5.7.15

Minor Improvements

• Avoid keyPressEvent interfering other widgets (#529)
• Fixed #508, sorting creator plugins by label (#530)
• Fixed #536, Scene Inventory now can select entire rows from action (#537)
• Implemented #544, Workfiles now closes itself on scene open (#545)
• Added QWidget.activateWindow() call on {tool}.show() to guarantee window shown on launch (#522)

5.7.0

Major Changes

• ✨ Avalon Loader load subset with options (#482, #518, #521)
• 🚀 Fixed issues in Maya 2020 (#486, #488)
• 💪 Refactored Avalon config import method into lib.find_submodule (#501, #515)

Minor Improvements

• Fixed Avalon tools not parenting in main window (#472)
• Fixed Loader in Houdini 18 returning version number as 'long' not 'int' (#479)
• Improved fps mapping in Maya (#475)
• Improved log message for no creator plugin been found (#490)
• Tiny Houdini lib.lsattr improvement (#491)
• Fixed OptionalMenu highlighting in Loader (#521)
• Fixed missing key AVALON_HIERARCHY in Avalon Session (#525)
• Tiny Avalon GUI styling improvement (#528)

Nuke Integration 🚧

The Avalon-Nuke integration is still in developement here, there's an PR #504 which is built on top of these changes and waiting for more tests and feedbacks. Welcome to give it a go !

• Re-implemented Nuke imprint (#473)
• Implemented Nuke's lib.read (#476)
• Porting #438 (#484)

5.6.2

Fix Avalon-Nuke implementation

See #471 for detail.

5.5.38

• Optimized Maya host.ls() method - #456

  Instead of using avalon.maya.lib.lsattr which scanning all nodes, implemented an internal method specifically for iterating only objectSet nodes for listing containers. The speed is about 5 times faster.

• Removed schema validation when parsing containers - #458

  The container data schema validation was dragging the process when the scene has lots of subset loaded. Removing it for performance.
  The flag validate=True in pipeline.parse_container has been removed.

• Fix avalon.lib.launch() for non-Windows platform - #459

  The subprocess.Popen flag creationflags only works on Windows, and cause error on non-Windows platform.

• Improve Python 3 compatibility - #462, #464, #469

  Improve for host which using Python 3+, e.g. Blender.

• Fix Workfiles app for working in Blender - #465

  This is a must for upcoming Avalon-Blender implementation. 🚀

• Minor fix

  • Fix tools debug mode - #470

5.5.15

• Exposed the "loader" column in Scene Inventory : #455
• Made the "No Task" entry clearly distinct from actual tasks by its icon : #452
• Fixed context manager allowing to switch to 'No Task' : #451
• Fixed QPushButton::menu-indicator being cut off : #450

5.5.0

• Resolved #246 : New document schema to store families in subset

  If you want to adopt this, the Pyblish Integrator plugin which responsible to write subset and version into the database, require to update the document schema name.

  And here's the simplest principle that worth to follow:

  • If subset already exists, do not use new schema on next version.
  • If subset will be created on integration, do use the new schema on both subset and version.

  Backward compatible, see #443 for detail.

• Resolved #407 : Refactor Work Files API

  • Renamed workio.open to workio.open_file in each host.
  • Renamed workio.save to workio.save_file in each host.

  This is backward compatible, see #449 for detail.

• Optimized AssetModel

  The launch time of Loader tool and Project Manager has been decreased tremendously.
  See #447 for detail.

• Resolved #179, #289, #445 : Improved Creator tool

  • Better visual feedback and status message from subset name input field
  • Validate inputted subset name immediately with GUI feedback
  • Fixed asset name validation bug

5.4.0

• Add contribution guidelines
• Repair Nuke-integration after previous release
• Repair Loader tool after #388
• Fix #189
• Fix #434 , also expose Project Manager to Maya's Avalon menu under System sub-menu
• Fix #350, see #431 for short detail
• Fix #411
• Refactor avalon.tools, see #440 for details

5.3.0

Merged #371 for AVALON_HIERARCHY and friends, along with #370 for an initial Nuke integration. Marking this pre-release until #417 is implemented.

5.2.9

Revert overly required members of schema

5.2.6

Preparing for integration with Allzpark