Interesting 'Daisy Chain' problem.

[Allan]

Hi,
I have a series of tasks whose run sequence is dictated by a 'state' number. Runs can also occur in parallel. Each sequence is terminated with a 'next state' of zero.
e.g.

Code: Select all

state    next-state    Task
   1         2         Job1
   2         3         Job2
   3         0         Job3
  55        77         Job4
  77        22         Job5
  22        23         Job6
  23         0         Job7
   4         0         Job8


This would run these jobs in parallel:

Code: Select all

Job1,Job2,Job3
Job4,Job5,Job6,Job7
Job8


The daisy-chaining from one state to another is easy enough, my question is, given just this dataset, what ECL returns a list of all the start states.
e.g., in this case, the result would be 1,4,55 (I've deliberately mixed the numbers up as the numbers given to states may well not be in sequence)

There is LOOP, but I wondered if there was something a bit clever?

Yours

Allan

[rtaylor]

Allan,

Here's how I would do it:

Code: Select all

rec := {UNSIGNED1 state, UNSIGNED1 next_state, STRING4 Task};
ds := DATASET([{  1,  2, 'Job1'},
               {  2,  3, 'Job2'},
               {  3,  0, 'Job3'},
               { 55, 77, 'Job4'},
               { 77, 22, 'Job5'},
               { 22, 23, 'Job6'},
               { 23,  0, 'Job7'},
               {  4,  0, 'Job8'}],rec);
StartTbl := TABLE(ds,{ds,BOOLEAN start := FALSE});
Flagged  := ITERATE(StartTbl,
                     TRANSFORM({StartTbl},
                               SELF.start := LEFT.next_state = 0,
                               SELF := RIGHT));
Starters := PROJECT(Flagged(start = TRUE),rec);
Starters;

HTH,

Richard

[Tony Kirk]

You're essentially looking for all those not referenced in next_state, right?

Code: Select all

ds(state not in set(ds, next_state));

JOIN logic could be used, especially if ds is large.

[rtaylor]

Tony,

Nice!!

Richard

[Allan]

Thanks, Both of you.

Tony, Whatever they're paying you it's not enough.

Allan