You want to form a target string of lowercase letters.

At the beginning, your sequence is target.length '?' marks. You also have a stamp of lowercase letters.

On each turn, you may place the stamp over the sequence, and replace every letter in the sequence with the corresponding letter from the stamp. You can make up to 10 * target.length turns.

For example, if the initial sequence is “?????", and your stamp is "abc", then you may make “abc??", “?abc?", “??abc” in the first turn. (Note that the stamp must be fully contained in the boundaries of the sequence in order to stamp.)

If the sequence is possible to stamp, then return an array of the index of the left-most letter being stamped at each turn. If the sequence is not possible to stamp, return an empty array.

For example, if the sequence is “ababc”, and the stamp is "abc", then we could return the answer [0, 2], corresponding to the moves “?????” -> “abc??” -> “ababc”.

Also, if the sequence is possible to stamp, it is guaranteed it is possible to stamp within 10 * target.length moves. Any answers specifying more than this number of moves will not be accepted.

Example 1:

Input: stamp = "abc", target = "ababc"
Output: [0,2]
([1,0,2] would also be accepted as an answer, as well as some other answers.)

Example 2:

Input: stamp = "abca", target = "aabcaca"
Output: [3,0,1]

Note:

1 <= stamp.length <= target.length <= 1000
stamp and target only contain lowercase letters.

Accepted 10.9K Submissions 23K

Skipping the worst case scenario trying all possible combination within the 10 stampings, I think I would start with finding the full stamp sequence first.

With the example 1, take ??abc and put the position 2 into a first in last out stack (doesn’t really matter, just keep track of the order).

Then we have ab???, which we’ll do a substring match with the stamp. We put the position 0 into the stack, and output the result: [0, 2]

This quickly becomes complex as the target.length increases. The problem limits it to 1000, and stamp.length is always smaller or equal to the target.length. Having 10 stamp limit also reduces the complexity.

Todo:

• Is finding the full stamps first a good strategy?
  Can some stamp configuration produce no solution, whilst globally a solution exists?