def _stringTimeToIntegerTimeInSeconds(self, stringTime):
integerTime = 0
# Multiplier if we are in seconds, minutes or hours
unitMultiplier = 1
# Multiplier according to the position of the figures in the number
decimalMultiplier = 1
for idxCharForward in range(len(stringTime)):
# Browse the string in reverse order
idxChar = len(stringTime)-idxCharForward - 1
char = stringTime[idxChar]
if (char >= "0" and char <= "9"):
figure = int(char)
integerTime = integerTime + \
(unitMultiplier * decimalMultiplier * figure)
# For the next step (if it is a figure and not ":" or out of range), add a power of 10
# Do not change the unit
decimalMultiplier = 10*decimalMultiplier
else:
if (char == ":"):
# Reset decimalMultiplier for next figure
decimalMultiplier = 1
# Unit will change on next figure
# Seconds to minutes and minutes to hours have the same multiplier
unitMultiplier = 60*unitMultiplier
else:
logPrint.printError(
char + " is not a figure or a comma, it is not possible to convert the time "+stringTime+" in seconds. Exiting.")
exit(-6)
return integerTime
def _findLineZero(self):
lineNb = 0
# Flag to stop the search for 0:00
isLineAtTimeZero = False
logPrint.printLog("Parsing the description field")
for line in self.descriptionLinesList:
if (not isLineAtTimeZero):
# Not find yet: perform the search
idxReturnedFromFind = line.find("0:00")
if idxReturnedFromFind != -1:
# "0:00" found. We store the line number and the character index
if idxReturnedFromFind > 0:
# Consolidate: check again if idxReturnedFromFind>0, i.e. if it is really 00:00 and not for example 10:00
isLineAtTimeZero = self._checkIfDetectedZeroIsReallyZero(
line, idxReturnedFromFind)
else:
# 0:00 at index 0 is for sure a time zero
isLineAtTimeZero = True
if isLineAtTimeZero:
# self.idxCharZeroZero = idxReturnedFromFind
self.lineNbZeroZero = lineNb
lineNb = lineNb+1
if not isLineAtTimeZero:
logPrint.printError("0:00 not found.")
return isLineAtTimeZero
def main(outputFilePath, argv):
prefix = ""
youtubeApiInst = youtubeApiWrapper(sys.argv[1])
descrFromYoutubeApi = youtubeApiInst.getDescriptionField()
# descriptionParser accepts only list of str
descrListFromYoutubeApi = descrFromYoutubeApi.splitlines()
# First attempt of chapter parsing using what Youtube API returned
descriptionParserInstFromYoutube = descriptionParser(
descrListFromYoutubeApi)
succesfullParsing = descriptionParserInstFromYoutube.parse()
if (not succesfullParsing):
logPrint.printInfo(
"Another attempt can be given using pasted chapter list.")
# Another parsing is attempted using stdin
logPrint.printInfo(
"Copy/paste the description list you want, then hit Ctrl+Z")
# returns a list of str
descrListFromCopyPaste = sys.stdin.readlines()
# strip this line from "\n"
strippedDescrListFromCopyPaste = [
"" for idx in range(len(descrListFromCopyPaste))
]
for idx in range(len(descrListFromCopyPaste)):
strippedDescrListFromCopyPaste[idx] = descrListFromCopyPaste[
idx].strip(" \n")
# Create a new descriptionParser with the pasted list stripped from the final \n
descriptionParserInstFromCopyPaste = descriptionParser(
strippedDescrListFromCopyPaste)
succesfullParsing = descriptionParserInstFromCopyPaste.parse()
if (not succesfullParsing):
logPrint.printError("Second attempt failed. Exiting.")
exit(-5)
else:
chaptersMatrix = descriptionParserInstFromCopyPaste.getChaptersMatrix(
)
else:
chaptersMatrix = descriptionParserInstFromYoutube.getChaptersMatrix()
if len(sys.argv) == 3:
prefix = sys.argv[2]
logPrint.printLog(
"The prefix \"" + prefix +
"\" read from the program's arguments, will be added before each title."
)
else:
prefix = ""
logPrint.printLog("No prefix read from the program's arguments.")
# Instantiate musicDisplayer to launch a timer and print the correct music name
musicDisplayerInst = musicDisplayer(outputFilePath, chaptersMatrix, prefix)
# Launch the timer
musicDisplayerInst.start()
def _buildPublicChaptersMatrix(self):
returnStatus = True
self.publicChaptersMatrix = np.empty(
(self.chaptersMatrixSize, 2), dtype=object)
previousTimeInSeconds = -1
for idxMatrix in range(self.chaptersMatrixSize):
timeinSeconds = self._stringTimeToIntegerTimeInSeconds(
self.chaptersMatrix[idxMatrix, 0])
if timeinSeconds < previousTimeInSeconds:
# Something went wrong and the matrix is not in crescent order
logPrint.printError(
self.chaptersMatrix[idxMatrix, 0] + " is smaller than the previous time, "+self.chaptersMatrix[idxMatrix-1, 0]+", the matrix is not ordered in a crescent order.")
returnStatus = False
self.publicChaptersMatrix[idxMatrix, 0] = timeinSeconds
self.publicChaptersMatrix[idxMatrix,
1] = self.chaptersMatrix[idxMatrix, 1]
return returnStatus
def __init__(self, message):
self.message = message + ". Exiting."
