses from services. It's a way to ensure that your workflow behaves as expected when it starts receiving actual dynamic data.

No-Operation (Noop) action

A noop action does nothing. It's often used for logic or as a placeholder in the workflow.

• Parameters: None.

• Output: None

Parse HTML action

The Parse HTML action is a versatile tool within Rewst, geared to pinpoint and extract specific elements or data from HTML documents. It leverages the power of BeautifulSoup, a Python library recognized for extracting data from HTML and XML files effectively.

Use cases

This action is particularly beneficial in these situations:

• Data Extraction: capturing specific information from the response of an HTTP request.

• Content Clean-Up: sieving out only the necessary data from complex HTML content.

• Web Scraping: automating the extraction of specific information from various web pages using defined tags, classes, or identifiers.

Action Parameters

The Parse HTML action accepts the following parameters:

• HTML: The HTML content to be parsed. This could be HTML content from a webpage, obtained using the HTTP Request Core Action.

• Class: Optionally finds HTML elements based on their class attribute.

• ID: Optionally searches for HTML elements based on their ID attribute.

• Query: Employs BeautifulSoup filters to specify the operation type:

  • find_all returns all instances of the defined HTML tag.

  • find returns only the first instance of the defined HTML tag.

  • select enables the use of CSS selectors for nested HTML tags.

• String: Optionally searches for specific text within the HTML content.

• Value: Identifies the tag or selector to search for in the HTML content. For example, a would find all anchor (<a>) tags in the HTML content.

Practical Use Case: Extracting Links from 'Hacker News'

This example involves making a GET request to the Hacker News website and parsing the returned HTML to extract all <a> links.

The first step uses the Core HTTP Request action to fetch the HTML content:

publish_result_as: news
request_method: GET
URL: https://news.ycombinator.com

You can then use the Parse HTML action to extract all <a> links. The parameters for this action would be set as follows:

html: {{ CTX.news }}
style: find_all
value: a

Example Workflow Results

Here's an example of how the Parse HTML action's input and output might look like on the workflow results page:

Input from the HTTP Request:

html: {
  cookies: {},
  data: "<html>...</html>",
  headers: {...},
  status_code: 200
},
query: {
  style: "find_all",
  value: "a"
}

Result:

[
  "<a href=\"https://news.ycombinator.com\"><img height=\"18\" src=\"y18.svg\" style=\"border:1px white solid; display:block\" width=\"18\"/></a>",
  "<a href=\"news\">Hacker News</a>",
  "<a href=\"newest\">new</a>",
  "<a href=\"front\">past</a>",
  "<a href=\"newcomments\">comments</a>",
  "<a href=\"ask\">ask</a>",
  "<a href=\"show\">show</a>",
  "<a href=\"jobs\">jobs</a>",
  "<a href=\"submit\">submit</a>",
  "<a href=\"login?goto=news\">login</a>"
]

This result contains all <a> tags found in the HTML content.

To further refine this output, returning only links for externally referenced pages, use the select query style, along with advanced CSS filters:

html: {{ CTX.news }},
style: select
value: .titleline a[href^='https://']

The result is a list of strings containing all <a> tags that meet the newly specified criteria:

[
  "<a href=\"https://arxiv.org/abs/2308.00676\" rel=\"noreferrer\">Electronic Structure of LK-99</a>",
  "<a href=\"https://www.science.org/content/blog-post/room-temperature-superconductor-new-developments\" rel=\"noreferrer\">A room-temperature superconductor? New developments</a>",
  "<a href=\"https://sophiehoulden.com/randomstuff/epitime/?revised\" rel=\"noreferrer\">Epicycle Clock</a>",
  "<a href=\"https://howardism.org/Technical/Emacs/new-window-manager.html\" rel=\"noreferrer\">Emacs is my new window manager</a>",
  "<a href=\"https://ploum.net/2023-08-01-splitting-the-web.html\" rel=\"noreferrer\">Splitting the Web</a>",
  "<a href=\"https://twitter.com/zebulgar/status/1686498517227814912\" rel=\"noreferrer\">Unconfirmed video showing potential LK-99 sample exhibiting the Meissner effect</a>",
  "<a href=\"https://magicloops.dev\" rel=\"noreferrer\">Show HN: Magic Loops – Combine LLMs and code to create simple automations</a>",
  "<a href=\"https://arxiv.org/abs/2307.08378\" rel=\"noreferrer\">eGPU: A 750 MHz Class Soft GPGPU for FPGA</a>"
]

To further understand CSS selectors, you can refer to this w3schools article.

Tip: Parse HTML's functionalities include finding elements by tags (<h1>), class (class_="abc"), text (string="The content"), or id ({"id": "abc"}). When string is the sole argument, only the text is returned, not the whole element, which can help you fine-tune data extraction.

Parse XML action

The Core Parse XML action in Rewst is designed to locate and extract specific elements or data from XML documents. This powerful tool, backed by an efficient Python library, facilitates precise data extraction from XML files, simplifying the process of parsing complex data structures. For additional understanding on XPath expressions, refer to this w3schools article.

Use cases

Consider using the Parse XML action in these scenarios:

• Data Extraction and Content Clean-Up: Capture specific information or filter out necessary data from XML-formatted content. This is particularly useful in processing responses from HTTP requests or handling complex XML documents.

• Web Scraping: Automate the extraction of specific information from various XML sources using defined tags, attributes, or identifiers. It enables you to precisely target the data you need from web resources.

