SLiMSuite REST Server


Links
REST Home
EdwardsLab Homepage
EdwardsLab Blog
SLiMSuite Blog
SLiMSuite
Webservers
Genomes
REST Pages
REST Status
REST Help
REST Tools
REST Alias Data
REST API
REST News
REST Sitemap

SLiMSuite REST Help

Programmatic access to SLiMSuite REST services

By default, the REST services return human-readable HTML when queued or running. Default output is formatted full server output.

Individual components can be accessed in plain text using &rest=X, where X is the chosen server output. Details of server output can be found in the main Help page for universal outputs, or by running the chosen program with &rest=outfmt. REST outputs are also listed in the main documentation for the program, accessed with &rest=docs.

However, even when plain text output is being required, the main program run may still need to go into a queue and/or may take several minutes to complete. To avoid proxy timeouts, these outputs will still go via a human-readable intermediate queue page.

Programmatic access is therefore a three step process:

  1. Set the job going and retrieve the JobID by calling the program with &rest=jobid. This will return the job ID in the form of an 11-digit number, e.g. 15010900002.

  2. Check the progress of your job using check&jobid=X, where X is the job ID returned in step 1. This will return Queued, Running or Finished, dependent on the status of the job. If something has gone wrong, Failed will be returned. Please cycle and check periodically until Finished is return. We request that you do not make this cycle period too short - once every few seconds/minutes (depending on the job) is recommended.

  3. Once Finished, retrieve the desired output using retrieve&jobid=X&rest=X. To get all outputs, use &rest=full. The tool SLiMParser can be used to split the full output into separate files if desired. If there is a problem, the output should start with ERROR:.

Note: By default, jobs are unprotected. To eliminate the risk of other users retrieving your jobs, you can add &password=X to each of these calls.

SLiMParser can also be used to execute this process in full:

python SLIMSUITEPATH/tools/slimparser.py restin="RESTURL" restout=T restoutdir=PATH

Note: The double quotes around RESTURL are important. Without them, a truncated URL will be sent to SLiMParser, which will be set to run in the background due to the `&` in the URL.

More help

Please visit http://slimsuite.blogspot.com.au or email richard.edwards@unsw.edu.au.


© 2015 RJ Edwards. Contact: richard.edwards@unsw.edu.au.