import sys
from fuzzywuzzy import fuzz
from fuzzywuzzy import process

#arguments test : lectures_maj.txt structures.txt output.txt

def stop_err( msg ):
    sys.stderr.write( msg )
    sys.exit()

def reverse_strand(strand):
	#Given a RNA strand string, it returns its reversed strand
	rev_strand=[]
	i=1
	for base in strand :
		if base == "A" :
			rev_strand.append("U")
		if base == "U" :
			rev_strand.append("A")
		if base == "G" :
			rev_strand.append("C")
		if base == "C" :
			rev_strand.append("G")
		i=i+1
	return str(rev_strand[::-1]).replace(',','').replace("'","").replace('[','').replace(']','').replace(' ','')
	
	
def __main__():

	#converting input files into maps

	try:
		f_1=open(sys.argv[1])
	except:
		stop_err( 'Cannot open a file\n' )
	lect_maj={}
	for line in f_1:
		line=line.replace("\n","")
		if ">" in line :
			key=line[5:] 
		else :
			lect_maj[key]=line
	f_1.close()

	try:
		f_2=open(sys.argv[2])
	except:
		stop_err( 'Cannot open a file\n' )
	structures={}
	for line in f_2:
		line=line.replace("\n","")
		if ">" in line :
			key=line.replace(">","")
			structures[key]=[]
		else :
			structures[key].append(line)
	f_2.close()

	#for each structure, it returns the reads that are included at 75% or more in the structure
	f_output=open(sys.argv[3],"w")
	scores_list=[]
	for i in structures.keys():
		struct=structures[i][0]
		for j in lect_maj.keys() :
			lect=lect_maj[j]
			similarity_score_forward=fuzz.partial_ratio(lect,struct)
			similarity_score_reverse=fuzz.partial_ratio(reverse_strand(lect),struct)
			if similarity_score_forward > similarity_score_reverse :
				similarity_score=similarity_score_forward
			else :
				similarity_score=similarity_score_reverse
			if similarity_score>= 75 :
				line="%s\t%s\t%s\t%s\n"%(i,j,similarity_score,lect)
				f_output.write(line)
				scores_list.append(similarity_score)
	f_output.close()
	
	f_output_infos=open("informations.txt","w")
	f_output_infos.write("%s miRNA reads are found to be good candidates of hairpin structures.\n"%len(scores_list))
	f_output_infos.write("Some miRNA may be selected multiple times if they are good candidates for more than one hairpin structure.\n%s miRNA were tested with %s hairpin structures.\n"%(len(lect_maj.keys()),len(structures.keys())))
	f_output_infos.write("The minimal similarity score is %s.\n"%min(scores_list))
	f_output_infos.write("The maximal similarity score is %s and is found %s times.\n"%(max(scores_list),scores_list.count(max(scores_list))))
	f_output_infos.write("The average similarity score is %s."%(sum(scores_list)/len(scores_list)))
	f_output_infos.close()


if __name__ == "__main__":
    __main__()