Affine ciphers

From: Neil Smith Date: Wed, 21 May 2014 17:19:22 +0000 (+0100) Subject: Fixed the merge conflict X-Git-Url: https://git.njae.me.uk/?p=cipher-training.git;a=commitdiff_plain;h=ee8bb0b7953e32c988f4faa72c0eb4c6cd6e4ea8 Fixed the merge conflict --- ee8bb0b7953e32c988f4faa72c0eb4c6cd6e4ea8 diff --cc slides/keyword-encipher.html index e96de86,0000000..7258e9c mode 100644,000000..100644 --- a/slides/keyword-encipher.html +++ b/slides/keyword-encipher.html @@@ -1,187 -1,0 +1,198 @@@ + + + + Affine ciphers + + + + +

+
 +# Keyword ciphers
 +
 +a | b | c | d | e | f | g | h | i | j | k | l | m | n | o | p | q | r | s | t | u | v | w | x | y | z
 +--|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|--
 +k | e | y | w | o | r | d | a | b | c | f | g | h | i | j | l | m | n | p | q | s | t | u | v | x | z
 +
 +* Taking a more Pythonic approach
 +
 +---
 +
 +# The cipher
 +
 +* Still character-by-character substitution, still monosubstitution.
 +
 +Ciphertext alphabet: start with a keyword, write out the rest of the alphabet, removing duplicates.
 +
 +## Three variants
 +
 +Write out the rest of the alphabet...
 +
 +1. ...starting from 'a' (keywordabcf...)
 +2. ...starting from the last letter of the keyword (keywordfgh...)
 +3. ...starting from the largest letter of the keyword (keywordzabc...)
 +
 +---
 +
 +# A more Pythonic way
 +
 +_string_`.translate()` and _string_`.maketrans()`
 +
 +* Make the 'ciphertext' alphabet, relate to the 'plaintext' alphabet (`string.ascii_lowercase`)
 +* Use those to make the translation table
 +* Enciphering is simply applying `plaintext.translate(enciphering_table)`
 +* Deciphering just uses a different table
 +
 +---
 +
 +# Making the cipher alphabet from a keyword
 +
 +Three challenges:
 +
 +1. How to say which type of cipher alphabet to use
 +2. Where to start the rest of the alphabet
 +3. Removing duplicate letters
 +
 +Solutions:
 +
 +1. Keyword arguments for procedures
 +2. sort and slices
 +3. Use something like an ordered set
 +
 +Both enciphering and deciphering need the same keyword-based alphabet, so pull this out into another procedure.
 +
 +---
 +
 +# Keyword arguments
 +
 +Used to:
 +1. give a default value for a parameter
 +2. allow parameters to be named (not just positional)
 +
 +Give our `keyword_encipher` and `keyword_decipher` procedures a keyword parameter of `wrap_alphabet`.
 +
 +Pass this parameter to the `keyword_alphabet` procedure.
 +
 +## Note: `Enum` introduced in Python 3.4. This is a better solution. 
 +
 +---
 +
 +# Deduplicating a sequence
 +
 +We need
 +
 +* Something set-like
 +* Something ordered
 +
 +No ordered set in Python, but do have an ordered dict.
 +
 +* Keys of a dict are a set. 
 +* Keys in an ordered dict retain their order (subsequent instances are ignored)
 +
 +`deduplicated_list = list(collections.OrderedDict.fromkeys(list))`
 +
 +---
 +
 +# Sorts and slices
 +
 +## Recap 
 +Write out the rest of the alphabet...
 +
 +1. ...starting from 'a' (keywordabcf...)
 +2. ...starting from the last letter of the keyword (keywordfgh...)
 +3. ...starting from the largest letter of the keyword (keywordzabc...)
 +
 +* Santitise the keyword before we use it
 +
 +## Cases
 +1. As we're deduplicating anyway, just add the entire alphabet to the end of the keyword, then deduplicate. 
 +`deduplicate(keyword + string.ascii_lowercase)`
 +
 +2. and 3. How to find the appropriate letter of the keyword.
 +
 +Indexing pulls out letters. `'keyword'[0]` = 'k' ; `'keyword'[3]` = 'w' ; `'keyword'[-1]` = 'd'
 +Slices pulls out substrings. `'keyword'[1:4]` = 'eyw' ; `'keyword'[:3]` = 'key' ; `'keyword'[5:]` = 'rd'
 +
 +`deduplicate(keyword + string_ascii_lowercase[from:] + string.ascii_lowercase)`
 +
 +Question: why not take a slice of the second alphabet copy?
 +
 +Question: what do we use as the last letter of 'character'? 'r' or 'e'?
 +
 +`sorted()` will put a string in lexicographical order.
 +`.find()` will find an item in a sequence
 +
++---
++
++# Keyword enciphering
++
++Now we've got the keyword-based cipher alphabet, simply use `.translate()` to do the enciphering/deciphering.
++
++Use `''.maketrans()` to make the translation table.
++
++Sorted!
++
++Does it pass the tests?
 +
 +

+ + + + + + +