<tool id="hcluster_sg_parser" name="hcluster_sg parser" version="0.2.0">
    <description>converts hcluster_sg 3-column output into lists of IDs</description>
    <command detect_errors="exit_code">
<![CDATA[
python '$__tool_directory__/hcluster_sg_parser.py' '$inputFile'
#if str($min_elems)
    -m $min_elems
#end if
#if str($max_elems)
    -M $max_elems
#end if
'$discarded'
]]>
    </command>
    <inputs>
        <param name="inputFile" type="data" format="tabular" label="hcluster output file" help="Tab-separated format: cluster_id [other optional columns] cluster-size cluster-members" />
        <param name="min_elems" type="integer" value="" min="0" optional="true" label="Minimum number of cluster elements" />
        <param name="max_elems" type="integer" value="" min="2" optional="true" label="Maximum number of cluster elements" />
    </inputs>
    <outputs>
        <collection name="ids_lists" type="list" label="${tool.name} on ${on_string}">
            <discover_datasets pattern="(?P&lt;designation&gt;.+)_output\.txt" ext="txt" />
        </collection>
        <data name="discarded" format="txt" label="${tool.name} on ${on_string}: discarded IDs" />
    </outputs>
    <tests>
        <test>
            <param name="inputFile" ftype="tabular" value="hcluster_sg.tabular" />
            <output_collection name="ids_lists" type="list">
                <element name="0" file="0_output.txt" ftype="txt" />
                <element name="1" file="1_output.txt" ftype="txt" />
                <element name="2" file="2_output.txt" ftype="txt" />
                <element name="3" file="3_output.txt" ftype="txt" />
            </output_collection>
            <output name="discarded" file="empty.txt" />
        </test>
        <test>
            <param name="inputFile" ftype="tabular" value="hcluster_sg.tabular" />
            <param name="min_elems" value="6" />
            <output_collection name="ids_lists" type="list">
                <element name="0" file="0_output.txt" ftype="txt" />
                <element name="1" file="1_output.txt" ftype="txt" />
            </output_collection>
            <output name="discarded" file="discarded.txt" />
        </test>
        <test>
            <param name="inputFile" ftype="tabular" value="hcluster_sg1.txt" />
            <output_collection name="ids_lists" type="list">
                <element name="0" file="hcluster_sg1_0_output.txt" ftype="txt" />
                <element name="1" file="hcluster_sg1_1_output.txt" ftype="txt" />
                <element name="2" file="hcluster_sg1_2_output.txt" ftype="txt" />
                <element name="3" file="hcluster_sg1_3_output.txt" ftype="txt" />
                <element name="4" file="hcluster_sg1_4_output.txt" ftype="txt" />
                <element name="5" file="hcluster_sg1_5_output.txt" ftype="txt" />
                <element name="6" file="hcluster_sg1_6_output.txt" ftype="txt" />
                <element name="7" file="hcluster_sg1_7_output.txt" ftype="txt" />
                <element name="8" file="hcluster_sg1_8_output.txt" ftype="txt" />
            </output_collection>
            <output name="discarded" file="empty.txt" />
        </test>
    </tests>
    <help>
<![CDATA[
A simple parser to convert the hcluster_sg output into lists of IDs, one list for each cluster.

When a minimum and/or maximum number of cluster elements are specified, the IDs contained in the filtered-out clusters are collected in the "discarded IDS" output dataset.
]]>
    </help>
    <citations>
    </citations>
</tool>