# Improving testing data management for "self-test"

**URL:** <https://discourse.slicer.org/t/improving-testing-data-management-for-self-test/5014>\
**Category:** Development\
**Created:** [December 8, 2018, 7:48am UTC](https://discourse.slicer.org/t/improving-testing-data-management-for-self-test/5014 "2018-12-08T07:48:42Z")\
**Posts on this page:** 1\
**Showing post:** 4

<div class="post-metadata">

**Author:** ![lassoan](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lassoan/32/13_2.png) [@lassoan](https://discourse.slicer.org/u/lassoan)\
**Post date:** [December 10, 2018, 12:24am UTC](https://discourse.slicer.org/t/improving-testing-data-management-for-self-test/5014/4 "2018-12-10T00:24:52Z")

</div>

> [@pieper](#):
>
> unified caching infrastructure

Should that be high-level feature implemented in SampleData/Python or low-level feature implemented at MRML level in C++?

I tend to prefer the former option (remove networking from MRML and add it at higher level), as I don’t think network communication could be part of MRML (https, various authentication protocols, etc. would be too complicated to implement; simple anonymous public data download over http would be too limited). We could keep cache management (finding cached files, cleaning up old files, etc) in C++ and add checksum support.

---

_[View the full topic](https://discourse.slicer.org/t/improving-testing-data-management-for-self-test/5014)._