ParaView 3 Telecon 05/08/2007: Difference between revisions
No edit summary |
No edit summary |
||
| (2 intermediate revisions by one other user not shown) | |||
| Line 17: | Line 17: | ||
| 6 || Moreland || [[Animation GUI]] | | 6 || Moreland || [[Animation GUI]] | ||
|} | |} | ||
== V&V Scripting == | |||
Greg Weirs is a user of PV who knows little of PV and VTK code. Greg has been the primary V&V driver. Most of our V&V request come indirectly from him. | |||
He is trying to use PV to do quantitative analysis. He wants the dynamic interaction that PV provides while being able to write scripts in the programmable filter. Once he gets the information he desires through PV, he is trying to repeat that for other data. Right now, he is using Python scripting to do so. | |||
The scripts written so far are the same basic operations done in the GUI. They use the programmable filter fairly heavily. The chalanging parts: | |||
* The pipeline/visualization model is different that what Greg and his colleagues are used to. It is difficult for him to match his problem description (e.g. pull some values out of an array) to the PV paradigms. | |||
* The multiple passes of the pipeline (REQUEST_INFORMATION, REQUEST_DATA, etc) are confusing. | |||
* PV and VTK tend to think in the global sense in that you attach filters to the entire data set. In the V&V analysis world, you are often focused on the cell level (it's values, its neighbors, etc.). | |||
Only major request: there is no direct access form the GUI to the data model (cells, points, connectivity, fields, etc.). | |||
== Deliverables == | |||
=== Better testing === | |||
Clint has started on it. He is testing and fighting bugs in the XDMF reader. | |||
They python stuff should be included. That is testing pvpython and testing the programmable filter. | |||
=== Display Rearchitecture === | |||
It goes. Right now has the superclass and architecture lain out. | |||
=== Multi-block === | |||
Dave D. is still dealing with the last few open PV 3.0 bugs. | |||
=== Exodus === | |||
Eric is also finishing up PV 3.0 bugs and some work on another project. | |||
=== Animation === | |||
Utkarsh has not been working on this. | |||
=== Selection === | |||
Not started. | |||
=== Annotation === | |||
Started adding font support for text widgets. Will check something in in a day or two. | |||
Berk is gathering ideas for labeling. | |||
=== Spreadsheet view === | |||
Done nothing. | |||
=== Line chart/bar chart === | |||
Mark started working on that. Finished the "better zoom" so that the axis are always visible. Working on getting it into PV. | |||
Dave K. also brought up that you should be able to access the chart widgets outside of PV. We all pretty much agree that it should be hosted in its own repository. | |||
=== Multiple Output === | |||
Berk has not done nothing. | |||
== Log scales == | |||
We will have a text edit next to the log checkbox that allows the user to set the value to be the minimum value to use as the left part of the range. | |||
Latest revision as of 13:35, 8 May 2007
| Item | People | Description |
|---|---|---|
| 1 | All | ParaView 3.2 Deliverables Progress |
| 2 | Weirs | V&V Scripting |
| 3 | Karelitz | Use Logarithmic Scale greyed out |
| 4 | All | Open level 4 bugs |
| 5 | Moreland | Resolved bugs |
| 6 | Moreland | Animation GUI |
V&V Scripting
Greg Weirs is a user of PV who knows little of PV and VTK code. Greg has been the primary V&V driver. Most of our V&V request come indirectly from him.
He is trying to use PV to do quantitative analysis. He wants the dynamic interaction that PV provides while being able to write scripts in the programmable filter. Once he gets the information he desires through PV, he is trying to repeat that for other data. Right now, he is using Python scripting to do so.
The scripts written so far are the same basic operations done in the GUI. They use the programmable filter fairly heavily. The chalanging parts:
- The pipeline/visualization model is different that what Greg and his colleagues are used to. It is difficult for him to match his problem description (e.g. pull some values out of an array) to the PV paradigms.
- The multiple passes of the pipeline (REQUEST_INFORMATION, REQUEST_DATA, etc) are confusing.
- PV and VTK tend to think in the global sense in that you attach filters to the entire data set. In the V&V analysis world, you are often focused on the cell level (it's values, its neighbors, etc.).
Only major request: there is no direct access form the GUI to the data model (cells, points, connectivity, fields, etc.).
Deliverables
Better testing
Clint has started on it. He is testing and fighting bugs in the XDMF reader.
They python stuff should be included. That is testing pvpython and testing the programmable filter.
Display Rearchitecture
It goes. Right now has the superclass and architecture lain out.
Multi-block
Dave D. is still dealing with the last few open PV 3.0 bugs.
Exodus
Eric is also finishing up PV 3.0 bugs and some work on another project.
Animation
Utkarsh has not been working on this.
Selection
Not started.
Annotation
Started adding font support for text widgets. Will check something in in a day or two.
Berk is gathering ideas for labeling.
Spreadsheet view
Done nothing.
Line chart/bar chart
Mark started working on that. Finished the "better zoom" so that the axis are always visible. Working on getting it into PV.
Dave K. also brought up that you should be able to access the chart widgets outside of PV. We all pretty much agree that it should be hosted in its own repository.
Multiple Output
Berk has not done nothing.
Log scales
We will have a text edit next to the log checkbox that allows the user to set the value to be the minimum value to use as the left part of the range.
