Filling the Gaps |
At the largest conference on coding and cryptography the following theorem needed a proof or a counterexample: Suppose you are given a set of binary strings of equal length -- thus each consists of 0's, 1's and/or *'s. Furthermore suppose the pattern of *'s is different for all words in the set. By this we mean: if you replace all 0's and 1's by any single character, say $, you obtain a set of distinct strings.
The claim is: if you replace the *'s by 0's and 1's in all possible ways, then you obtain a set that is at least as big as the set you started with.
Example:
{ 10*, *0*, *00 } produces { 100, 101, 000, 001 }
{ 100, 101, 10* } produces { 100, 101 }
Notice that the set in the latter example does not satisfy the condidtion mentioned above, so it does not provide a counterexample.
You program is to help experimentally test this claim (not prove it). As such, it needs to do two things:
The input is a text-file that presents a sequence of sets. Each set is a single line of input and consists of its words delimited by spaces.
The output is a textfile that contains one line for each set, i.e., one output line for each input line. If the pattern of *'s is different for all the words in this set this line should contain YES (in uppercase), followed by a space and the number of distinct words obtained by expanding all of the *'s. Otherwise it should contain NO (in uppercase) only, with no following integer.
10* *0* *00 10** *011 *0*0 1100 1101 110* 10** *10*
YES 4 YES 7 NO