Dannon Baker wrote:
The 'step' in question is the actual workflow step id, since ordering of steps in a workflow is flexible and might be changed without realizing it by moving steps around in the editor. The easiest way to retrieve this identifier is to use the API and view the workflow in question.
Here's an example of that output, using scripts/api/display.py
$ python display.py <my_api_key> http://localhost:8080/api/workflows/ebfb8f50c6abde6d Member Information ------------------ url: /api/workflows/ebfb8f50c6abde6d inputs: {'16': {'value': '', 'label': 'Input Dataset'}} id: ebfb8f50c6abde6d name: Foo User's Workflow
For inputs, you have a dictionary containing (in this case just one) the workflow steps, their labels, etc. So, regarding the workflow_execute last argument, it's the step id from above, and the source is 'ldda', 'ld', or 'hda', depending on what you're using as an input. For inputs you're pulling out of a history, use 'hda'. The other options are for using datasets directly from data libraries.
-Dannon
On Sep 6, 2011, at 1:24 PM, Roxanne Tapia wrote:
I was trying to use the workflow_execute.py
And I can't understand the last argument:
usage: workflow_execute.py key url workflow_id history step=src=dataset_id
So in step=src=dataset_id step: I think equals the next step #in the history? src: ??? dataset_id: This has to be the dataset id from a data library which I can get by using display.py to query libraries & contents of libraries. Right?
In addition to what Dannon said above, there's currently a confusing difference with respect to the dataset_id. This is most easily pulled from the URL of a dataset in the library or history UI, rather than the API. For the details on why the ID in the API won't work, see this thread from a couple weeks ago: http://lists.bx.psu.edu/pipermail/galaxy-dev/2011-August/006310.html We'll have a fix for this at some point. --nate
Can anyone explain this to me?
Roxanne Tapia Software Developer 3 B-6, Genome Science Bioscience Division rox@lanl.gov (505)665-0206
___________________________________________________________ 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: