We are proud to announce the release of yt 3.2.1! yt (http://yt-project.org) is an open-source, community-developed toolkit for analysis and visualization of volumetric data of all types, with a particular emphasis on astrophysical and nuclear engineering simulations. Version 3.2.1 is a regularly scheduled bugfix release including fixes for a number of issues reported since the release of yt 3.2. These changes are summarized below and include a number of bugfixes, performance improvements, and documentation improvements. We urge all users of version 3.2 to upgrade to yt 3.2.1 immediately. Summary of changes ------------------ * Fixed an issue with reading ORION2 star particle data * The "particle_velocity_cylindrical_radius" field now uses the correct field definition. * Prevent unit test failure from occuring if scipy or astropy is not installed * Present better error messages when trying to load empty files or directories * Add support for the fabs ufunc to YTArray * Remove enzo installation option from the install script * Fixed issue with the EPS writer when it is supplied a ProfilePlot * Fixed reading of the NMSU-ART star particle creation time field * Use a stronger hashing algorithm for calculating selection hashes to fix issues with data object selection in Python 3 * Fix an error in the docstrings of Cosmology.t_from_z() * Fix python3 answer test failure for Enzo species field names * Fix incorrect use of "normal" instead of "center" in spherical coordinate fields. * Avoid NameError when comparing temperature units with offsets * Avoid creating YTQuantity instances with more than one value * Include requirements files and cuda sources in source distribution. Avoid including a full docs build. * Add search bar to main docs page * Add canonical link for documentation to improve google search results for yt documentation * Use nice latex representation for unit labels returned by DerivedField.get_label() * Add assert_allclose_units to avoid test failures in Numpy 1.10 and newer * Avoid unsafe casting error in Numpy 1.10 or newer * Fix incorrect plots created by scripts that call PlotContainer._setup_plots() directly. * Allow the creation of gradient fields based on fields with dimensionless units * Correct two incorrect paths in the documentation on loading data * The MaxLocation and MinLocation derived quantities now work correctly with non-cartesian data and particle fields. * Use OrderedDict in PlotWindow answer tests to avoid unstable test names in Python 3 * Speed improvements for the absorption_spectrum analysis module * Speed improvements for smoothed covering grids * Fixed a small CSS issue in the main docs page * Fix incorrect velocity and time units for Gadget cosmological datasets * Use truediv operator in unit object tests for Python 3 compatibility * Fix several bugs in the photon simulator analysis module. * Ensure camera width and center are in code units before rendering, avoiding a possible segmentation fault. * Ensure fields passed to process_octree during smoothing are contiguous. Fixes a regression in the values of smoothed SPH fields. Coming Soon! ------------ The next major release of yt will be version 3.3, which is slated to include an overhaul of the volume rendering system and support for analyzing and visualizing unstructured mesh data. Standard Installation Methods ----------------------------- As with previous releases, you can install yt from source using one of the following methods. 1) From the install script (http://yt-project.org/#getyt ): If you previously installed yt from the install script, you can update your installation in-place using the following command: # Update $ yt update If you are on the “stable” branch, updating will bring you from yt 3.2 to 3.2.1, incorporating all changes since 3.2, whereas if you are on the development “yt” branch, only the changes since your last update should be incorporated. To install from scratch, do the following: # Installation $ wget http://bitbucket.org/yt_analysis/yt/raw/stable/doc/install_script.sh $ bash install_script.sh 2) From pip (source or binary wheel, see below for more details): # Installation $ pip install yt # Update $ pip install -U yt 3) From the Anaconda Python Distribution (https://store.continuum.io/cshop/anaconda/): # Installation $ conda install yt # Update $ conda update yt Note that it might take up to a week for the conda package built by Continuum Analytics to be updated for yt 3.2.1. Installing Binary Packages via pip If you do not have compilers available, it is also possible to install binary packages (“wheels”) using pip on Windows and Mac OS X (64-bit only for both). This has the advantage of not needing to install yt from source using a proper compiler setup, which has caused occasional problems on both of these platforms and prevented us from installing yt easily on other Python distributions. Note that it may take several days for binary wheels for yt 3.2.1 to be uploaded. We have so far been able to install and run the binary distribution via pip on the following platforms and Python stacks: Note that it might take a day or two for the pip wheels to be updated. Windows x86_64: Enthought Canopy Python (https://www.enthought.com/products/canopy/) WinPython (http://winpython.sourceforge.net/) Mac OS X x86_64: Enthought Canopy Python (https://www.enthought.com/products/canopy/) Homebrew Python (http://brew.sh/) Python.org Python Mac OS X’s system Python MacPorts Python (https://www.macports.org/) This is somewhat experimental, so other distributions may work (or not), please submit bug reports or successes to the mailing list or to the Bitbucket issues page (http://bitbucket.org/yt_analysis/yt/issues). Both Python 2.7 and Python 3.4 are supported. The dependencies for installing yt via this method are the same as from source: NumPy h5py HDF5 SymPy Matplotlib IPython (not required, but strongly recommended) To install a new version of yt on one of these platforms, simply do $ pip install yt and you should get the binary distribution automatically. Also, if your python installation is system-wide (e.g., the Mac system Python) you might need to run pip with administrator privileges. For more information, including more installation instructions, links to community resources, and information on contributing to yt’s development, please see the yt homepage at http://yt-project.org and the documentation for yt-3.2 at http://yt-project.org/docs/3.2. yt is the product of a large community of developers and users and we are extraordinarily grateful for and proud of their contributions. Please forward this announcement on to any interested parties. As always, if you have any questions, concerns, or run into any trouble updating please don't hesitate to send a message to the mailing list or stop by our IRC channel. All the best, Nathan Goldbaum, on behalf of the yt development team