On Mon, 17 Jun 2002 10:02:40 -0500, Roy Brown <[log in to unmask]> wrote: >Thanks, Stan - though Greg gets the cigar for a solution that is *less* >characters than I was previously typing.. Greg definitely gets the cigar for finding it. I hadn't thought of using that syntax. >Your suggestion indicates that there might still be something awry with >QEdit, though.... > >Typing (when line 100 doesn't exist): > >/l 100+1 > 100.226 >RB020613 > >/l 100/100+1 > 100.217 ************************* >RB020613 > 100.226 >RB020613 > >Note that *whatever* line exists at or after 100 (100.217 in this case) >isn't listed by L 100+1 > >But L 100/100+1 shows two lines, neither of which is 100, and one of which >isn't 100+1. > >Anybody (from Robelle or elsewhere) know the logic behind this? List obviously has different behaviors depending on the line number(s) provided. I haven't looked at the actual code but here's my take on it at the moment. When the user specifies a precise line number, List insists on finding that one and nothing else. When the user specifies a relative number (100+1) or range, it uses the nearest line number. So, 1) /List 100: does not exist, does not return anything 2) /List 100+1: finds the line nearest to 100, gets 100.217, adds "+1" and returns 100.226. 3) /List 100/100+1: finds the line nearest to 100, gets 100.217, selects it as the start of range, adds "+1", gets 100.226 and uses it as the end of the range. With the planned Set List Nearest On, line 100 would *always* translate internally to 100.217. Case #1 would get 100.217. Case #3 would get 100.217 and 100.226 as it does now. If we keep the same logic for case #2, it would have to work the same way. To make things clearer for the user, we might get Qedit to display a warning: Warning: line 100 does not exist. Using 100.217. If you agree or disagreen, we are open to suggestions, opinions and ideas. Francois Desrochers Robelle Solutions Technology Inc. francois<dot>desrochers<at>robelle<dot>com * To join/leave the list, search archives, change list settings, * * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *