Commit 952eff39 authored by David E. DeMarle's avatar David E. DeMarle

Add release notes for 5.7.1.

parent bdf710ea
ParaView 5.7.1 Release Notes
============================
Major changes made since ParaView 5.7.0 are listed in this document. The full list of issues addressed by this release is available
[here](https://gitlab.kitware.com/paraview/paraview/-/milestones/16).
# General Notes
This patch release is mostly dedicated to various import and output bugs that were found in the 5.7.0 release. It does however have one new feature, a plugin that converts data into a tree based AMR format.
### New Features
* The new 'HyperTreeGridADR' plugin adds a 'Resample To HyperTreeGrid' filter that converts any data set into a tree based AMR data type. It does so by doing intelligent resampling that optionally takes into account attribute expressions that control the refinement. The algorithm follows the approach outlined in the Nouanesengsy et al's Analysis-Driven Refinement paper from LDAV 2014.
* ParaView can be now compiled with OpenImageDenoise releases other than 0.8.
* The Web exporter for standalone data viewing now preserves dataset names.
### Bug fixes
* There are a number of miscellaneous refinements to 5.7.0's new cmake infrastructure.
* A crash in the Cell Size filter of 5.7.0 is fixed.
* Two regressions in which HyperTreeGrid .htg writer did not support multicomponent arrays and could not be saved into .vtm collections are fixed.
* There are a number of miscellaneous bug fixes to HyperTreeGrid filters.
* Two bugs are fixed in the AMReX (formerly BoxLib) reader.
* The MPAS reader no longer issues warnings when reading datasets without time.
* Inconsistent time information and cell arrays in the Truchas reader are fixed.
* Incorrect 2D rectilinear grid orientation with XDMF readers is corrected.
* The new ExportNow feature now properly skips screenshots from deselected views.
* Catalyst scripts with Cinema image exports no longer skip deferred rendering attribute array images.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment