I was just looking if there is an easy way to get column headers from table using data scraping. After reading this subject it looks like there is no such built in feature. I think most of tables I used data scraping on did not use th tag, so I had to use while loop to change column names, then delete the first row. Exactly the same approach as @andrzej.kniola described.
you could solve it I have the same problem, please help me
I recently stumbled upon this problem as well. Working around this is doable, however it would be very beneficial to have at least some documentation regarding the <extract-table> XML of the Extract Metadata input property.
For example, when extracting the data table, my auto-generated Extract Metadata property looked as follows:
<extract-table get_columns_name='1' get_empty_columns='1' columns_name_source='Longest' />
The current documentation does not offer any explanation regarding the attributes of this XML and which of them we can add/remove/modify in order to achieve the required results.
I see that you can manually define the column header names, however a more generic way of defining what to extract as the column headers would be great.
To everyone else struggling with this problem: if you can assure that the parts of the table which need to be extracted are visible on the screen, you can also make use of the CV Extract Table activity. CV activities can come in handy very often, but it still would be more reliable to extract the table using the actual data of the application.
prashant1603765
ashokkarale
imad-pro
Matias_Clemente.Arg
Shah_Rukh