Marlin Mixon · Nov 4, 2021

IRIS Language Bridge Python

At the recent virtual conference I saw the demonstration on being able to run Python within ObjectScript and run ObjectScript within Python thanks to a feature called Language Bridge.  Is this available now? If so, am I able to install it?  I am currently running HealthShare 2020.1.  I have a little bit of experience using Irisnative with Python but was intrigued with the more seamless-looking features offered by the Language Bridge.

Thanks for any insight.

Marlin Mixon

Product version: HealthShare 2020.1
0 206
Discussion (5)3
Log in or sign up to continue

And as you might also have seen the version was 2021.2  which is not released yet.

Interesting to see if this will be a container-only release (as intermediate releases tend to be) or a full kit release.
In the Early Adaptor Program, there were full kits available, which would be my preference.

@Herman Slagman  - FYI ... we will be making full kits available for the 2021.2 CD (Continuous Delivery) release.  Stay tuned for more announcements coming this month about the Preview for that release.  

Embedded Python is a new feature being added in the upcoming IRIS 2021.2 and IRIS for Health 2021.2.  We're getting close to releasing a preview, but it's not yet available.  

Calling any interested Python developers: It would be interesting for someone to take a look at what I've done in JavaScript/Node.js to implement what I refer to as persistent JSON objects, essentially creating JavaScript objects/JSON that resides on disk rather than memory and which can also be manipulated, traversed and modified in-situ rather than being shuttled between disk and memory.  It's a concept I've referred to as QEWD-JSdb.  Given the native support for JSON that now exists in Python, it strikes me that the equivalent ought to be possible using the new embedded Python in IRIS.  I'm not a Python developer so I'm not the person to implement it, but all the logic for implementing this concept is available in my JavaScript repositories and my assumption is that it should be a matter of recasting the logic in Python.  The really cool part is the multi-model stuff you can then derive from it such as the persistent DOM against which a standard XPath library can then be applied for searches.  For information on what I'm referring to, see:

and a more broad discussion of the underlying concepts:

If anyone is interested in such a project, all I ask is for the appropriate attribution for the original concept.