<tool id="aws_sra" name="NCBI SRA AWS Fetch" version="0.4.0+gt+0.2" profile="23.0">
    <description>Fetches one or more SRA runs from AWS S3 and converts them to FASTQ</description>

    <requirements>
        <requirement type="package" version="2.34.8">awscli</requirement>
        <requirement type="package" version="3.2.1">sra-tools</requirement>
        <requirement type="package" version="2.8">pigz</requirement>
    </requirements>

    <version_command>fasterq-dump --version</version_command>

    <command detect_errors="aggressive"><![CDATA[
        ## Create a clean list of accessions from the user input
        echo "$accession" | sed 's/,/\n/g; s/ \+/\n/g' | grep . > accessions.txt &&

        ## Loop over each clean accession
        for acc in $(cat accessions.txt);
        do
            echo "Processing accession: $acc" &&

            ## 1. Create unique directories for this accession
            mkdir -p sra_cache_${acc} fastq_out_${acc} &&

            ## 2. Download the file from S3 using aws s3 cp
            aws s3 cp --no-sign-request "s3://sra-pub-run-odp/sra/${acc}/${acc}" ./sra_cache_${acc}/ &&

            ## 3. Convert with fasterq-dump
            fasterq-dump --outdir ./fastq_out_${acc} --temp . --threads \${GALAXY_SLOTS:-4} --split-files ./sra_cache_${acc}/${acc} &&

            ## 4. Compress with pigz
            pigz -p \${GALAXY_SLOTS:-4} ./fastq_out_${acc}/*.fastq &&

            ## 5. Move outputs for collection discovery
            #if str($layout) == 'paired'
                mv ./fastq_out_${acc}/${acc}_1.fastq.gz '$output_r1.files_path/${acc}_1.fastq.gz' &&
                mv ./fastq_out_${acc}/${acc}_2.fastq.gz '$output_r2.files_path/${acc}_2.fastq.gz'
            #else
                mv ./fastq_out_${acc}/${acc}.fastq.gz '$output_r1.files_path/${acc}.fastq.gz'
            #end if &&

            ## 6. Clean up
            rm -rf sra_cache_${acc} fastq_out_${acc}
        done
    ]]></command>

    <inputs>
        <param name="accession" type="text" multiple="true" label="SRA Accession(s)" help="Provide one or more accession numbers (separated by commas, spaces, or newlines). This field accepts a dataset list of accessions in a workflow."/>
        <param name="layout" type="select" label="Read layout" help="This setting is applied to all accessions.">
            <option value="paired" selected="true">Paired-end (R1 + R2)</option>
            <option value="single">Single-end</option>
        </param>
    </inputs>

    <outputs>
        <collection name="output_r1" type="list" label="FASTQ Reads (R1) for ${accession}">
            <discover_datasets pattern="(?P&lt;designation&gt;.+)_1\.fastq\.gz" format="fastqsanger.gz" />
        </collection>
        <collection name="output_r2" type="list" label="FASTQ Reads (R2) for ${accession}">
            <discover_datasets pattern="(?P&lt;designation&gt;.+)_2\.fastq\.gz" format="fastqsanger.gz" />
            <filter>layout == 'paired'</filter>
        </collection>
    </outputs>

    <tests>
        <test expect_num_outputs="2">
            <param name="accession" value="SRR13333333"/>
            <param name="layout" value="paired"/>
            <output_collection name="output_r1" type="list" count="1">
                <element name="SRR13333333_1" ftype="fastqsanger.gz" has_text="@SRR13333333"/>
            </output_collection>
            <output_collection name="output_r2" type="list" count="1">
                <element name="SRR13333333_2" ftype="fastqsanger.gz" has_text="@SRR13333333"/>
            </output_collection>
        </test>
    </tests>

    <help><![CDATA[
**NCBI SRA AWS Fetch**

Fetches one or more SRA runs from the public `sra-pub-run-odp` S3 bucket and converts them to gzip-compressed FASTQ using `fasterq-dump`. This tool uses `aws s3 cp` for direct downloads within the AWS environment.
    ]]></help>

    <citations>
        <citation type="bibtex">
@misc{ncbi_sra_aws,
  title        = {{NCBI} {SRA} on {AWS} Open Data},
  author       = {{National Center for Biotechnology Information}},
  howpublished = {\\url{https://registry.opendata.aws/ncbi-sra/}},
  note         = {Accessed via AWS S3 without credentials}
}
        </citation>
        <citation type="bibtex">
@article{sra_toolkit,
  title   = {The {NCBI} {SRA} and portable data in biology},
  author  = {Leinonen, Rasko and Sugawara, Hideaki and Shumway, Martin and
             {International Nucleotide Sequence Database Collaboration}},
  journal = {Nucleic Acids Research},
  volume  = {39},
  number  = {suppl\\\_1},
  pages   = {D19--D21},
  year    = {2011},
  doi     = {10.1093/nar/gkq1019}
}
        </citation>
    </citations>
</tool>