Permalink
Commits on May 23, 2018
  1. Also adds project files to gitignore! (Why did I wait so long to appl…

    tngreene committed May 23, 2018
    …y this stuff?) In addition, there are now tests (on another branch that require 2.78. Just one test really) so lets make the standard until we can all catch up together.
  2. In addition to being far far far out of date, a resource I have NEVER…

    tngreene committed May 23, 2018
    … used in developing this plug in, we're moving to using pydoc strings and Type annotations as we should have. This messes up github decision of what language we're using (which I admit can be changed without deleting this stuff), but its time to ditch this stuff anyway.
  3. Adds a bunch of dot folders to the gitignore file. The ignore, mypy, …

    tngreene committed May 23, 2018
    …PyCharm, VsCode, and the builds folder.
  4. Fixes README.md content

    tngreene committed May 23, 2018
  5. Merge branch 'master' into v3-4

    tngreene committed May 23, 2018
    # Conflicts:
    #	README.md
Commits on May 15, 2018
  1. Updated and fixed unit tests! Due to some floating point errors some …

    tngreene committed May 15, 2018
    …tests weren't passing anymore. TestCase5's blend file simple had a little cleanup of names and workspace window, also setting the Print OBJ Info checkbox to true. The only obj that really mattered in updating was mega_nested_mega.obj, but the rest were nice too. io_xplane2blender.tests now has assertMaticesEqual available to all now.
Commits on May 14, 2018
Commits on May 10, 2018
  1. Now that we know Travis CI can export this in about 42 to 44 seconds,…

    tngreene committed May 10, 2018
    … we can include this nice block of text. As I run this more I'll see if 63 is too high. For now, it seems okay.
  2. I want to run the test again and see how long it takes for two data p…

    tngreene committed May 10, 2018
    …oints, so, here's a small change.
  3. This is a two part commit. It adds the new unit test for this perform…

    tngreene committed May 10, 2018
    …ance fix, but, since we don't know how fast travis is going to run this thing, we'll have to see what it says. Hopefully it can run it in less than 63 seconds. Otherwise we'll need to add special code that can tell if it is part of a travis tests. Otherwise, we could just have travis ignore the performance tests since people aren't making scenery export on Travis CI servers anyway.
Commits on May 9, 2018
  1. Fix bug #350 - connected child bones with leftover translations are n…

    Ben Supnik
    Ben Supnik committed May 9, 2018
    …ot WYSIWYG
    
    Blender ignores translations when connected and child, but we have
    to simulate this because we are extracting keyframes directly.
  2. Fix bug #351 - bone animation parenting data block is not WYSIWYG

    Ben Supnik
    Ben Supnik committed May 9, 2018
    The old exporter would find the pre-animation matrix for DBs by walking
    forward from the parent and simulating the parent-child process.  This
    simulation was wrong for bones and generally error-prone since it
    required us to duplicate a big chunk of opaque Blender code.
    
    The new method walks backward from the datablock pose; since data block
    animations are last in Blender's transform algo, this is actually a lot
    less work, and more reliable.
Commits on May 4, 2018
  1. Although making indicies an array from the beginning didn't seem to h…

    tngreene committed May 4, 2018
    …ave a noticable impact on time, it is free and easy and adds a bit of typing to it which I don't mind.
  2. writeIndicies has been rewritten to not only be more efficient in tim…

    tngreene committed May 4, 2018
    …e, but to never have 10 dangling IDX lines when we could have one IDX10!
Commits on May 2, 2018
  1. By using an array to collect the data instead of a list, using the fa…

    tngreene committed May 2, 2018
    …ster % string formatter, and applying the formatting later, we can now export 100 spheres in 46 seconds (writeVertices takes 19 of them. The previous commit message's 36 seconds was referingto this as well)!
  2. Simply using array instead of list decreases times for ExportTest wit…

    tngreene committed May 2, 2018
    …h 100 spheres from 88.9 seconds to 36 seconds!
Commits on Apr 20, 2018
  1. Update of README to simplify and corrent Automatic Installation instr…

    tngreene committed Apr 20, 2018
    …uctions, changes minimum Blender level to 2.76 in README
Commits on Jan 2, 2018
Commits on Dec 23, 2017
  1. Removes xplane2blender_ver property from the data model and replaces …

    tngreene committed Dec 23, 2017
    …it with VerStruct.current()
Commits on Dec 20, 2017
Commits on Dec 19, 2017
  1. XPlaneDatarefs now placed

    tngreene committed Dec 19, 2017
Commits on Dec 18, 2017
Commits on Dec 15, 2017
  1. Updates fixture file and changes xplane_file's self.objects dictionar…

    tngreene committed Dec 15, 2017
    …y to be ordered to get deterministic output
Commits on Dec 14, 2017
  1. New unit test for testing the relationship between Armatures, bones, …

    tngreene committed Dec 14, 2017
    …and datablocks! Right now it seems not to want to pass because it looks like it is not deterministically iterating through the bones for the non-animated lights.