8:00 – 4:00 ESSO

• Tried out the test plan that JR and I put together last Friday. It went well, and she’s going to see how Vern wants it integrated into the actual test plan
• It looks like that the email-based system will be used for about 14 of the PMs until they are able to connect to the VisTool. So that means that we need to tweak that part of the tool
  • All emails for a peorson should be listed in the “cut and paste” email that’s generated. If there is no email in a particular field, then it should probably read “unset” rather than blank or some unprintable character, which is what’s happening noe.
  • August is misspelled.
  • We need to write a script that produces a table with user, project, description, status, and month (for the last 4 months maybe?) as rows. This will let us produce reports in VISIBILITY.
• Back to putting in help/guidance in FlexiChart. Done!
  • Add scrolling frame for series manipulation

8:30 – ESSO

• 10:30 Weekly Meeting
• 1:00 Test plan walk through. Well, that didn’t go well, in that it was a cut-and paste job from the manual and not a test plan at all. Will start writing an actual plan tomorrow.
• Installed Flashbuilder 4.5 from disk. That appears to have fixed my particular problem.

Instructions
use this panel to set up properties that apply to the entire chart

• X-Axis Field: Determines which row will be used to provide the labels for the X-axis in the chart. This will need to be adjusted only very rarely.
• Base at Zero. Plots values relative to zero. If more range is needed, uncheck, and results will be plotted so that the lowest value is on the X-axis
• Legend Enabled: Shows a legent identifying the chart components
• Legend Position: Lets you select between a legend placed at the bottom of the chart or to the right

}

8:00 – 4:30 ESSO

• Installed new IngestManager
• Tested Dong’s VSS compiled in 4.5 same lag, so not deployed.
• Working on adding text to Flexichart, but the IDE hangs every time it attempts an auto-complete. Going to have to re-install.
• Reinstalling had no effect. Let’s try a new workspace….
• That didn’t work either. It’s only in Flexichart that I’m having this problem. Since that’s the case, let’s try re-creating it.
• Here’s the file structure:
• 
• Ok, I’ve changed the IDE and the project, and ConfigScreen.mxml still causes eclipse to hang. Maybe it’s the com.fgm.flexichart.config package? All the pieces extend canvas, so I don’t see what the big deal is. The workaround is to edit in VI and refresh in the IDE. Lame. Utterly lame.

8:00 – 4:00 ESSO

• Backups of PPM and scripting databases. Should probably start Vis2 as well, as questions, desktops and slide shows start getting stored.
• So I brought in my test programs today, and they all worked. Yay! This also led to the main bug that was causing the whole problem. The test programs allow the endpoint to be entered from the app. When I was writing the code, I had the default string set to ‘amfsecure’, which is how I had changed all the endpoints after having problems with Visibility Scripting. When I went back to the SettingsXML file, I discovered that the endpoints had gone back to amf, rather than amfsecure. Odd. Maybe from part of a restore that I didn’t know about? Anyway, I tried vis2ff, and that worked, so the version I’m currently building (3.6 and linked projects) should hopefully work now.
• Adding names to the tooltips for the value charting accordion, and matching the formatting.
• Chased down some poor settings.xml loading tied to IE (again!)

8:00 – ESSO

• Backups and check in with Jessica
• The version of Vis2 from yesterday would not run – same problems.
• Going to try to make a new project(s) that run various tests. If they work, then I’ll add more and more of vis2 until either everything is working or we find the culprit.
• This is version 1.
• 
• V2 will have local and remote logging. Done.
• V3 will access the SetupXML. Done
• V4 will use the login functions.
• If all those work, I’ll build a new AccountManager project file and copy over the code. If that runs then doing the same for Vis2 should work.

