Anonymous user
Word break problem: Difference between revisions
→Functional: improve.
(→{{header|Javascript}}: Updated tokenTrees to match the Python version) |
(→Functional: improve.) |
||
Line 745:
=={{header|Python}}==
===Functional===
{{improve|Python|Reads like machine translated Haskell rather than idiomatic Python}}
The '''tokenTrees''' function recursively builds a tree of possible token sequences, using a list monad ('''concatMap''' with a function which returns its result wrapped in a list – an empty list where a parse has failed) to discard all branches which lead to dead ends. This allows us to return more than one possible word-break parse for a given lexicon and input string. (Searches for 'monadic parsing in Python' will yield references to more sophisticated uses of this general approach).
| |||