In the meantime, the suggestion to create a script and then upload it to IRIS seems like the best advice. Is there any way for us to create a script for downloading metabolic models (even if the users needed to edit it a bit to get exactly what they wanted? We could perhaps give them a skeleton of a script?).
Nomi
On Aug 29, 2014, at 7:26 AM, Michael Mundy (JIRA) <
help@kbase.us> wrote:
|
This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira |
There is a way to script in IRIS but it is limited. You can save commands in a file and then run all of the commands from the file using "execute". Unfortunately there are no variables or control flow features in IRIS scripting. You could generate the file with your favorite scripting language on your computer and then upload the script to IRIS. Run "whatsnew" from the IRIS command line to get some details on the scripting support.
Using the restful API or command line is difficult right now with no pre-built client image available. Building your own client image can be done but that is not in a very good state at the moment. The fba modeling service has dependencies on the typecomp, kb_seed, workspace_service, auth, idserver, kb_model_seed, probabilistic_annotation, and workspace_deluxe services. I'm pretty sure the workspace_service and auth repos are not moving to github so you can't build fba modeling unless you have a developer account. And you would still need to track down a runtime image and I don't think there is a current image that is publicly available.
Do we know what language the user wants to use when accessing the restful API? With Python it might be possible to grab the Client.py file and get something to work. Also, what type of system is the user running on (Linux, Mac, etc.)?