Fernando D. Bozzo wrote on 2017-02-28:
Hi Ken:
Never seen something like this, but if you are distributing an exe, then you should md5 it so you can detect any change comparing with a saved md5.
This way I've found viruses that changed parts of the code and antiviruses have cleaned them up, but leaving the exes in a different state than the original, sometimes even unrunnable.
Best regards
El 28/2/2017 4:58 p. m., "Ken Dibble" krdibble@stny.rr.com escribió:
I swear, my code is haunted.
I have some code in a subclass of my custom-built list control class. I provide enough code here to indicate the context, but I don't think I need to explain its purpose to illustrate what has just happened:
FOR m.x = 1 TO ALEN(THISFORM.oMail.aSortList,1) ** 2/28/17: ** ** On this date, a user had an "Array dimensions are invalid" ** error at this point in the code. However, the trace gave the ** offending line of code as follows: ** ** thekey = THISFORM.oMail.aSortList(m.x,10) ** ** The parent code doesn't contain any lines like that, at all. ** The array has 5 hard-coded columns, not 10. It has always ** had 5 columns since I wrote this code many years ago. ** There is NO WAY that the hard-coded scalar value of 5 ** in this code could be changed to 10. That is really, really, ** REALLY weird.
Ken,
I've seen this when I have had a second function/procedure in a PRG based class of the same name. The second one runs, not the first.
I've also seen strange things happen when classes are based off different classes with the same name from different libraries. It was something odd that way. It's been a long time.
Tracy Pearson PowerChurch Software