logPrint.printError(message)
exit(-2)
def _parseChapterList(self):
idxSplitted = 0
idxOfZeroInSplittedLine = -1
line = self.descriptionLinesList[self.lineNbZeroZero]
splittedLineList = re.split(r'([0-9:]+:[0-9:]+)', line)
for splittedWord in splittedLineList:
if re.search(r'[0-9:]+:[0-9:]+', splittedWord):
idxOfZeroInSplittedLine = idxSplitted
idxSplitted = idxSplitted+1
# Analyze pattern
self.patternId = self._analyzeChaptersPattern(
splittedLineList, idxOfZeroInSplittedLine)
logPrint.printLog("Pattern detected: "+str(self.patternId))
idxMatrix = 0
# To avoid false end-of-chapters due to problems of formatting in the description, another attempt is given to each non-match
numberOfRemainingAttemptsToParseEachLine = self.MAX_ATTEMPTS_NB_TO_PARSE_TIME_AND_TITLE
logPrint.printLog("While parsing the chapters, a maximum of "+str(
self.MAX_ATTEMPTS_NB_TO_PARSE_TIME_AND_TITLE)+" not matching line(s) is accepted.")
offsetForIdxMatrixBecauseOfFalseErrors = 0
# Max length of the matrix
while (idxMatrix < self.MAX_SIZE_OF_MATRIX):
# Avoid overflow
if((self.lineNbZeroZero + idxMatrix + offsetForIdxMatrixBecauseOfFalseErrors < len(self.descriptionLinesList)) and numberOfRemainingAttemptsToParseEachLine > 0):
line = self.descriptionLinesList[self.lineNbZeroZero +
idxMatrix + offsetForIdxMatrixBecauseOfFalseErrors]
if re.search('[0-9:]+:[0-9:]+', line):
# Reset for the next formatting error found
numberOfRemainingAttemptsToParseEachLine = self.MAX_ATTEMPTS_NB_TO_PARSE_TIME_AND_TITLE
# FIXME Does not work for "[2:20] the past inside the present"
splittedLine = re.split(r'([0-9:]+:[0-9:]+)', line)
# Too specific, commented and will be deleted after testing
# offsetForSplittedLineReading = 0
# if splittedLine[0] == "":
# # re.split created an empty cell at the beginning, ignore it
# offsetForSplittedLineReading = 1
# Pattern 0 could include the case where time is not the first word, the shift to do is idxOfZeroInSplittedLine
# We here take thy hypothesis that the shift of the 0:00 line is the same for every line
if self.patternId == 0:
time = splittedLine[0 + idxOfZeroInSplittedLine]
titleAndDelim = splittedLine[1 +
idxOfZeroInSplittedLine]
elif self.patternId == 1:
time = splittedLine[1]
titleAndDelim = splittedLine[0]
else:
logPrint.printError(
"Chapters pattern not found. Exiting.")
exit(-4)
logPrint.printLog("Matching line found, line "+str(self.lineNbZeroZero +
idxMatrix + offsetForIdxMatrixBecauseOfFalseErrors)+", time "+time)
# The title is stripped to not store useless characters in the beginning and in the end
title = titleAndDelim.strip(" -_[]#:")
# Strip remaining parenthesis around timestamp e.g. "(25:32) title" without cutting the included parenthesis of one title
# e.g. "(25:32) title (featuring Pitbull)"
title = title.lstrip(") ")
title = title.rstrip("( ")
self.chaptersMatrix[idxMatrix, 0] = time
self.chaptersMatrix[idxMatrix, 1] = title
# Shift the index only if we have found something to avoid holes because of formatting errors
idxMatrix = idxMatrix+1
else:
if numberOfRemainingAttemptsToParseEachLine > 0:
# Second chance
logPrint.printLog("Not matching line found, line "+str(self.lineNbZeroZero +
idxMatrix + offsetForIdxMatrixBecauseOfFalseErrors)+", another chance is given.")
numberOfRemainingAttemptsToParseEachLine = numberOfRemainingAttemptsToParseEachLine - 1
offsetForIdxMatrixBecauseOfFalseErrors = offsetForIdxMatrixBecauseOfFalseErrors + 1
else:
# No more time is detected and the formatting error margin is consumed, the chapter list is over
logPrint.printLog("No more matching line found, line "+str(self.lineNbZeroZero +
idxMatrix + offsetForIdxMatrixBecauseOfFalseErrors)+", the chapter list parsing is over.")
self.chaptersMatrixSize = idxMatrix
# End the while loop
idxMatrix = self.MAX_SIZE_OF_MATRIX
logPrint.printDebug("chaptersMatrix, " +
str(self.chaptersMatrixSize)+" lines: ")
for idxMatrix in range(self.chaptersMatrixSize):
logPrint.printDebug("|" +
self.chaptersMatrix[idxMatrix, 0]+"|"+self.chaptersMatrix[idxMatrix, 1]+"|")
)
else:
prefix = ""
logPrint.printLog("No prefix read from the program's arguments.")
# Instantiate musicDisplayer to launch a timer and print the correct music name
musicDisplayerInst = musicDisplayer(outputFilePath, chaptersMatrix, prefix)
# Launch the timer
musicDisplayerInst.start()
if __name__ == "__main__":
# Parsing config.ini
internalConfigInst = internalConfig()
logLevel = internalConfigInst.getVerbosityLevel()
outputFilePath = internalConfigInst.getOutputFilePath()
# Log level = debug
logPrint(logLevel)
logPrint.printDebug("Config.ini: logLevel=" + str(logLevel))
logPrint.printDebug("Config.ini: outputFilePath=" + outputFilePath)
if len(sys.argv) >= 2 and len(sys.argv) <= 3:
main(outputFilePath, sys.argv[1:])
else:
logPrint.printError("Usage: timeSnip.py <YoutubeURL> [prefix]")
exit(-1)