How to use 'Structure data extraction action' command in Recorder Package

​@SILPA P R I made a video on the topic. Find it here:

https://upskill.automationanywhere.com/getting-started-with-structured-data-extraction

The same can be found for nearly any topic in the Automation Anywhere universe. Use this page to search:

https://upskill.automationanywhere.com/

Thank you for sharing this documentation. It means a lot.

But in my scenario i have around 13000 records which spread across in multiple pages. Like in one page around 100 records. Using the method shown in the vedio it only captures 8 records. How can i move to next page and also capture entire data in the single page instead of capturing few records.

​@SILPA P R The video shows either 21 or 25 records. It captures multiple screens of data on a progressively-loading page.

Now, if you have separate pages (e.g., a page counter with a “Next” button), then you won’t use the structured data extraction option. You will use something like this:

Loop… while object exists (e.g., your Next button)
    Grab the data using the Recorder package
    Put the data into an output file like Excel
    Click on the Next button using the Recorder package
End loop
Grab the last page of data
Put the data into the output file

​@SILPA P R  I would suggest to look this approach in your scenairo 

The best-performing solution is usually:

1. Inspect the Network tab.
2. Identify the API used by the grid/table.
3. Use Automation Anywhere's REST Web Service package instead of Recorder.
4. Parse the JSON response.
5. Loop through API pages until all records are collected.

​@rbkadiyam That approach works well for applications that use JavaScript fetch calls to populate the page. That client-side REST initiation is how Automation Anywhere’s Control Rooms work.

For other pages that are generated server side, that method won’t work. Consider this test page:

https://developer.automationanywhere.com/pagelist.php

Since the pages are generated server side, you cannot use REST calls to get the data.

Thank you ​@rbkadiyam and ​@Aaron.Gleason for your valuable suggestion. But what i understood is that the records which are in the screen in not in table format. It is in div tags and I don’t think these methods will work. And also i understood is while scrolling down to extract next set data domx path is changing again, which is harder to extract. is there any other way to extract the data at once in a page instead of capturing single item one by one.

​@SILPA P R Is there a way you can give us a redacted snippet of code so we know what we’re working with? Maybe a redacted screen shot?

​@Aaron.Gleason , Did you mean the source code of the page. If it is please find the attached file.

​@Aaron.Gleason , if you are ok can we connect and see if there is any solution for this. 

This is my email ID silpa.pr@albahar.com. We can have a teams meeting. This would be really helpful for me.

​@SILPA P R Thank you for the source code. It does appear things are in <div> and <span> tags, which isn’t all that unusual. What it does mean is that you may have to get each individual piece of data using variable DOMXPaths in the Recorder: Capture actions.

I might recommend watching this video for assistance:

https://upskill.automationanywhere.com/accelerating-with-the-recorder-package-repeating-web-objects

Since I am not on the Professional Services team, I cannot have a direct meeting with you. I would recommend contacting your internal RPA support team first, then if things aren’t going well, they can contact Professional Services for further assistance.