Action parameters

The Core Parse XML action requires the following parameters:

• XML: The XML content that needs to be parsed. This could be XML content from an API response, obtained using the HTTP Request Core Action.

• Attributes: (Optional) Allows you to find XML elements based on their attribute key.

• ID: (Optional) Permits searching for XML elements based on their ID attribute.

• Selector: Determines the operation type. Options include:

  • find: Returns only the first instance of the defined XML tag.

  • find_all: Returns all instances of the defined XML tag.

  • select: Enables the use of XPath expressions for nested XML tags or conditional searches.

• String: (Optional) Allows you to search for specific text within the XML content.

• Value: Specifies the tag or selector to search for in the XML content.

Practical Use Case: Extracting Books from a Bookstore's XML Data

Before diving into parsing XML data, you'll need to fetch the XML file. In this use case, the XML file is fetched from a public URL which contains bookstore data in XML format. The first task in the workflow, called get_books, uses the Core HTTP Request action to fetch this XML content:

Input parameters:

url: http://books.toscrape.com/catalogue/category/books_1/index.html
request_method: GET
publish_result_as: get_books

The result from this task will look something like this:

{
  "cookies": {},
  "data": "<bookstore>...</bookstore>",
  "headers": {...},
  "status_code": 200
}

The data field contains the XML content, which is the input for the Parse XML action. The XML content is passed using the Context (CTX) object as CTX.books.data.

Finding the first book

In this scenario, we are using the find operation to return the first book element in the XML:

Input parameters:

input:
  xml: {{CTX.books.data}}
  selector: find
  value: book

The result from this task will look something like this:

{
  "value": "<book category=\"cooking\"><title lang=\"en\">Everyday Italian</title><author>Giada De Laurentiis</author><year>2005</year><price>30.00</price></book>"
}

The output includes the first book element in the XML content.

Selecting all 'Children' category books

For a more complex operation, we can use the select operation with an XPath expression to extract all book tags where the category attribute is children.

Input parameters:

xml: {{CTX.books.data}}
selector: select
value: book[category='children']

Result:

{
  "value": "<book category=\"children\"><title lang=\"en\">Harry Potter</title><author>J K. Rowling</author><year>2005</year><price>29.99</price></book>"
}

The result includes all book tags where the category attribute is children.

Send Mail action

Allows for the sending of an email.

• Parameters: This action requires a sender prefix (sender), with multiple options available, the recipient's email address (to), the subject of the email (subject), the title of the email (title), and the message body (message). It also has the option to render markdown as HTML (render_markdown). You can also fully control the HTML of the email with (Custom HTML). This can also reference a template using the {{ template(“guid”) }} function.

  • Note that if using the Custom HTML field, the message and title fields will be ignored

  • You cannot upload images to Rewst, so any image will need to be externally referenced

  • Emails will still be sent from the rewst.io domain

• Output: The task doesn't yield an output upon success. It will fail if there are any errors during the process of sending the email.

Send SMS action

Allows you to send a text message to a specified phone number.

• Parameters: This action requires the recipient's phone number (phone_number) and the text message (message) to be sent.

• Output: The output of this action will depend on the implementation details. Usually, it will return a confirmation message or an error message.

Test All Parameter Types action

This action serves as a comprehensive example and testing tool that showcases all the different input field types that can be used in Rewst actions. It's primarily used for:

• UI testing and validation

• Demonstrating parameter field capabilities

• Training and educational purposes

• Testing form rendering and validation

Inputs

The task accepts a wide variety of input parameters representing every field type available in Rewst:

Basic fields:

• simple_string (required) - Basic text input

• optional_string - Optional text field

• string_with_default - Text with default value "default_value"

• simple_integer (required) - Whole number (default: 42)

• simple_number - Decimal number (default: 3.14)

• simple_boolean - Toggle switch (default: false)

• required_boolean (required) - Required toggle

Advanced string fields:

• string_enum (required) - Dropdown with options (option_1, option_2, option_3)

• string_enum_objects - Dropdown with custom labels and descriptions

• radio_selection (required) - Radio button group

• secret_string - Password/masked field

• multiline_text - Large text area

• textarea_content - Code editor area

Date/Time fields:

• date_field - Date picker

• time_field - Time picker

• datetime_field - Date and time picker

Array fields:

• simple_array - Array of strings

• array_with_enum - Multi-select dropdown

• array_of_objects - List of complex objects with name, enabled, and quantity fields

Object fields:

• simple_object - JSON object field

• nested_object - Complex nested object with host, port, SSL settings, and credentials

Special fields:

• number_with_range - Integer between 1-100 (default: 50)

• local_reference_template - Reference to local templates

• remote_reference_example - Reference to remote API options

• remote_reference_with_dependency - Dependent reference field

UI state fields:

• tooltip_field - Field with helpful tooltip

• hidden_field - Hidden from UI

• disabled_field - Cannot be edited

• readonly_field - Read-only display

• deprecated_field - Marked as deprecated

• staff_only_field - Only visible to staff users

Outputs

The action returns a simple object with two properties:

• message (string) - A test result message describing the outcome

• success (boolean) - Whether the test execution was successful

UUID action

Generates a new UUID (Universally Unique Identifier).

• Parameters: UUID type (options include uuid1 and uuid4, defaults to uuid4).

• Output: The generated UUID.