requirements for conflict markers to be placed

[Todd Denniston]

I was hit by some users today claiming 'CVS did not mark a physical conflict'.

While looking at it, I thought I remembered (as did they) 'if there are 
changes within 10 lines of each other, that is a physical conflict'. 
Question, was this ever true as the default case?  Are there options that can 
be passed to CVS to make it true?

but with some testing 1.11.19 shows that, if it is possible for  diff to find 
one line between the two changes, then no conflict.

i.e.,
base version (both branches came from this):
 for (i=1; i < BEOP; i++)
 { // Calc CS
   CS += B[i];
 }

Branch 1 version:
 for (i=1; i < BEOP; i++)
 { // Calc CS
      if (B[i] == 0x01)
      { //skip fo 01. Nowatc!
        ++i;
      }
   CS += B[i];
 }

Branch 2 version:
 for (i=1; i < BEOP; i++)
 { // Calc CS
   CS += B[i];
      if (B[i] == 0x01)
      {
        //skip tnbais0
        i = i + 1;
      }
 }

result of setting in a Branch 1 sandbox and doing update -jBranch2:
 for (i=1; i < BEOP; i++)
 { // Calc CS
      if (B[i] == 0x01)
      { //skip fo 01. Nowatc!
        ++i;
      }
   CS += B[i];
      if (B[i] == 0x01)
      {
        //skip tnbais0
        i = i + 1;
      }
 }



Thanks for any ClueByFours, and I do have a set of scripts to generate a test 
repo showing the above if anyone wants them.

--
Todd Denniston
Crane Division, Naval Surface Warfare Center (NSWC Crane)
Harnessing the Power of Technology for the Warfighter