Scrap a specific text from html

[lliu_western]

Hello community,

I have a flow that gets the html of a website.

Part of the html out that I got from getting the html is follow:

The "sessionDataKey" is the unique text within the html that stays consistent and closest to the key that I need to scap.

So I have a substring formula as follw:

substring
(
substring(outputs('html'), indexOf(outputs('html'), 'sessionDataKey')),
9,
indexOf(substring(outputs('html'), indexOf(outputs('html'), 'sessionDataKey')), '"')
)

I counted from the word "sessionDataKey" to the first letter of the key that I need to scrap, there are nine spaces between then, and I want it to stop when it encounter the quotation mark. 

But the flow fails, can anyone sees any obvious mistakes in my substring code?

Thanks in advance 

[Paulie78]

Try:

split(substring(outputs('html'), indexof(outputs('html'), 'name="sessionDataKey"')),'"')[3]

[lliu_western]

@Paulie78 

still not liking it. Can't figure out where it got wrong either. 

[Paulie78]

sounds like it is looking for a step called "html" and you don't have one.

[lliu_western]

@Paulie78 , I did have the step "html" before that. I renamed it again and it worked, I was able to run the flow but it failed with the error below:

Unable to process template language expressions in action 'Get_DataKey_Token' inputs at line '1' and column '7271': 'The template language function 'substring' parameter is out of range: 'start index' must be non-negative integer and should be less than the length of the string. Please see https://aka.ms/logicexpressions#substring for usage details.'.

[Paulie78]

That means it did not find 

name="sessionDataKey"

in the string. Try adding a compose before which just has:

indexof(outputs('html'), 'name="sessionDataKey"')

and see what it comes out with. Sounds like it is producing -1

[lliu_western]

@Paulie78  I got the substring working just fine now, but I got more than I needed in terms of split length but I should be able to figure out from here, thanks!!

[lliu_western]

@Paulie78 

@yashag2255 

this whole image is the return value I got:

by using this formula:

But I only need the highlighted text, how can I get the substring top at the first single quote it encounters? 

[Paulie78 · Accepted solution]

It is because the expression I gave you was searching for a double quote as per original post. This string has single quotes so try this instead:

split(substring(outputs('htmlContent'), indexof(outputs('htmlContent'), 'name=''sessionDataKey''')),'''')[3]

If you get stuck, lose the [3] off the end of the split expression and see what array it outputs. The [3] just selects the 3rd element of the array.

[lliu_western]

@Paulie78  it is still saying invalid expression even if I take out the [3]

[lliu_western]

@Paulie78 

this is what I have currently that gets me exactly what I need. But it is not dynamic enough. I don't want to use "32", I want to use a expression that would stop when it encounters the quotation. 

[Paulie78]

When I tested it, the expression I provided worked perfectly. But it is really hard to test it properly without actual sample data. 

[lliu_western]

@Paulie78  I still couldn't get it work by using split. But I was able to get it work by using Substring again and just use indexof formula to calculate where I want the substring to stop. But thank you for your help, you have been very helpful!