Hello,
I'm new to Galaxy, and I've been asked to tweak some tools so that output file names are related to input file names. For example, an input file of myData.abc might lead to an output file named myData_afterMyTool.xyz.
I think the user is really just interested in the labels showing up in the history panel, so that gives me some flexibility.
In my development environment, I've changed one tool's xml file's output tag to this:
<outputs> <data name="out_file1" format="bed" label="${input1}_afterTool.xyz"/> </outputs>
... and that seems to work, but it's using an internal file name, instead of the label for the input file.
So - does anyone know if there's an easy way to get the label from the input file into the output label, instead of the path/dataset_n.dat file format for this name? If there's a better way to approach this, I'd love any suggestions on that, too.
Thanks,
Darin
Hi,
You can use label="$input.name"
Cheers, Joachim
Joachim Jacob, PhD
Rijvisschestraat 120, 9052 Zwijnaarde Tel: +32 9 244.66.34 Bioinformatics Training and Services (BITS) http://www.bits.vib.be @bitsatvib
On 10/26/2012 04:38 PM, Darin wrote:
Hello,
I'm new to Galaxy, and I've been asked to tweak some tools so that output file names are related to input file names. For example, an input file of myData.abc might lead to an output file named myData_afterMyTool.xyz.
I think the user is really just interested in the labels showing up in the history panel, so that gives me some flexibility.
In my development environment, I've changed one tool's xml file's output tag to this:
<outputs> <data name="out_file1" format="bed" label="${input1}_afterTool.xyz"/> </outputs>
... and that seems to work, but it's using an internal file name, instead of the label for the input file.
So - does anyone know if there's an easy way to get the label from the input file into the output label, instead of the path/dataset_n.dat file format for this name? If there's a better way to approach this, I'd love any suggestions on that, too.
Thanks,
Darin ___________________________________________________________ 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:
and if you use something like;
label="#echo os.path.splitext ( str ( $input.name ) ) [ 0 ] #-BWA.bam"
You can change "SomeFile.fastq" into "SomeFile-BWA.bam".
This allows you to remove the extension (.fastq in this case), insert some extra text after the original name and set the right extension (Not needed for Galaxy, but we are all used to seeing those labels)...
I do this for all the tools in galaxy (bit of a pain, but worth it) so they produce more intelligent names than "12: X on data 10" etc.
Thon
On Oct 26, 2012, at 08:12 AM, Joachim Jacob joachim.jacob@vib.be wrote:
Hi,
You can use label="$input.name"
Cheers, Joachim
Joachim Jacob, PhD
Rijvisschestraat 120, 9052 Zwijnaarde Tel: +32 9 244.66.34 Bioinformatics Training and Services (BITS) http://www.bits.vib.be @bitsatvib
On 10/26/2012 04:38 PM, Darin wrote:
Hello,
I'm new to Galaxy, and I've been asked to tweak some tools so that output file names are related to input file names. For example, an input file of myData.abc might lead to an output file named myData_afterMyTool.xyz.
I think the user is really just interested in the labels showing up in the history panel, so that gives me some flexibility.
In my development environment, I've changed one tool's xml file's output tag to this:
<outputs> <data name="out_file1" format="bed" label="${input1}_afterTool.xyz"/> </outputs>
... and that seems to work, but it's using an internal file name, instead of the label for the input file.
So - does anyone know if there's an easy way to get the label from the input file into the output label, instead of the path/dataset_n.dat file format for this name? If there's a better way to approach this, I'd love any suggestions on that, too.
Thanks,
Darin ___________________________________________________________ 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:
___________________________________________________________ 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:
On Fri, Oct 26, 2012 at 7:53 PM, Anthonius deBoer thondeboer@me.com wrote:
and if you use something like ...
I do this for all the tools in galaxy (bit of a pain, but worth it) so they produce more intelligent names than "12: X on data 10" etc.
Thon
These kind of tricks are quite handy, but in general I'd like Galaxy to make the ancestry of each dataset more visible (I would say "history" but Galaxy already uses that for the collection of current datasets), both in the peek view, but also perhaps a detail view reusing some of the graph layout code used in workflows?
i.e. This isn't something each tool author should be doing, but something general that should be in Galaxy itself.
Peter
On Fri, Oct 26, 2012 at 3:47 PM, Peter Cock p.j.a.cock@googlemail.com wrote:
i.e. This isn't something each tool author should be doing, but something general that should be in Galaxy itself.
Yes, we would very much like a visual/graphical indicator of dataset ancestry, and possibly an alternative history view that was graph based. On the big list once the new History UI framework is done.
(I imagine people will still want to customize labels though).
-- jt
galaxy-dev@lists.galaxyproject.org