def generateJSONForD3(filemanager, mergedSet):
"""
Generates the data formatted nicely for the d3 visualization library.
Args:
mergedSet: Boolean saying whether to merge all files into one dataset or, if false,
create a list of datasets.
Returns:
An object, formatted in the JSON that d3 needs, either a list or a dictionary.
"""
chosenFileIDs = [int(x) for x in request.form.getlist('segmentlist')]
activeFiles = []
if chosenFileIDs:
for ID in chosenFileIDs:
activeFiles.append(filemanager.files[ID])
else:
for lFile in filemanager.files.values():
if lFile.active:
activeFiles.append(lFile)
if mergedSet: # Create one JSON Object across all the chunks
minimumLength = int(request.form['minlength']) if 'minlength' in request.form else 0
masterWordCounts = {}
for lFile in activeFiles:
wordCounts = lFile.getWordCounts()
for key in wordCounts:
if len(key) <= minimumLength:
continue
if key in masterWordCounts:
masterWordCounts[key] += wordCounts[key]
else:
masterWordCounts[key] = wordCounts[key]
if 'vizmaxwords' in request.form:
maxNumWords = int(request.form['maxwords'])
sortedwordcounts = sorted(masterWordCounts, key=masterWordCounts.__getitem__)
j = len(sortedwordcounts) - maxNumWords
for i in xrange(len(sortedwordcounts) - 1, -1, -1):
if i < j:
del masterWordCounts[sortedwordcounts[i]]
returnObj = general_functions.generateD3Object(masterWordCounts, objectLabel="tokens", wordLabel="name",
countLabel="size")
else: # Create a JSON object for each chunk
returnObj = []
for lFile in activeFiles:
returnObj.append(lFile.generateD3JSONObject(wordLabel="text", countLabel="size"))
return returnObj # NOTE: Objects in JSON are dictionaries in Python, but Lists are Arrays are Objects as well.
def generateD3JSONObject(self, wordLabel, countLabel):
"""
Generates a JSON object for d3 from the word counts of the file.
Args:
wordLabel: Label to use for identifying words in the sub-objects.
countLabel: Label to use for identifying counts in the sub-objects.
Returns:
The resultant JSON object, formatted for d3.
"""
wordCounts = self.getWordCounts()
return general_functions.generateD3Object(wordCounts, self.label, wordLabel, countLabel)
def generateJSONForD3(filemanager, mergedSet):
"""
Generates the data formatted nicely for the d3 visualization library.
Args:
mergedSet: Boolean saying whether to merge all files into one dataset or, if false,
create a list of datasets.
Returns:
An object, formatted in the JSON that d3 needs, either a list or a dictionary.
"""
chosenFileIDs = [int(x) for x in request.form.getlist('segmentlist')]
activeFiles = []
if chosenFileIDs:
for ID in chosenFileIDs:
activeFiles.append(filemanager.files[ID])
else:
for lFile in filemanager.files.values():
if lFile.active:
activeFiles.append(lFile)
if mergedSet: # Create one JSON Object across all the chunks
minimumLength = int(request.form['minlength']) if 'minlength' in request.form else 0
masterWordCounts = {}
for lFile in activeFiles:
wordCounts = lFile.getWordCounts()
for key in wordCounts:
if len(key) <= minimumLength:
continue
if key in masterWordCounts:
masterWordCounts[key] += wordCounts[key]
else:
masterWordCounts[key] = wordCounts[key]
if 'vizmaxwords' in request.form:
# Make sure there is a number in the input form
checkForValue = request.form['maxwords']
if checkForValue == "":
maxNumWords = 100
else:
maxNumWords = int(request.form['maxwords'])
sortedwordcounts = sorted(masterWordCounts, key=masterWordCounts.__getitem__)
j = len(sortedwordcounts) - maxNumWords
for i in xrange(len(sortedwordcounts) - 1, -1, -1):
if i < j:
del masterWordCounts[sortedwordcounts[i]]
returnObj = general_functions.generateD3Object(masterWordCounts, objectLabel="tokens", wordLabel="name",
countLabel="size")
else: # Create a JSON object for each chunk
returnObj = []
for lFile in activeFiles:
returnObj.append(lFile.generateD3JSONObject(wordLabel="text", countLabel="size"))
return returnObj # NOTE: Objects in JSON are dictionaries in Python, but Lists are Arrays are Objects as well.
