def encodeDataToWordList(data, bytesForSize, markovChain, wordsPerState=1):
# encode the data length first
lenData = utils.convertNumberToByteList(len(data), bytesForSize)
bitsField = bigBitField.BigBitField(lenData)
if wordsPerState == 1:
lastWord = config.startSymbol
elif wordsPerState == 2:
lastWord = (config.startSymbol, config.startSymbol)
lenDataCode = fixedSizeCode.encodeBitsToWordList(bitsField, markovChain,
lastWord, wordsPerState)
# compute last word (or bigram)
if wordsPerState == 1:
lastWord = lenDataCode[-1]
elif wordsPerState == 2:
if len(lenDataCode) <= 1:
lastWord = (config.startSymbol, lenDataCode[-1])
else:
lastWord = (lenDataCode[-2], lenDataCode[-1])
if lastWord[1] == config.startSymbol:
lastWord = (config.startSymbol, config.startSymbol)
# encode the actual message
bitsField = bigBitField.BigBitField(data)
mainDataCode = fixedSizeCode.encodeBitsToWordList(bitsField, markovChain,
lastWord, wordsPerState)
return lenDataCode + mainDataCode
def testExample(data, markovChain):
print "---"
print "input: " + repr(data)
miniBF = bigBitField.BigBitField(data, False)
print "output: " + repr(
fixedSizeCode.encodeBitsToWordList(miniBF, markovChain,
config.startSymbol, 1))
def decodeWordListToBits(words,
maxDigits,
markovChain,
previousWord=config.startSymbol,
wordsPerState=1):
bitsRange = ["0", "1"]
bitsField = bigBitField.BigBitField()
wordsUsed = 0
markovChainDict = utils.markovChainToDictionary(markovChain)
lastTime = time.time()
secondsForStatusPrint = 20
for word in words:
bitsRange = decodeWordToBitsRange(
word, previousWord, markovChainDict,
maxDigits - bitsField.totalFieldLen(), bitsRange)
wordsUsed = wordsUsed + 1
# compute previous word (or bigram) for next iteration
if wordsPerState == 1:
previousWord = word
elif wordsPerState == 2:
if word == config.startSymbol:
previousWord = (config.startSymbol, config.startSymbol)
else:
previousWord = (previousWord[1], word)
# simplify range, remove bits and add them to the field
bitsRange2 = utils.removeCommonBitsInRange(bitsRange)
bitsRemovedLen = len(bitsRange[0]) - len(bitsRange2[0])
if bitsRemovedLen + bitsField.totalFieldLen() > maxDigits:
bitsRemovedLen = maxDigits - bitsField.totalFieldLen()
bitsField.pushQueueNBits(bitsRange[0][0:bitsRemovedLen])
bitsRange = bitsRange2
if time.time() - lastTime > secondsForStatusPrint:
print(" - decoded bits so far: " + repr(bitsField.totalFieldLen()))
lastTime = time.time()
# we exit when our range describes only one number
if bitsField.totalFieldLen() == maxDigits:
break
if bitsField.totalFieldLen(
) == maxDigits - 1 and bitsRange[0][0] == bitsRange[1][0]:
bitsField.pushQueueNBits(bitsRange[0][0])
break
return (bitsField, wordsUsed)
break
return words
if __name__ == '__main__':
print "testing fixedSizeCode.py"
testMarkov = config.testMarkov
testMarkovDict = utils.markovChainToDictionary(testMarkov)
testMarkov2 = config.testMarkov2
testMarkovDict2 = utils.markovChainToDictionary(testMarkov2)
# this is "01000110 01011010 11111111"
testBitField = bigBitField.BigBitField([70, 90, 255])
print "A:"
print encodeBitsToWord(testBitField, ["0", "1"], config.startSymbol,
testMarkovDict) == ('A', ('00', '01'))
print "A2:"
print encodeBitsToWord(testBitField, ["0", "1"],
(config.startSymbol, config.startSymbol),
testMarkovDict2) == ('A', ('00', '01'))
print "B:"
print encodeBitsToWord(testBitField, ["0", "1"], "A",
testMarkovDict) == ('B', ('01', '01'))
print "B2:"
print encodeBitsToWord(testBitField, ["0", "1"], (config.startSymbol, "A"),
testMarkovDict2) == ('B', ('01', '01'))
print "C:"