Here's the Situation.
I use EasyWeather (EW) to export data in .xls format.
I then paste this data into my own SpreadSheet (See attached file)
I use this data to produce graphs and run data analysis on. (See attached file)
I have attached screen shots of a graph I produce (with anomalous data-point 18/10/14 -10:00), my Excel Data and an example of the EW.dat file which is now messed up.
The problem is.
I got a massive rain spike (18/10/14 10:00) when it wasn't raining. I've no idea why this happened.
I went into EW to try to correct the data. The only option I could find in EW (unless I'm missing something) was to wipe ALL the data off the PC. Not individual points.
I obviously didn't do this.
I opened up the EW.DAT file to have a look to see if I could simply remove the offending data-point. It looked as though EW hadn't simply recorded the raw data with a time-stamp, it actually attempted to track cumulative data and store rudimentary calculations as TEXT files!! Therefore it was impossible to simply remove the single anomalous data-point as the subsequent data in the .DAT file text would be wrong. So I didn't change or save anything.
However, when I next opened EW, it started loading thousands of non-existent records and scrambled the .dat file.
Is there any way I can rebuild the EW.DAT from within EW? or from my spreadsheet data?
Is it possible to correct the data-point from within Cumulus then rebuild the EW.DAT from the corrected data from within Cumulus?
I can't reinstall EW from scratch because all the cumulative data and totals (trends, monthy data, chronological data etc...) would be forever wrong and I'd never be able to run analysis on my data. My Spreadsheet would be trashed.
I cant simply rewrite the incorrect lines of the EW.DAT because of the insane way it stores calculations and cumulative data in TEXT format!
.I'd have to work out all the equations used by EW, re-calculate them myself and enter 1000's of lines of code by hand!
If anyone can help solve this it would save months of data and hard work! TY