Pynbody

Latest version: v1.6.0

Safety actively analyzes 629959 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 5 of 7

1.0.3

Bug fixes
=======

* Resolve an incompatibility with python 3.9 in the PyPI source distribution (due to historical Cython incompatibility)
* Fix bug with GadgetHDF where incorrect units could be inferred in some circumstances
* Fix bug in the integration limits for 2D projected density maps (elehcim)
* Correct definition of 'dyn' unit

Enhancements
===========

* Add ability to read starlog meta-data as written by recent versions of Changa (emapple)
* Automatically wrap/tile particles over the simulation box size when using `sph.to_3d_grid` (as previously implemented for `sph.image`)
* Add support for adaptahop catalogues (cphyc)
* Improve support for storing profiles (Martin-Rey)
* Add ability to calculate SPH curl and divergence (elehcim)
* Add support to read tracer particles from ramses snapshots (cphyc)

1.0.2

This fixes a bug in the initial pip release of 1.0.1, where the README file was not included in the package, leading to an error on executing setup.py (575).

1.0.1

*This version ends support for Python 2.7, in favour of supporting Python 3.5, 3.6, 3.7 and 3.8*

The increase in the major version number reflects this, rather than the addition or change of any major functionality. Pip should automatically install older versions if you attempt to use pynbody with Python 2.7.

Bug fixes
--------

* Restores the Kravtsov and Hudson Stellar-Mass-Halo-Mass relations (bwkeller, 568)
* Fixes a bug with addition of dimensionless units (565)
* Fixes compatibility with Python 3.8

0.48

Bug fixes
--------

* Fix issue in fast_kde (519) where floating point numbers were accidentally used as an array sizze
* Compatibility fix for shared memory on macOS Catalina

Performance enhancement
-----------

* Improve ramses reading times by reimplementing in Cython (cphyc, 548)

Features
----------

* Add option to write non-cosmological tipsy files (524)
* Improve ability of ramses reader to automatically run part2birth utility and use the results for a sensible `tform` array ( Martin-Rey, 520)
* Add support for AdaptaHOP halo finder (cphyc, 546)

0.47

Enhancements
===========

* Add a halo mass function analysis tool in `pynbody.analysis.hmf` (Martin-Rey)
* Add halo shape analysis tools (Matt Orkney)
* Introduce option to force gadget loader to use non-cosmological units (`ignore_cosmo=True`; elehcim)
* Add ability to use a custom CMD luminosity file in `pynbody.analysis.luminosity`
* Fix major performance issue with ramses files in Python 3.x (which arose due to the interaction between numpy and Python 3.x IO caching system)
* Increase number of fields read by the grafIC reader

Bug fixes
=======

* Allow use of AHF reader in read-only folders (issue warning instead of exception that cache cannot be written)
* Allow SPH smoothing to work even if units of box size are unknown
* Fix issue where IndexedSimArray with converted units could crash Python 3.x (elehcim)
* Compatibility fix for iteration in Python 3.7
* Fix issue where AHF load checking would ignore a specified `ahf_basename`, preventing `f.halos(ahf_basename='...')` from working
* Fix mismatched signature in AHF reader's `writehalos` method, which caused it to fail

0.46

*Enhancements:*
* Add default mapping for black holes in GadgetHDF snapshots
* Better dependency specification in `setup.py` for automated installation (rokroskar)
* Handles Eagle-like GadgetHDF snapshots which store their halo/group catalogue as two arrays (specifying the group and subgroup number)
* Add ability to save back individual arrays in GadgetHDF format using `snap['arname'].write()` or `snap.family['arname'].write()` syntax
* Performance enhancements to GadgetHDF implementation

*Bug fixes:*
* Fixed an issue where mu (relative atomic mass) was defined incorrectly in some cases (mtremmel / trquinn)
* Fixed an issue in IndexedSubSnap, where an entire family array could be incorrectly returned instead of an empty array when no family members were present in the IndexedSubSnap
* Fixed an issue where inconsistent smoothing information would prevent a KDTree from generating density estimates; now KDTree insists on using its own smoothing lengths to prevent this occurring.

Page 5 of 7

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.