import rrEval
import os

acc = [] #[sep6, sep12, sep24] values for each sequence separation see rrEval.py for more details
cov = [] #[sep6, sep12, sep24] values for each sequence separation see rrEval.py for more details

trueDir = os.getcwd() + "/true" #change to where the .RR files are located (empty dir except .RR files)
predDir = os.getcwd()+ "/svmIndividual" #change to the predictions. empty dir except for prediction files. Must start with 5 character name matching the .RR file name beginning
seqDir = os.getcwd() + "/seq"
L = ((1)) #multiplies the sequence length which determines the num of preds evaluated. 
out = open('svm_plusReg_L.txt', 'w')

trueFiles = []
scores = [] #list of lists of [acc6, acc12, acc24, cov6, cov12, cov24]
seqsAndSizes = []# [targetID, target length in residues]

for file in os.listdir(trueDir):
    if file.endswith('.RR'):
        trueFiles.append(trueDir + "/" + file)
    
for file in os.listdir(seqDir): #directory target .fasta files (no headers. only sequence. one line)
    if file.endswith('.fasta'):
        check = open(seqDir + "/" + file, 'r')
        n = ''
        for line in check:
            if line[0] == '>':
                n = line[1:6]
                continue
            
            seqsAndSizes.append([n, len(line)]) #targetID, L

for file in os.listdir(predDir):
    name = file[:5] #the target name
    try:
        tf = [x for x in trueFiles if name in x]
        tf = tf[0]
        proLen = [x[1] for x in seqsAndSizes if name in x][0]

        scores.append(rrEval.main([tf, (predDir + "/" + file), name, int(L * proLen)]))
        scores[-1].append(name)
    except:
	print name + " didn't work\n"
        continue

#calc the averages of the scores array
numOfTargets = len(scores)
sums = [0, 0, 0, 0, 0, 0]


for x in scores:
    out.write( "-----------------------------------------------\n")
    out.write(str(x[6]) + "\n")

    out.write("acc6: " + str(x[0]) + "\n") #acc6
    out.write("acc12: " + str(x[1]) + "\n") #acc12
    out.write("acc24: " + str(x[2]) + "\n" + "\n") #acc24

    out.write("cov6: " + str(x[3]) + "\n") #cov6
    out.write("cov12: " + str(x[4]) + "\n") #cov12
    out.write("cov24: " + str(x[5]) + "\n") #cov24
    out.write("-----------------------------------------------\n")

    sums[0] = sums[0] + x[0]
    sums[1] = sums[1] + x[1]
    sums[2] = sums[2] + x[2]
    sums[3] = sums[3] + x[3]
    sums[4] = sums[4] + x[4]
    sums[5] = sums[5] + x[5]

sums[0] = sums[0]/numOfTargets
sums[1] = sums[1]/numOfTargets
sums[2] = sums[2]/numOfTargets
sums[3] = sums[3]/numOfTargets
sums[4] = sums[4]/numOfTargets
sums[5] = sums[5]/numOfTargets

out.write( "evaluation complete on: " + str(numOfTargets) + " prediction files\n")
out.write ("averages: [acc6, acc12, acc24, cov6, cov12, cov24]"+"\n")
out.write(str(sums))