import os

tsvFile = open("Eligibility.tsv","w",encoding="utf8")

rootPath = "/Users/sriramnarayanan/Downloads/AllPublicXml/"

subFolders = os.listdir(rootPath)

subFolders = subFolders[2:]

for subFolder in subFolders:

    files = os.listdir(rootPath+subFolder)

    for fname in files:

        file = open(rootPath+subFolder+"/"+fname, encoding = "utf-8")

        fileText = file.read()

        index = fileText.find("<criteria>")

        if index == -1: #look into alternate looping methods

            file.close()

            continue

            #go to next file

        index2 = fileText[index:].find("Exclusion Criteria")

        if index2 == -1:

            index2 =0

        #keep the nct number, inclusion/exclusion criteria, item, corresponding drug/intervention    

        index2 += index

            #go to next file

        index3 = fileText[index2:].find("</textblock>")

        if index3 == -1:

            file.close()

            continue

        index3 += index2

        if index2 == index:

            index2=index3

            #go to next file

        textBlock1 = fileText[int(index):int(index2)].replace("\n","").replace("\t","") #ask if this will lose information

        textBlock2 = fileText[int(index2):int(index3)].replace("\n","").replace("\t","")

        textBlock3 = str(fname.replace(".xml",""))

        textBlock1 = textBlock1.replace("<criteria>","").replace("<textblock>","")

        textBlock2 = textBlock2.replace("<criteria>","").replace("<textblock>","")

        tsvFile.write(textBlock1+"\t") #will this lose information

        tsvFile.write(textBlock2+"\t")

        tsvFile.write(textBlock3+"\n")

        #print(index, index2, index3)

        #print("TextBlock1:"+"\n"+textBlock1+"\n")

        #print("TextBlock2:"+"\n"+textBlock2+"\n"+"\n")

        file.close()

#check if index = -1, if it does move to next file

#if else find index of next occurence of inclusion criteria

#then find index next occurrence of exclusion criteria

#then find index of next occurrence of <textblock>

#use python array indexing to 

#remove the "/ns" and "/ts" from those pieces of text

#NCT "tab" inclusion "tab" exclusion"