Hi,

What I am trying is,

A)     to write an exec_after_process for tool1(Scatterer tool). This tool1 gives multiple outputs(in the form of files names generated by Galaxy).

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

scatterer_postprocess.py

def exec_after_process(app, inp_data, out_data, param_dict, tool, stdout, stderr):

        print "Collected Dataset[Primary] is %s"% param_dict.get('__collected_datasets__')['primary']['output1'];

Output:

Collected Dataset[Primary] is {'output2': <galaxy.model.HistoryDatasetAssociation object at 0xb0648ac>, 'output3': <galaxy.model.HistoryDatasetAssociation object at 0xb15170c>, 'output4': <galaxy.model.HistoryDatasetAssociation object at 0xaf499ac>}

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

I could see these values successfully. Next,

B)      I wanted to view these multiple values in the exec_prejob hook of tool2(Gatherer)

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

gatherer_prejob.py
def exec_before_job(inp_data, out_data, param_dict, tool):

        print "******************* collected dataset is %s"% param_dict.get('__collected_datasets__')['primary']['output1'];

Output:

TypeError: Error in 'Gather' hook 'exec_before_job', original message: exec_before_job() got multiple values for keyword argument 'inp_data'

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

I am not very savy with Python for now. Can someone point me out to any possible resolution to this issue. The reason I am writing these hooks is to read the multiple inputs in tool2 thrown out by tool1.

Thank you for your time.

Regards,

Sonali Amonkar

Hi,

Currently I am building a workflow which splits a file into multiple chunks and then recreates it back into one.

Following the document on https://bitbucket.org/galaxy/galaxy-central/wiki/ToolsMultipleOutput, I wrote the following code.

The following is my Splitter tool:

<tool id="sc" name="sc" version = "1.0.0" force_history_refresh="True">

        <description> My tool 1 </description>

        <requirements/>

                <command>java -cp /Splitter1.jar /Common.jar SplitterModule $input $output1 $output1.id $__new_file_path__</command>

        <inputs>

                 <param format="txt" name="input" type="data" label="reads">

                      <label>Reads</label>

                 </param>

        </inputs>

        <outputs>

                <data format="interval" name="output1" metadata_source="input"/>

        </outputs>

        <help>

        </help>

</tool>

The following is my Integrator tool:

<tool id="gt" name="gt" version = "1.0.0">

        <description> My Tool 2  </description>

        <requirements/>

                <command>

                        java -cp Integrator1.jar:Common.jar IntegratorModule $input $output

                </command>

        <inputs>

                 <param format="interval" name="input" type="data" label="pp">

                      <label>File Name</label>

                 </param>

        </inputs>

        <outputs>

                <data format="text" name="output"/>

        </outputs>

        <help>

                This tool reads the reads.pp.txt file and creates smaller sized batches of it.

        </help>

</tool>

When I try to connect and run the workflow, this is what happens:

1.       The files with the names generated from the parameters are created in the tmp folder temporarily

2.       Then these are removed and converted to .dat files and placed in database/files/000 folder with the galaxy generated numbers file names.(eg. dataset_89.dat)

3.       However, when I check the parameters in the IntegratorModule program, I receive new dataset files generated by Galaxy and NOT the ones which were created and should have been passed from the Splitter module

Is there a format I need to follow when writing the Integrator Module’s input parameters? Any pointer to any document which mentions this will be helpful.

Thank you for your time.

Warm Regards,

Sonali Amonkar