Thanks Peter,

  Yeah, looking over unified_genotyper.xml, you would probably just
want to replace:

   #for $i, $input_bam in enumerate( $reference_source.input_bams ):
       -d "-I" "${input_bam.input_bam}" "${input_bam.input_bam.ext}"
"gatk_input_${i}"
       #if str( $input_bam.input_bam.metadata.bam_index ) != "None":
           -d "" "${input_bam.input_bam.metadata.bam_index}"
"bam_index" "gatk_input_${i}" ##hardcode galaxy ext type as bam_index
       #end if
   #end for

with

   #for $i, $input_bam in enumerate( $reference_source.input_bams ):
       -d "-I" "${input_bam}" "${input_bam.ext}" "gatk_input_${i}"
       #if str( $input_bam.metadata.bam_index ) != "None":
           -d "" "${input_bam.metadata.bam_index}" "bam_index"
"gatk_input_${i}" ##hardcode galaxy ext type as bam_index
       #end if
   #end for

And:

        <repeat name="input_bams" title="BAM file" min="1"
help="-I,--input_file &amp;lt;input_file&amp;gt;">
            <param name="input_bam" type="data" format="bam" label="BAM file">
              <validator type="unspecified_build" />
              <validator type="dataset_metadata_in_data_table"
table_name="gatk_picard_indexes" metadata_name="dbkey"
metadata_column="dbkey" message="Sequences are not currently available
for the specified build." /> <!-- fixme!!! this needs to be a select
-->
            </param>
        </repeat>

with

            <param name="input_bams" type="data" multiple="true"
format="bam" label="BAM file">
              <validator type="unspecified_build" />
              <validator type="dataset_metadata_in_data_table"
table_name="gatk_picard_indexes" metadata_name="dbkey"
metadata_column="dbkey" message="Sequences are not currently available
for the specified build." /> <!-- fixme!!! this needs to be a select
-->
            </param>


And:

        <repeat name="input_bams" title="BAM file" min="1"
help="-I,--input_file &amp;lt;input_file&amp;gt;">
            <param name="input_bam" type="data" format="bam" label="BAM file" >
            </param>
        </repeat>

with

            <param name="input_bams" multiple="true" type="data"
format="bam" label="BAM file" >
            </param>

I have not tested these specific changes, so your millage may vary. I
have no clue if those validators in that second block are going to
work with multiple="true". If you test it out and there is some
problem, please let me know I can try to fix it.

I don't know if the Galaxy team wants to start replacing these blocks
in its tools - this change would break existing workflows built on the
unified genotyper and I am not sure the Galaxy team has any interest
in using these kind of data blocks going forward. They are working out
great for us on Galaxy-P though.

-John


On Tue, Oct 1, 2013 at 7:57 AM, Peter Cock <p.j.a.c...@googlemail.com> wrote:
> On Tue, Oct 1, 2013 at 1:27 PM, Ulf Schaefer <ulf.schae...@phe.gov.uk> wrote:
>> Dear all
>>
>> We frequently find ourselves in situations where a tool needs to be run
>> with a lot of input files. For example, run the GATK UnifiedGenotyper
>> with easily dozens of bam files.
>>
>> Using the <repeat> in this case requires quite a bit of clicking. Is
>> there a more conventient way fo doing this? Maybe similar to the
>> multi-file-select that is possible for workflow inputs?
>>
>> I saw some older discussions on this or similar issues, but I am a bit
>> lost what the current official stable proposed solution for this is.
>>
>> Thanks for your help
>> Ulf
>
> Use <param type="data" multiple="true" ... > instead of using
> <repeat ...><param type="data" ...></repeat>.
>
> I asked John Chilton about this recently via Twitter, and it is
> now on the wiki,
> http://wiki.galaxyproject.org/Admin/Tools/ToolConfigSyntax
>
> Simple example here:
> https://bitbucket.org/galaxyp/galaxyp-toolshed-iquant/src/tip/iquant.xml?at=default
>
> Peter
> ___________________________________________________________
> Please keep all replies on the list by using "reply all"
> in your mail client.  To manage your subscriptions to this
> and other Galaxy lists, please use the interface at:
>   http://lists.bx.psu.edu/
>
> To search Galaxy mailing lists use the unified search at:
>   http://galaxyproject.org/search/mailinglists/
___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:
  http://lists.bx.psu.edu/

To search Galaxy mailing lists use the unified search at:
  http://galaxyproject.org/search/mailinglists/

Reply via email to