<tool id="dada2_primerCheck" name="dada2: primer check" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description></description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="bio_tools"/>
    <expand macro="requirements"/>
    <expand macro="stdio"/>
    <expand macro="version_command"/>
    <command detect_errors="exit_code"><![CDATA[
Rscript '$dada2_script'
    ]]></command>
    <configfiles>
        <configfile name="dada2_script"><![CDATA[
#import re
library(Biostrings, quietly=T)
library(ShortRead, quietly=T)

FWD <- "$forward_primer"
REV <- "$reverse_primer"

allOrients <- function(primer) {
    # Create all orientations of the input sequence
    dna <- DNAString(primer)  # The Biostrings works w/ DNAString objects rather than character vectors
    orients <- c(Forward = dna, Complement = Biostrings::complement(dna), Reverse = Biostrings::reverse(dna), RevComp = Biostrings::reverseComplement(dna))
    return(sapply(orients, toString))  # Convert back to character vector
}
FWD.orients <- allOrients(FWD)
REV.orients <- allOrients(REV)

primerHits <- function(primer, fn) {
    ## Counts number of reads in which the primer is found
    nhits <- vcountPattern(primer, sread(readFastq(fn)), fixed = FALSE)
    return(sum(nhits > 0))
}

df <- NULL;
#for $i, $read in enumerate($paired_cond.reads):
    #set elid = re.sub('[^\w\-\.]', '_', str($read.element_identifier))
    #if $paired_cond.paired_select == "single"
        #set fwd_reads = $read
    #else
        #set fwd_reads = $read.forward
        #set rev_reads = $read.reverse
    #end if
    df <- rbind(df, c('$elid', 'FWD', 'FWD', sapply(FWD.orients, primerHits, fn = '$fwd_reads')))
    df <- rbind(df, c('$elid', 'REV', 'FWD', sapply(REV.orients, primerHits, fn = '$fwd_reads')))
    #if $paired_cond.paired_select != "single"
        df <- rbind(df, c('$elid', 'FWD', 'REV', sapply(FWD.orients, primerHits, fn = '$rev_reads')))
        df <- rbind(df, c('$elid', 'REV', 'REV', sapply(REV.orients, primerHits, fn = '$rev_reads')))
    #end if
#end for
colnames(df) <- c('Sample', 'Primer', 'ReadDir', 'Sequence', 'Complement', 'Reverse', 'RevComp')
write.table(df, "$out", quote=F, sep="\t", row.names = F, col.names = T)
    ]]></configfile>
    </configfiles>
    <inputs>
        <expand macro="fastq_input" multiple="True" collection_type="list:paired" argument_fwd="fl" argument_rev="fl"/>
        <param name="forward_primer" type="text" label="Forward primer sequence">
            <validator type="empty_field" message="You need to specify a forward primer sequence"/>
        </param>
        <param name="reverse_primer" type="text" label="Reverse primer sequence">
            <validator type="empty_field" message="You need to specify a reverse primer sequence"/>
        </param>
    </inputs>
    <outputs>
        <data name="out" format="tabular"/>
    </outputs>
    <tests>
        <!-- paired data in paired collection -->
        <test expect_num_outputs="1">
            <conditional name="paired_cond">
                <param name="paired_select" value="paired"/>
                <param name="reads">
                    <collection type="list:paired">
                        <element name="F3D0_S188_L001">
                            <collection type="paired">
                                <element name="forward" value="F3D0_S188_L001_R1_001.fastq.gz" ftype="fastqsanger.gz"/>
                                <element name="reverse" value="F3D0_S188_L001_R2_001.fastq.gz" ftype="fastqsanger.gz"/>
                            </collection>
                        </element>
                        <element name="F3D141_S207_L001">
                            <collection type="paired">
                                <element name="forward" value="F3D141_S207_L001_R1_001.fastq.gz" ftype="fastqsanger.gz"/>
                                <element name="reverse" value="F3D141_S207_L001_R2_001.fastq.gz" ftype="fastqsanger.gz"/>
                            </collection>
                        </element>
                    </collection>
                </param>
            </conditional>

            <param name="forward_primer" value="ACCTGCGGARGGATCA"/>
            <param name="reverse_primer" value="GAGATCCRTTGYTRAAAGTT"/>
            <output name="out">
                <assert_contents>
                    <has_n_lines n="9"/>
                    <has_n_columns n="7"/>
                </assert_contents>
            </output>
        </test>
        <!-- single end data -->
        <test expect_num_outputs="1">
            <conditional name="paired_cond">
                <param name="paired_select" value="single"/>
                <param name="reads" value="F3D0_S188_L001_R1_001.fastq.gz" ftype="fastqsanger.gz"/>
            </conditional>
            <param name="forward_primer" value="ACCTGCGGARGGATCA"/>
            <param name="reverse_primer" value="GAGATCCRTTGYTRAAAGTT"/>
            <output name="out">
                <assert_contents>
                    <has_n_lines n="3"/>
                    <has_n_columns n="7"/>
                </assert_contents>
            </output>
        </test>
    </tests>

    <help><![CDATA[
Description
...........

Simple check for primer sequences in sequencing data. The tool counts the number
of occurrences of the primer sequence, its complement, the reverse and the
reverse complement.

See also: https://benjjneb.github.io/dada2/ITS_workflow.html#identify-primers

Usage
.....

**Input** FASTQ datasets and forward and reverse primers

**Output** a table listing the counts of the different occurrences in the read files.


@HELP_OVERVIEW@
    ]]></help>
    <expand macro="citations"/>
</tool>