Difference between revisions of "Developer Meetings/20120515"

From Slicer Wiki
Jump to: navigation, search
 
(3 intermediate revisions by the same user not shown)
Line 1: Line 1:
Attendees:  
+
Attendees: Alex, Chris, Jc, Julien, Nicole, Steve
  
 
== To discuss ==
 
== To discuss ==
Line 13: Line 13:
 
** Every developer should make sure the issue assigned to him will be resolved 3 weeks before release date (August 1st)
 
** Every developer should make sure the issue assigned to him will be resolved 3 weeks before release date (August 1st)
 
* Road map is forming: http://na-mic.org/Mantis/roadmap_page.php
 
* Road map is forming: http://na-mic.org/Mantis/roadmap_page.php
** VS2010: Slicer 4.2.0 ?
+
** VS2010: Slicer 4.2.0 ? http://www.na-mic.org/Bug/view.php?id=2002
 
** ITKv4: Slicer 4.3.0 ? http://na-mic.org/Mantis/view.php?id=2007
 
** ITKv4: Slicer 4.3.0 ? http://na-mic.org/Mantis/view.php?id=2007
 
** Qt 4.8.1: Slicer 4.3.0 ? See http://www.na-mic.org/Bug/view.php?id=2046
 
** Qt 4.8.1: Slicer 4.3.0 ? See http://www.na-mic.org/Bug/view.php?id=2046
Line 45: Line 45:
  
 
== Conclusion ==
 
== Conclusion ==
 +
 +
* Discussed release date for 4.1.1 and 4.2
 +
** 4.1.1
 +
*** Planned for June 1st. No feature freeze is expected.
 +
*** It will consist in creating a branch of 4.1 and integrating tweaks mainly related to extensions support. Related issues are listed [http://www.na-mic.org/Bug/search.php?project_id=3&sticky_issues=on&target_version=Slicer+4.1.1&sortby=last_updated&dir=DESC&hide_status_id=-2 here].
 +
** 4.2
 +
*** Need to clearly define which features and bug fixes we want to integrate. As of today, the list is [http://www.na-mic.org/Bug/search.php?project_id=3&sticky_issues=on&target_version=Slicer+4.2.0+AHM+Summer+2012&sortby=last_updated&dir=DESC&hide_status_id=-2 here].
 +
*** No special release will be done for RSNA (November). The plan will be to use 4.2.
 +
*** As of today, all issues labelled "crash" have been targeted for 4.2.
 +
*** '''TBD:''' Before *NEXT* Tuesday, developers are expected to review issue assigned to them and targeted for 4.2. If you believe an issue can't be resolved by mid-August / beginning of September, please let me know.
 +
 +
* Discussed update of DCMTK
 +
** Integration of DCMTK RT components/classes into official repository will be done at project week.
 +
 +
* Discussed VS2010 Support
 +
** Currently targeted for 4.2.0
 +
** Issue [http://www.na-mic.org/Bug/view.php?id=2002 #2002]
 +
 +
* Discussed ITKv4 Support
 +
** Currently targeted for 4.3.0
 +
** Issue [http://www.na-mic.org/Bug/view.php?id=2007 #2007]
 +
 +
* Discussed Qt 4.8.1 Support
 +
** Currently targeted for 4.3.0
 +
** Issue [http://www.na-mic.org/Bug/view.php?id=2046 #2046]
 +
 +
* Discussed transition to Git
 +
** Will definitively happen before the end of they year
 +
** Probably for 4.3.0
 +
** Issue [http://na-mic.org/Mantis/view.php?id=2059 #2059]
 +
** Before doing so, testing data will be moved into Midas. See [http://www.na-mic.org/Bug/view.php?id=2060 #2060]
 +
 +
* Discussed directory layout of scripted module in Slicer4
 +
** Scripted module will be moved into dedicated directory. Will be similar to scripted extension
 +
** Issue: [http://www.na-mic.org/Bug/view.php?id=2058 #2058]
 +
 +
* Discussed the layout of extensions
 +
** As of today, it will always be assumed a CMakeLists.txt will be available at the top level of your extension

Latest revision as of 18:09, 16 May 2012

Home < Developer Meetings < 20120515

Attendees: Alex, Chris, Jc, Julien, Nicole, Steve

To discuss

= Volume streaming and MRML

From Laurent:

Hello,

I've seen Andriy is using vtkMRMLDiffusionWeightedImageNode to store several volumes in his 4D module.
I would like to use it with the same purpose, but with streaming volumes.

The major difference is I don't know number of volumes I will receive. I looked at the vtkMRMLDiffusionWeightedImageNode and, when we call SetNumberOfGradient, it seems, it's reseting BValue array and gradient array. 

If I understood it, we use gradient array to store our ImageData. 

My first thought has been, when I  receive a volume, increase the value of NumberOfGradients, then, add the new volume in the gradient array.
However, if SetNumberOfGradient is reseting these arrays, this will erase my previous volumes store in it.

I still can copy them before, then SetNumberOfGradient, then put them in the array (with the new size) and add my new volume at the end, but this is not efficient at all (especially when number of volumes is increasing).

Please, could you let me know if I misunderstood something. If not, is there any workaround ?

Thank you.

Conclusion

  • Discussed release date for 4.1.1 and 4.2
    • 4.1.1
      • Planned for June 1st. No feature freeze is expected.
      • It will consist in creating a branch of 4.1 and integrating tweaks mainly related to extensions support. Related issues are listed here.
    • 4.2
      • Need to clearly define which features and bug fixes we want to integrate. As of today, the list is here.
      • No special release will be done for RSNA (November). The plan will be to use 4.2.
      • As of today, all issues labelled "crash" have been targeted for 4.2.
      • TBD: Before *NEXT* Tuesday, developers are expected to review issue assigned to them and targeted for 4.2. If you believe an issue can't be resolved by mid-August / beginning of September, please let me know.
  • Discussed update of DCMTK
    • Integration of DCMTK RT components/classes into official repository will be done at project week.
  • Discussed VS2010 Support
    • Currently targeted for 4.2.0
    • Issue #2002
  • Discussed ITKv4 Support
    • Currently targeted for 4.3.0
    • Issue #2007
  • Discussed Qt 4.8.1 Support
    • Currently targeted for 4.3.0
    • Issue #2046
  • Discussed transition to Git
    • Will definitively happen before the end of they year
    • Probably for 4.3.0
    • Issue #2059
    • Before doing so, testing data will be moved into Midas. See #2060
  • Discussed directory layout of scripted module in Slicer4
    • Scripted module will be moved into dedicated directory. Will be similar to scripted extension
    • Issue: #2058
  • Discussed the layout of extensions
    • As of today, it will always be assumed a CMakeLists.txt will be available at the top level of your extension