Version 1.0.8

• make compatible with scipy >= 1.8
• various updates to plotting in cla
  • update logic and options for the "magpct" plots
  • avoid using matplotlib.pyplot unless showing plots on screen
• add ytools.max_complex_vector_sum
  • compute maximum complex vector from x and y components
• miscellaneous docstring updates & bug fixes

Version 1.0.7

Minor update, primarily to add ytools.max_complex_vector_sum.

Version 1.0.6

Added ytools.compmat - a matrix comparison tool
Moved read_text_file/write_text_file decorators from ytools to guitools

Version 1.0.5

This is a relatively minor update:

• Updated some of the documentation examples to be compatible with Matplotlib 3.5.0.
• Updated how the pre_eig option is handled in SolveUnc to make it smarter
• Officially supports Python 3.9
• Fixed some bugs, mainly the bug in pyyeti.nastran.n2p.replace_basic_cs. It incorrectly used "0" for the rectangular type when it should use 1.

Version 1.0.2

Fixed an issue with the output4 reader (pyyeti.nastran.op4) that caused it to fail on some reads. (pyYeti should be able to read all formats.) Thanks to Jeremy Priest for the bug report!

Also did some code cleaning & refactoring here and there.

Version 1.0.1

Made the inner loop of the diagonal solver SolveUnc more efficient in terms of both speed and memory usage. Numba is used if available. Many thanks to Jeremy Priest for instigating this update! :) More updates like this are planned in the near future.

Documentation updates

Version 1.0.0

Added the "drms" option to cla.rpttab and cla.rptext
Enhanced n2p.find_xyz_triples to be more robust
In psd.rescale, changed "frange" default start to be 1.0 instead of 25.0
Documentation updates

Version 0.99.9

Primarily a code cleaning update along with some minor enhancements & bug fixes.

Changed the "incrb" option in the frequency domain solvers to be a string (like "dva") instead of the impossible to remember integer 0, 1, or 2.

Version 0.99.8

Various updates to srs.srs_frf; especially the documention to show
frequency domain base drive transfer function derivation

Updated cb.mk_net_drms handling of the cglf category when s/c axial
doesn't line up with l/v axes

Updated cyclecount.binify to be much faster in some cases by using
numba. Many thanks to Daniel Kowollik for spearheading this update!

Lots of minor documentation updates, code clean-ups & bug fixes

Version 0.99.7

Another bug fix for the op2 reader: the matrix reader didn't properly check for word size.