A script to perform a case- and variable-merge of two SPSS datasets

mergeSPSSDatasets.py

import sys
import copy
import spss

#This program takes two sets and appends the second to the first, including ALL cases and variables.
#It does this by adding the number of cases in the second set to the first set, cross-referencing and rectifying all 
#value labels, and then finally populating the newly-constructed cases in the first dataset with the data from
#the second. 

#For running outside of spss-- this gets the two datasets. 
spss.Submit("""
     GET
   FILE='D:\\binar\\Pictures\\Downloads\\dataset1.sav'.
DATASET NAME DataSet1 WINDOW=FRONT.""")
spss.Submit("""
     GET
   FILE='D:\\binar\\Pictures\\Downloads\\dataset2.sav'.
DATASET NAME DataSet2 WINDOW=FRONT.""")

#Clears python/spss environment from any previous runs
spss.EndDataStep()
#Args:
#set1,set2-- spss.Dataset objects. varname1, varname2-- String objects from spss.Variable.name field
#Purpose:
#Checks if Value Levels from SPSS dictionary are the same. If so, does nothing.
#If not, checks for conflicts. If there are conflicts, it remaps variables and levels in set2.
#In either case, it makes the value labels of varname1 the union of the labels of varname1 and varname2
def modifyVarLevels(set1, set2, varname1, varname2):
#Note-- varname* MUST be a string
    if(set1.varlist[varname1].valueLabels == set2.varlist[varname2].valueLabels):
        return None
    else:
        for i in dict.keys(set2.varlist[varname2].valueLabels):
            if i in dict.keys(set1.varlist[varname1].valueLabels):
                if(set1.varlist[varname1].valueLabels[i] != set2.varlist[varname2].valueLabels[i]):
                #if there is any value which is incongruent between sets
                    newVal = max(max(dict.keys(set1.varlist[varname1].valueLabels)),max(dict.keys(set2.varlist[varname2].valueLabels)))+1
                    #This guarantees a unique value-- what this line is doing is finding the maximum value across both sets and adding one to it, guaranteeing uniqueness
                    replaceVals(set2, set2.varlist[varname2].index, i,newVal)
                else:
                    set1.varlist[varname1].valueLabels[i] = set2.varlist[varname2].valueLabels[i]
        return None
#Args:
#setObj-- spss.Dataset object. index-- integer drawn from the "index" field of an spss.Variable object.
#oldval-- value at the index at each case to replace. All other values are ignored.
#newval-- value put in place of oldval when oldval is detected
def replaceVals(setObj,index,oldVal,newVal):
    for i in setObj.cases:
        if(i[index] == oldVal):
            i[index] = newVal
    return None
#Args: 
#newSet-- spss.Dataset object where values are transferred to. oldSet-- spss.Dataset object from where values
#are transferred
#destVar-- spss.Variable object representing column where values are transferred to. #origVar-- same, but origin of val
#startIndex-- ending of original dataset 1
#Purpose-- takes all values in the origVar column in oldSet and appends them to the destVar column in newSet starting
#at startIndex
def appendVals(newSet,oldSet,destVar,origVar,startIndex):
#Note-- startIndex must be inclusive.The first case will be written to the startindex
    for i in range(len(oldSet.cases)):
       # print(i)
        print(len(newSet.cases))
        print(len(oldSet.cases))
        newSet.cases[startIndex+i,destVar.index] = oldSet.cases[i,origVar.index]
    return None
#Args:
#set[1|2]-- spss.Dataset object. var-- spss.Variable object. endSet1-- integer representing original length of set1
#function to be executed if variable in set2 is not found by name in set1. See init and inVarFunction
def notInVarFunction(set1,set2,var,endSet1):
    print(var.label)
    for j in set2.varlist:
        print(j)
        if j.label.lower() == var.label.lower():
            inVarFunction(set1,set2,var,j,endSet1)
            return None
    set1.varlist.append(copy.deepcopy(var))
    appendVals(set1,set2,var.name,var.name,endSet1)
    return None
#set[1|2]-- spss.Dataset object. var[1|2]-- spss.Variable object. endSet1-- original length of set1
def inVarFunction(set1,set2,var1,var2,endSet1):
    appendVals(set1,set2,var1,var2,endSet1)
    return None
#Args: 
#set1-- first set (see above). set2-- second set (see above). Both spss.Dataset objects. 
def init(set1,set2):
    lengthSet1 = len(set1.cases)
    lengthSet2 = len(set2.cases)
    print("Lengths found")
    #appending relevant number of cases
    for i in range(lengthSet2):
        set1.cases.append()
    print(str(lengthSet2)+" cases appended")
    for i in set2.varlist:
        bool = False
        for j in set1.varlist:
            print(i)
            print(j)
            if i.name == j.name:
                print(i.name+" being transferred")
                inVarFunction(set1,set2,i,i,lengthSet1)
                bool = True
                break
        if bool == False:
            print(i.name+" being added")
            notInVarFunction(set1,set2,i,lengthSet1)
    return None


spss.StartDataStep()
print("Started")

#imports two datasets found above to Python
ds1 = spss.Dataset(name="DataSet1")
ds2= spss.Dataset(name="DataSet2")
print("Imported")
#runs main function
init(ds1,ds2)
#Closes out spss.DataStep object
spss.EndDataStep()