8:00 – 4:00 ESSO

• Weekly status
• Trying out new versions of vis2
• Thursday 10:30 meeting. It looks like we’ll create a new, clean DB and keep the old one with a PPM version pointing at it. New version keeps all the meta information about contracts, people and such.
• Build yet another version of Mike’s code. This time I included as much of the original settings as I could. He appeared to have the vis2 project point at the other projects rather than the SWCs in the repo, so I’m trying that.
• Tried the Fiddler web dibugger and discovered that the message to the remote object didn’t appear to be going out. Going to try the Charles Debugger and see if there is any difference. Nope. Fiddler appears to be a bit better. Not as good as the Flashbuilder debugger though.
• Just checked and there is a getVersion() method in DbRemoteObject. If the current (re)build of Mike’s workspace fails, I’m going to try a minimum flex app that opens the remote object and calls that method. The code should be grabable from VisibilityScripting

8:00 – 2:30 ESSO

• Backups on site
• Tried the Flex 3.5B version of the code. Neither vis2 or AccountManagers worked. I find that particulalry vexing since AM was compiled entirely with 3.5B.
• Dong got XP up and running with the proper versions of IE7 and Flash player. Of course, the code works just fine. Sigh
• Trying Mike’s workspace. Was able to compile, though maven errors will have to be dealt with in the future if this works
• Added the datavisualization swc to the 3.5B version of vis2 build path. THis was probably the reason that the version yesterday would crash here. Dong has built a 3.5B and 3.6 version of AccountManagers to try as well.

2:30 – 6:00 FP

• PAD Meeting
• Wired up new mouse, and found a laser USB mouse that can be sacrificed

8:00 – 4:00 ESSO

• Ran Dong’s test SWFs, and the culprit appears to be the settings.
• Deployed the new PPM and PA
• Working on adding in the Label Field for numeric plots. Done
• Adding _labelFieldColumnName to xml save and restore. Done
• Added a tool tip to the combobox, which is not coming up. Odd. Checked in anyway. Something for Monday.
• Looks like the compiler is the culprit. Downloaded Flex 3.5 and added it to my SDK list, but didn’t install it while I was working on the labeling.

8:00 – 4:00 ESSO

• Bringing in many combinations of vis2. Let’s see if any fix the IE7  problem!
• Well, my build didn’t work, so that means that it’s probably not Maven. Dong’s going to make copy the workspace on Mike’s machine and try the following:
  • Workspace 1: Change the project settings only.
  • Workspace 2: Update the code, but not the project settings
  • Workspace 3: Update everything
  • I may also try building in his workspace to see if my version will run.
• Then I’ll try running those on the test server and see if there is any pattern.
• Looking at a way to label the _plotAccordian charts using text from another column. pieLabelFunction2 in WidgetBase should be able to look at a given column. The name of that column will have to be stored out in the xml as well.
  • looks like I can use the nameField attribute (as shown here).
• Burned a new disk of Dong’s updates for tomorrow.

8:00 – 5:00 ESSO

• Spent all day on site
• Dong’s
• Did a presentation of the Project Assistant and PPM to the project PMs
• Spent the rest of the day working on ingesting Cognos data and processing it into Flexichart-able tables
  •  Need to make the layout_xml blob larger
  • Need to fix the ‘manual’ desktop xml loading (from pasting) so that it can save
  • IE version of code still hanging on remote object access, even for Dong’s code. Ingest Manager works though, which points to a possible library issue.

8:00 – 4:00 ESSO

• Got all the scripts up and running on the production server
• Added a backup batch file for VSS. Unable to copy files in the batch file, which is odd. The network was slow today, so that could have something to do with it. Since it’s exactly the same code (that looks at its own name for behavior), what works for move should work for copy.
• Talked to JR about reports. In addition to what we have now, she’s like to be able to show the zero, non-zero, and null values on a chart. Dong’s going to look into making that query.
• Still unable to get IE to open Data Navigator. I think I might have found the answer in LoginScreen. Mike had changed the call to usersAndRoles.getUserInfo(userName.text) so that there was a boolean that said to use the DN lookup. Set that to false. This makes sense, but now I’m confused why FF logged in. Burning a disk to bring in tomorrow. Will also add Dong’s scripts