Frequency distribution of processes

sameFunctions.py

import csv
import sys

csv.field_size_limit(sys.maxsize)
#allBlocks.csv is from https://github.com/TUDelftScratchLab/ScratchDataset
with open('allBlocks.csv','r') as rawcsv:
    reader = csv.DictReader(rawcsv,fieldnames =[
        'projectID', 'scriptCoordinates', 'scriptIndex',
        'stageSpriteindication', 'spriteName', 'indentation',
        'blockIndex', 'blockType', 'parameters'])
    largeMsgs = []

    procDict = {}
    #for each block
    for (i,row) in enumerate(reader):
        #if block is the first block of a process
        if (row['stageSpriteindication']=='procDef' and row['blockIndex']=='0'):
            #this project interests us (has process)
            if row['projectID'] in procDict:
                #actually row['spriteName'] is the name of the process
                #if the same process appears many times it must be 
                #in different sprites (can't tell from csv but it's safe to assume)
                if row['spriteName'] in procDict[row['projectID']]:
                    procDict[row['projectID']][row['spriteName']] += 1
                else:
                    procDict[row['projectID']][row['spriteName']] = 1
            else:
                procDict[row['projectID']] = {row['spriteName']:1}
freqDict = {}
for projectID in procDict:
    for process in procDict[projectID]:
        processAppearances = procDict[projectID][process]
        if processAppearances in freqDict:
            freqDict[processAppearances] += 1
        else:
            freqDict[processAppearances] = 1

for freq in sorted(freqDict.keys()):
    print(";".join([str(freq),str(freqDict[freq])]))