Data Scraping not working in the sequence, but working when I just test it on an already opened web page. Can someone please help me with this issue?

I have had a very similar problem but documented the circumstances more completely, I just didn’t spam tag people and no one responded so I didn’t feel obliged to post my solution. Replacing ExtractDataTable every browser session If this sounds similar to your problem then I may have a solution:

Occasionally UiPath will truncate css selectors after a certain number of characters and replace them with a wildcard ‘*’, rendering them invalid, as there are often other child elements. This appears when looking through the UiExplorer. I can enter a complete path into UiExplorer, in my case that path looks something like:
…;div>div>div>table’ tag=‘TABLE’ />

entering this directly into UiExplorer and clicking the validate button tells us that the selector is valid, but it immediately tries to shorten it to
…div>div>di*’ tag=‘TABLE’ />, which of course is invalid.

Data Extraction makes workarounds particularly tricky, as it is nearly impossible to select the table item that the extraction requires, but it is possible to preserve the full string by running the data extraction wizard, and copying the Selector in the Target portion of Properties. Before running anything, this selector should be complete. I then check after each data scrape when using the automation if the resulting table is empty with a callout in a retry scope, and if it is, I rerun the extraction but pass in what I know is the full selector instead of the one generated by the wizard, which somehow corrupts over time/sessions of use.

Additionally, make sure that you have the appropriate extensions installed, including Java and Silverlight, and that your .NET framework is at least version 4+.

2 Likes