Blog from October, 2009

Work Completed

  • Continued working on the table model design, added basic support for long running loading of large shapefiles. After making several changes to the table model, I fixed the drag and drop support for the Data Catalog table viewer for both shapefiles and basic delimited text files stored as beans.
  • Started working on new GIS Map View for viewing shapefile data that is stored as beans. I will continue this work next week.

Work Planned

  • Work on Data Catalog table model design, keep all tabular data types current as the design changes
  • Work on GIS Map View for viewing shapefile data that is stored as beans in the Data Catalog

Comments

This week went as planned. I took half of a personal day on Friday October 23, 2009. I will be in Champaign on Monday October 26, 2009 for the MAEviz Lunch.

Work Completed

  • Reviewed the Statement of Work (SOW) and offered input on the SOW we submitted this week to Argonne for the TRANSMIS/TRANSviz project
  • Worked on the design of the Data Catalog table model. I am adding support so models that should be able to display tabular data (e.g. shapefile dbf, delimited text files, etc) can be dropped onto a table view and display their contents. I am also working on allowing table models to handle sorting for operations where a Database backed dataset might want to handle sorting inside the database and pass the results back to the view.
  • Meeting with Shawn using Google Video Chat to discuss the table model design and meeting minutes from the Monday staff/budget meeting I missed.
  • Can Unen found a bug where the Turkish version of MAEviz has an issue running an analysis, but the English version does not. I asked him to file a bug report and then Amr Elnashai will have to determine if it is a bug that we address for them. With a potential project with KISTI, we may end up addressing the issue if that project receives funding since this could simply be an internationalization bug.

Work Planned

  • Work on Data Catalog table model design.
  • Review and Comment on TRANSIMS/TRANSviz Statement of Work

Comments

I took a personal day on Monday October 12, 2009. This week went as planned. Next week I will continue working on the Data Catalog and anything that might arise for TRANSIMS.

Work Completed

  • Started pulling together MAEviz Jira entries and text between October 1, 2008 and the present for the final report to the MAE Center. This work is nearly finished and will need some contributions from Jong for the work he has done. This document should be finished when the final MAE features/bug fixes are completed. I believe the target finish date is October 15, 2009.
  • Worked on Data Catalog with Shawn. Some unit tests and classes had to be updated to conform to API changes and test new features.

Work Planned

  • Pull together final report for MAE Center
  • Work on Data Catalog

Comments

This week went as planned. I will be taking a personal day on October 12, 2009.

Work Completed

  • Exchanged emails with Can Unen who is still having trouble running the MAEviz pipeline analysis. I successfully ran the latest daily build at the time he said he was having an issues so I will have to debug further by email with him. I asked him for more information about his setup, including which daily build he was using. I also requested that he unzip and run the daily build as-is instead of using an old batch file launcher file since I tweaked the build to run without a batch file. I believe the batch file could be corrupting the launch by loading files from an old MAEviz installation. I have not heard back from him since my last email.
  • Finished adding basic tabular data support (some more changes may be required as the API changes), including wizard pages to ingest tabular data and the appropriate UI actions to interpret the physical model to a logical model. Added a TableViewer class with drag and drop support for the tabular logical model with backing of Nat Table and Glazed Lists. Also added a sample right click menu that is only active for tabular data.
  • Video Chat with Shawn and Jong to discuss possible KISTI projects.

Work Planned

  • Add Logical and Physical Models for tabular data to the data catalog
  • Add drag and drop support from the data catalog to an eclipse view that can understand tabular data
  • Add right click menus declaratively to the data catalog that operate on tabular data models

Comments

This week went as planned. I took a personal day on October 1, 2009. Next week will be more work on the data catalog and contacting Can to see if he is still having problems with the Pipeline Analysis.