Dear J enthusiasts, I try to learn state machine parsing and try to use 4 or 5 action code in practice. Here is an example to be concrete. 1. So we have the file: $ cat test1.txt 1000ddd 2000 ab3000
11111xxx 11 22 2. now I want to ideally have three boxes, one containing 1000, 2000, 3000, the second containing 11111 and the last one containing 11, 22. 3. So I create m where 1 is LF, 2 are digits, and 0 is rest m =. a. e. LF m=: m + 2* a. e. '0123456789' m 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 2 2 2 2 2 2 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4. Now I start with defining s (firstly with undesired result but still) s =. 1 3 2 $ 0 0 1 0 2 1 s =. s , 3 2 $ 0 0 0 0 2 1 s =. s , 3 2 $ 0 3 1 3 2 0 5. Running the machine gives: d=: fread jpath './test1.txt' (0;s;m;0 _1 0 _1) ;: d ┌────┬────┬────┬─────┬──┬──┐ │1000│2000│3000│11111│11│22│ └────┴────┴────┴─────┴──┴──┘ which is as expected basing on s. Now I would like to have parser that produces something like that: ┌────────────────┬───────┬───────┐ │┌────┬────┬────┐│┌─────┐│┌──┬──┐│ ││1000│2000│3000│││11111│││11│22││ │└────┴────┴────┘│└─────┘│└──┴──┘│ └────────────────┴───────┴───────┘ Can I use 4/5 action code in s to achieve that in one state machine definition? Basically the moment that separates numbers is (row) LF -> (column) LF. Thanks ! Pawel ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
