Unanswered

Apply_to_each is bypassed

(0) Share

Report

Posted on by MattCB

Hi everyone, I've built a flow that has run successfully a few times. Today I've gone to run the process again and it bypasses my "Apply_to_each" so none of the actions within are processed.

The flow is triggered by a file being added to a SharePoint folder. The data is captured from the file, and then I use the "Apply_to_each" to process the data. There's definitely data there as I have checked the Compose output before the "Apply_to_each".

Does anyone know why this might be happening? The flow "succeeds" but none of the actions actually happen.

Compose is:

split(outputs('Get_file_content')?['body'], decodeUriComponent('%0D%0A'))

Apply_to_each is:

skip(outputs('Compose'),1)

Compose 5:

replace(item(),' ',',')

Compose 2:

split(outputs('Compose_5'),',')

Categories:

Building flows

I have the same question (0)

All responses (11)

Answers (0)

Pstork1 68,697 Most Valuable Professional on at

Like (1)

Report

The results of the run suggest that the apply to each is running on an empty array. What array of values are you running the loop on? I would double check again that the array actually contains data.

Was this reply helpful? Yes No
MattCB 14 on at

Like (0)

Report

Thanks @Pstork1 - I've just updated the post with additional information.

Apply_to_each is:
skip(outputs('Compose'),1)

Compose 5:
replace(item(),' ',',')

Compose 2:
split(outputs('Compose_5'),',')

It has run fine before - I've been able to check the outputs after the "Apply_to_each" is run and you're right, nothing there, but there's definitely outputs from "Compose".

All the variables intialised beforehand are appended to in the Apply to each.

Was this reply helpful? Yes No
Pstork1 68,697 Most Valuable Professional on at

Like (0)

Report

I would put a compose in just prior to the Apply to each set to

skip(outputs('Compose'),1)

I suspect Compose is returning an array of length 1. When you skip the first element you are ending up with an empty array. That suggests the input document isn't finding the %0D%0A (carriage return and linefeed).

Was this reply helpful? Yes No
Pstork1 68,697 Most Valuable Professional on at

Like (0)

Report

I would put a compose in just prior to the Apply to each set to

skip(outputs('Compose'),1)

I suspect Compose is returning an array of length 1. When you skip the first element you are ending up with an empty array. That suggests the input document isn't finding the %0D%0A (carriage return and linefeed).

Was this reply helpful? Yes No
MattCB 14 on at

Like (0)

Report

I've added that in but still the Apply_to_each doesn't fire.

Was this reply helpful? Yes No
MattCB 14 on at

Like (0)

Report

Taking out any skip expression does return the data - although I've used this before and it worked great. Same file type, same headers to the data set, but different result?

Was this reply helpful? Yes No
MattCB 14 on at

Like (0)

Report

@Pstork1 might you know why the input document isn't finding the carriage return and linefeed?

Was this reply helpful? Yes No
Pstork1 68,697 Most Valuable Professional on at

Like (1)

Report

It can depend a lot on the document. I've seen cases where one or the other doesn't show up, or the spacing is off, or the order is backwards. There can be a variety of reasons.

Was this reply helpful? Yes No
MattCB 14 on at

Like (0)

Report

Thank you, this has really helped. Do you know how I might include a check for one or both within the same flow?

Was this reply helpful? Yes No
MattCB 14 on at

Like (0)

Report

I think I know the answer to my own question - parallel branch?

Was this reply helpful? Yes No