To: <<EMAIL>>

********** U R G E N T **********

*** VHF Test Update #009: January 27, 1999 ***

I have further investigated the report of some problems with the VHF Test Editor program and have confirmed that there are data-corrupting bugs depending upon the edits you make.  I have thus far found two problems:

(1) When changing a grid and the user enters a four-character grid, the editor DOES NOT pad the grid, thus throwing off the alignment of the record and putting garbage in the file.  This has been fixed.

(2) The editor also allows times without the leading zeroes: for example an entry of "19" for "0019" is accepted, but written as "19" and, as in #1, causes a misalignment of data and makes a mess out of things.  This has also been fixed.

At this time I am unsure if there are lost QSOs.  I have not confirmed the particulars, but the one sample log I have shows corruption when changing a grid.  Much of the data for the QSO is there, although out of alignment, but the frequency is missing.  It looks to me the user determined something went wrong and relogged the same QSO three minutes later.  I base this on the last record in the backup "VT@" file and the first QSO after the junk in the active log.

One way of knowing there is corrupt data is the warning message when reloading the file that tells of incorrect record lengths.  Also, it's obvious when loading the .VT3 file in a text editor.

At this point in time I can only suggest that anyone with corrupt files via the editor make a backup copy of the file, then delete the junk records unless you can supply the missing info.  If you wish to e-mail me a copy of the log to investigate, I will see what condition things are in.  Any logs sent to me will be handled in the utmost of confidence and will be removed from my system when this problem is resolved.

I don't remember any time in VHF Test's history when a bug corrupted data, and I am disturbed about this.  I suspect the only resolution is even more testing on my part!

73, Allen, WG3E