Difference between revisions of "Developer Meetings/20130205"

Revision as of 18:51, 8 February 2013

Hide from editors flag on mrml nodes, bug 2906
- model hierarchy optimizations causing unexpected behaviour
Failing tests - ITKv4 and otherwise

Wiki updates
- Re-organized FAQs
  - Sub section of FAQ (UserInterface, DICOM, ...) can be included where it matters. For example see here and here
  - Created a developer FAQ. See here
  - FAQ page now redirect to Documentation/Nightly/FAQ. See here
- Added a Roadmap & priority page - See http://www.slicer.org/slicerWiki/index.php/Roadmap
Build instruction: Pre-requisites are not included directly in the page. (Still need to consolidate Qt prereqs)

Discussed issue bug 2906
- After 4.3.0 is released, the default value associated with HideFromEditor will be set to false in the base class.
- This is captured in the "API changes roadmap". See http://www.slicer.org/slicerWiki/index.php/Roadmap#4.4.0_2
- Also discuss that the search/filter box associated with the data model could be extended to understand more complex expression. This would allow to easily hide or show element of the tree based on regular expression.

Discuss MRML scene versionning
- Since Sept 2012, there is a property named MRML_SUPPORT_VERSION in Libs/MRML. See here
- Possibility of having a XSD schema for MRML scene that would be updated for each change in MRML API. Would the introduction of such a schema imply major change in the current XML structure ?

Nicole discussed the recent changes associated with ModelMaker.
- It can now take a "template model hierarchy" as an argument. More details associated with the commit message: here

Look at the list of test failing tests.
- Alex and Julien will be looking at the failing test vtkSlicerModelsLogicAutoRemoveTest. Julien will also enter an issue to document/track some what should be done: http://www.na-mic.org/Bug/view.php?id=2916 2916]
- Steve reported that the failling test "py_RSNA2012Quant" is probably due to some memory leaks in the Slice Displayable Manager.

Discussed issue that Steve was having when running CTest
- This was due to the fact he updated PATH manually to include some other tools that were conflicting with ctest.

Steve reported a problem when using 4.5.2 to build Slicer.
- As documented on the wiki, waiting CMake 2.8.11 is release. Used the a nightly version of CMake to build on MacOSX 10.8 + XCode >= 4.5.X. See http://www.slicer.org/slicerWiki/index.php/Documentation/Nightly/Developers/Build_Instructions#MacOSX

Worked on issue 2915
- Identified some design flaw in the way "Shared library" CLI are executed. Investigation are on-going. Make sure to follow the issue to get update.

@@ Line 29: / Line 29: @@
 * Look at the list of test failing tests.
-** Alex and Julien will be looking at the failing test [http://slicer.cdash.org/testDetails.php?test=3371208&build=62445 vtkSlicerModelsLogicAutoRemoveTest]. Julien will also enter an issue to document/track some what should be done.
+** Alex and Julien will be looking at the failing test [http://slicer.cdash.org/testDetails.php?test=3371208&build=62445 vtkSlicerModelsLogicAutoRemoveTest]. Julien will also enter an issue to document/track some what should be done: http://www.na-mic.org/Bug/view.php?id=2916 2916]
 ** Steve reported that the [http://slicer.cdash.org/testDetails.php?test=3391876&build=62483 failling test "py_RSNA2012Quant"] is probably due to some memory leaks in the Slice Displayable Manager.