# H5diff functionality/h5compare

**URL:** <https://forum.hdfgroup.org/t/h5diff-functionality-h5compare/5296>\
**Category:** HDF5 Library\
**Created:** [January 15, 2019, 2:33pm UTC](https://forum.hdfgroup.org/t/h5diff-functionality-h5compare/5296 "2019-01-15T14:33:50Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![elmer.vangeijn](https://forum.hdfgroup.org/letter_avatar_proxy/v4/letter/e/eada6e/32.png) [@elmer.vangeijn](https://forum.hdfgroup.org/u/elmer.vangeijn)\
**Post date:** [January 15, 2019, 2:33pm UTC](https://forum.hdfgroup.org/t/h5diff-functionality-h5compare/5296/1 "2019-01-15T14:33:50Z")

</div>

Hi!

I intended to suggest an option to suppress the ‘0 differences found’ results of h5diff, but then stumbled upon an [old post with the exact same question](https://forum.hdfgroup.org/t/suppressing-0-differences-found-from-h5diff-r/2583). The [issue [HDFFV-8088]](https://jira.hdfgroup.org/browse/HDFFV-8088) created in response to that post denies the request, but contains the promising words:

> 1. MOST RELEVANT: We are developing a new tool, h5compare, which will  
> include a uniform way for a user to filter out uninteresting  
> information

Since the above post and issue are quite old, I am now wondering whether this statement still reflects the current stance on the issue and what the development status of the mentioned h5compare tool is?

Thanks!

Elmer van Geijn

---

<div class="post-metadata">

**Author:** ![epourmal](https://forum.hdfgroup.org/letter_avatar_proxy/v4/letter/e/c67d28/32.png) [@epourmal](https://forum.hdfgroup.org/u/epourmal)\
**Post date:** [February 18, 2019, 2:40am UTC](https://forum.hdfgroup.org/t/h5diff-functionality-h5compare/5296/2 "2019-02-18T02:40:00Z")

</div>

We would love to drop h5diff. It looks like every time we try to solve an h5diff problem we create a new one. The root cause of this situation is a lack of initial well-thought-through design and documentation to maintain the tool.

Unfortunately, our development of h5compare is on hold. If you are interested in what we had in mind as a replacement, I created HDFFV-10704 issue (it should be open to everyone) and attached the design document (work in progress) to the issue. At least we started to think what does it mean to compare HDF5 objects 😉

Thank you!

Elena