[Edserplo_users] Update: Edteva stuff mostly

[Steve Loch]

We've been making reasonable progress in removing at least two showstoppers over the last few days although it is fair to note that in the case of Edteva functionality we're actually where I had expected to be last Christmas and not this (blame Rapid Real Time for that). This has meant we have lost a year of operational experience in this area.

1) We can now expect data edits to survive in the dump file (a definite showstopper - no pre-existing coding for this!)

2) Changes in data values as well as flag values are reflected on the PortPage (red lettering for the port).

3) The port lettering reverts to black after an intermediate save.

4) Settings file  "remembrance of change" is no longer reflected on the PortPage.

5) The speed of output of (random-access  including QXF) files on the OutputPage is now back to what it should be.

6) The inner loop of the PortTimPage has been rewritten. Large files (say 20MB) could not be plotted on Windows. They can now, with no degradation of speed (effected by removing a truly massive garbage generator!). This was another showstopper.

7) The missing cyan dot on the border of a gap has been inserted.

8) Rounding problems on date display on the PortPage have been fixed.

Quyen is getting close to completing the delivery of tooltips via the Web.  Hopefully we'll incorporate this and his other developments before Christmas. Whilst not wanting to tempt fate I haven't seen a freezing event recently (almost certainly caused by thread deadlock) one cause of which I removed previously.

Unfortunately there's still an enormous amount to do

a) The interpolation code needs looking at.

b) Stats generation needs checking.

c) Duxmod functionality needs to be checked out.

d) Tidal analysis has come a lot closer but needs to be finished 

e) The inner plotting loop of SeriesTimPage needs to be rewritten.

f) Dump files need to be hardened against class change (there's a danger that you will not be able to read the file after making a change to the software). 

g) Some progress has been made on supporting span dissection and output to a specified interval but it needs to be finished

h) Tutorial needs to be written 

Another couple of good weeks should see us give a yea to not replacing the SG maintenance licence.