ability to spell check a very close plaintext to perfectly plausible plaintext about cipher-project-1

[evanrichter]

• create a branch with skeleton code

ready to go for @amp813 ! See the spell-check branch

[evanrichter]

a helpful library for various string similarity functions: strsim

I made a single word spell checker using Levenshtein "edit" distance in this dictionary method:

cipher-project-1/src/dict.rs

Lines 49 to 64 in b495829

	/// Find the closest word by Levenshtein distance.
	///
	/// Returns (dictionary_word, edit_distance)
	///
	/// The lower the score, the fewer edits needed to match the dictionary word.
	pub fn best_levenshtein(&self, word: &str) -> (&str, usize) {
	// iterate over words in dictionary
	self.words
	.iter()
	.map(|s| s.as_str())
	// create tuples of &str and the respective levenshtein distance
	.map(|s| (s, strsim::levenshtein(word, s)))
	// return the best word-score tuple
	.min_by_key(|x| x.1)
	.expect("spell correct with an empty Dictionary")
	}

[evanrichter]

@amp813 have you taken a look at the file words/plaintext_dictionary_test1.txt? It looks like we know 5 possible plaintexts in their entirety, not just the wordlist.

If test 1 must encrypt one of these five plaintexts, then you should make a special case if the near-plaintext is close to these, then pick one exactly.

I can make helper functions to parse out the sentences from this file, and make them available.