Channel.fromFilePairs issue in DSL2 (process has been already used)

[ferygood]

Hello everyone,
I am starting to learn Nextflow using DSL2 and have problem using .Channel.fromFilePairs() loading {1,2}.fastq files to my process. I hope you can give me some suggestions on this since I cannot figure it out by myself reading the documentation and examples...

#!/usr/bin/env nextflow
nextflow.enable.dsl =2

reads_ch = Channel.fromFilePairs( '/home/data/*_R{1,2}.fastq' )

process foo {
    input:
        tuple val(sample_id) file(reads)
    script:
        """
        echo $sample_id
        echo ${reads[0]}
        echo ${reads[1]}
        """

workflow {
    reads_ch.view()
    foo(reads_ch)
}

If I only run with reads_ch.view() in my workflow, my terminal will print out the tuple structure of my fastq files correctly. However, if I use the process with channel foo(reads_ch), my terminal will show the error below and fail to print out the sample_id, first and second fastq files.

Process 'foo' has been already used -- If you need to reuse the same component include it with a different name or include in a different workflow content

I want to make sure I can access the file name before going for further analysis. However, I was stuck for this issue for a while. Any suggestions or solutions are appreciated. Many thanks!

[ferygood]

I realize I have made an obvious mistake in the input field. I missed a comma between val() and file() using the tuple input qualifier.
It should be:

input:
    tuple val(sample_id), file(reads)

I don't know why missing a comma will produce the "Process 'foo' has been already used" error. However, adding the comma fix my problem. Thanks to anyone who might have started reading my question. I also hope my post can help others who encounter the same situation.