annotate data_reader.xml @ 0:b737d0ed42be draft default tip

Uploaded

<tool id="directory_table_reader" name="Directory Data Reader" version="0.2">
    <description>Reads data from preconfigured directories table.</description>
    <command interpreter="python">
        <![CDATA[
        directory_copier.py  
            --ending .${directory.fields.original_extension} 
            --new_ending .${directory.fields.galaxy_extension} 
            #if $results.required=="data"
                --new_ending .${directory.fields.galaxy_extension} 
                --decompress ${directory.fields.decompress} 
                #if $results.start
                    --start "$results.start"
                #end if      
                #if $results.last
                    --last "$results.last"
                #end if      
                #if $results.regex
                    --regex "$results.regex"
                #end if      
            #end if      
            --path ${directory.fields.path} 
            --list ${listing}
        ]]>
    </command>
    <inputs>
        <param name="directory" type="select" label="Directory to import data from">
            <options from_data_table="directory_data"/>
            <validator type="no_options" message="No Data Directory Setup"/>
        </param>
        <param name="list_name" type="text" size="25" label="output name" value="input data"/>
        <conditional name="results">
            <param name="required" type="select" label="Download data or just directory listing" help="Select type of action required.">
                <option value="data" selected="true">Data and listing of selected type</option>
                <option value="listing">Get listing of selected file types </option>
            </param>
            <when value="data">
                <param name="start" type="text" value="" label="String which must be at the start of each file name" />
                <param name="last" type="text" value="" label="String which must be at the end of the file name (excluding the file type)" />
                <param name="regex" type="text" value="" label="Regex pattern which must somewhere in the file name (excluding the file type)" >
                    <sanitizer>
                        <valid initial="string.printable"/>
                    </sanitizer>
                </param>
            </when>
            <when value="listing"/>
        </conditional>
    </inputs>
    <outputs>
        <data format="txt" name="listing" label="List of files in $list_name">
        </data>
        <!-- Ideally galaxy can get the type based on the file extensions. If so just add the type here -->
        <collection type="list" label="$list_name" name="data_collection">
            <filter>(results['required'] == 'data')</filter>
            <discover_datasets pattern="__designation_and_ext__" directory="output" visible="true" />
        </collection>
    </outputs>
    <tests>
        <test>
            <param name="directory" value="fastq.gz_files_id" />
            <param name="list_name" value="test_files" />
            <param name="results|required" value="listing"/>
            <output name="listing">
                <assert_contents>
                    <has_line line="sample1.fastqsanger" />
                    <has_line line="other.fastqsanger" />
                </assert_contents>
            </output>
         </test>
        <test>
            <param name="directory" value="fastq.gz_files_id" />
            <output name="listing_fastq_gz">
                <assert_contents>
                    <has_line line="sample1.fastqsanger" />
                </assert_contents>
            </output>
            <output_collection name="data_collection" type="list">
                <element name="sample1" ftype="fastqsanger" file="sample1.fastq" />
                <element name="other" ftype="fastqsanger" file="other.fastq" />
           </output_collection>
         </test>
        <test>
            <param name="directory" value="fastq_files_id" />
            <param name="results|start" value="sam" />
            <output name="listing_fastq">
                <assert_contents>
                    <has_line line="sample1.fastq" />
                    <not_has_text text="other.fastq" />
                </assert_contents>
            </output>
            <output_collection name="data_collection" type="list">
                <element name="sample1" ftype="fastq" file="sample1.fastq" />
           </output_collection>
         </test>
        <test>
            <param name="directory" value="fastq_files_id" />
            <param name="results|regex" value="le.?" />
            <output name="listing_fastq">
                <assert_contents>
                    <has_line line="sample1.fastq" />
                    <not_has_text text="other.fastq" />
                </assert_contents>
            </output>
            <output_collection name="data_collection" type="list">
                <element name="sample1" ftype="fastq" file="sample1.fastq" />
           </output_collection>
         </test>
    </tests>

    <help>
<![CDATA[
This tool will lookup files on the Galaxy server machine, including mounted directories.

Only directories and ending combinations set up by a Galaxy admin can be listed or loaded in this way.
These endings are case senitive.

====

The data options will look for all files that have a particular ending in the selected directory.

The tool will return two things.

1. A Dataset collection of all the detected files. (If data requested)

2. A file with the names of all the detected files. These will be sorted in the same order as galaxy builds the dataset collection. 

The files can be filtered by setting a specific start string for the file name. 
Only files that start with this string (case senstive) will be included.

Files can also be filter for the last part before the file extsentions.

Files can also be filtered by a regex pattern.  
Only files that contain the regex string will be included.
This uses the python search funtion so as long as the Regex pattern is found somewhere in file name (excluding extension).

The three filter start, last and regex if supplied work indepently, so only files that pass all supplied test will be included.

Note: Before applying the last and regex test the ending (includig the . just before the ending are removed)

Assuming the directory has:
C01_R1_001.fasta   C01_R2_001.fatsa   C02_R1_001.fasta   C02_R2_002.fatsa

Setting start C01 will return just the C01 files:   C01_R1_001.fasta   C01_R2_002.fatsa

Setting last R1_001 will return the read1 files:   C01_R1_001.fasta   C02_R1_001.fasta

Setting regex R2_00.$ will return the R2 files:   C01_R2_001.fatsa   C02_R2_002.fatsa

As Galaxy detects the file type based on the extension this tool will change the exstension as setup by the admin.

This tool will unzip gz files if requested to by the admin,

]]>
    </help>
    <citations>
    </citations>

</tool>