#!/usr/bin/env python
# -*- coding: utf-8 -*-
#A program to parse DrReposER webserver
import sys, os, re, requests, ast, optparse

"""
Description of DrReposER webserver:
The webserver consists of 3 interfaces;

interface1 = "Search for amino acid arrangements similar to known drug binding sites in known target structures / protein-drug complexes"
interface2 = "Search for drug binding interfaces in protein-drug complexes"
interface3 = "Search a protein structure for amino acid residue arrangements similar to a known drug binding site "
--------------------------------------------------------
FIRST_INTERFACE (Similar patterns of amino acids)
--------------------------------------------------------
1) Search by PDB ID ('PDB_ID'):
General view page--> http://27.126.156.175/drreposed/get_by_assam.php?query_id=PDB_ID&query_assam=1eqc&submit=Submit
Results in tabulated form--> http://27.126.156.175/drreposed/drreposer_man_assam.php?myresult=PDB_ID,1eqc

2) Search by Ligand ID ('HETATM_RESIDUES'):
General view page--> http://27.126.156.175/drreposed/get_by_assam.php?query_id=HETATM_RESIDUES&query_assam=CTS&submit=Submit
Results in tabulated form--> http://27.126.156.175/drreposed/drreposer_man_assam.php?myresult=HETATM_RESIDUES,CTS

--------------------------------------------------------
SECOND_INTERFACE (Known binding sites)
--------------------------------------------------------
1) Search by PDB ID ('pdb'):
General view page--> http://27.126.156.175/drreposed/get_by_pdbid.php?pdbid=1mxd&submit=Submit
Results in tabulated form--> http://27.126.156.175/drreposed/drreposer_man_pdb.php?myresult=1mxd

2) Search by Ligand ID ('ligand'):
General view page--> http://27.126.156.175/drreposed/get_by_ligandid.php?ligandid=ACR&submit=Submit
Results in tabulated form--> http://27.126.156.175/drreposed/drreposer_man_ligand.php?myresult=ACR

3) Search by Drug ('drug'):
General view page--> http://27.126.156.175/drreposed/get_by_drugbankid.php?drugbankid=acarbose&submit=Submit
Results in tabulated form--> http://27.126.156.175/drreposed/drreposer_man_drug.php?myresult=ACARBOSE

4) Search by Keyword('keywords'):
a) Drug Indication (Results in tabulated form): http://27.126.156.175/drreposed/drreposer_man_keywords.php?myresult=query_indication,diabetes
b) Source Organism (Results in tabulated form): http://27.126.156.175/drreposed/drreposer_man_keywords.php?myresult=query_organism,Aspergillus%20awamori
c) Macromolecule name (Results in tabulated form): http://27.126.156.175/drreposed/drreposer_man_keywords.php?myresult=query_macromoleculename,GLUCOAMYLASE-471
d) Pfam annotation (Results in tabulated form): http://27.126.156.175/drreposed/drreposer_man_keywords.php?myresult=query_pfamannotation,7tm_1

--------------------------------------------------------
THIRD_INTERFACE (Search for potential motifs in a protein structure)
--------------------------------------------------------

1) Upload a PDB file
2) Or enter a 4-letter PDB ID

"""

def get_html_text(urll):
	r = requests.get(urll)
	return r.text.encode('utf-8')

def get_results_list(urll):
	try:
		r_text = get_html_text(urll)
		r_text_post = re.findall(r"<textarea name='export_data' style='display: none;'>.*</textarea>",r_text)
		export_data_value = r_text_post[0].split(">")[1].split("<")[0].replace('null','"null"')
		export_data_value = ast.literal_eval(export_data_value)
		return export_data_value
	except:
		return "Error! Either no matches found or could not generate results. "

def FIRST_INTERFACE(args):
	#e.g. 'PDB_ID,1eqc' or 'HETATM_RESIDUES,CTS'
	if len(args.split(","))==2:
		query_id=args.split(",")[0]
		query_term=args.split(",")[1]
		if query_id =="PDB_ID" or query_id =="HETATM_RESIDUES":
			urll="http://27.126.156.175/drreposed/drreposer_man_assam.php?myresult={},{}".format(query_id,query_term)
			return get_results_list(urll)
		else: return "Error! First argument must be 'PDB_ID' or HETATM_RESIDUES."
	else: return "Error! First interface requires two arguments joined by comma. e.g. 'PDB_ID,1eqc' or 'HETATM_RESIDUES,cts'\nSearch for PDB ID: 'PDB_ID,$query'\nSearch for Ligand ID: 'HETATM_RESIDUES,$query'"

def SECOND_INTERFACE(args):
	#e.g. 'pdbid,1mxd' or 'ligandid,ACR' or 'drug,acarbose' or 'keywords,drugindication,diabetes'
	if len(args.split(","))==2:
		query_id=args.split(",")[0]
		query_term=args.split(",")[1]
		#if query_id != "keywords":
		if query_id =="pdb" or query_id =="ligand" or query_id =="drug":
			urll="http://27.126.156.175/drreposed/drreposer_man_{}.php?myresult={}".format(query_id,query_term)
			return get_results_list(urll)
		if query_id =="keywords":
			return "Search by keywords requires three arguments joined by comma. e.g. 'keywords,indication,diabetes'"
		else: return "First argument must be 'pdb', 'ligand', 'drug' or 'keywords'."

	if len(args.split(","))==3:
		query_id=args.split(",")[0] #keywords
		if query_id=="keywords":
			query_category=args.split(",")[1]
			query_term=args.split(",")[2]
			urll="http://27.126.156.175/drreposed/drreposer_man_keywords.php?myresult=query_{},{}".format(query_category,query_term)
			return get_results_list(urll)
		else:
			return "First argument for 'Search by keywords' must be 'keywords'"
	else: return "error! Second interface requires two or three arguments joined by comma. e.g. pdbid,1mxd' or 'ligand,ACR' or 'drug,acarbose' or 'keywords,indication,diabetes'\n" \
				 "Search for PDB ID: 'pdb,$query'\nSearch for Ligand ID: 'ligand,$query'\nSearch for Drug: 'drug,$query'\nSearch for Keywords:\n" \
				 "	Drug indication-->'keywords,indication,$query'\n	Organism-->'keywords,organism,$query'\n	Macromolecule-->'keywords,macromoleculename,$query'\n	Pfam annotation-->'keywords,pfamannotation,$query'"

def THIRD_INTERFACE(args):
	if len(args)==4:
		urll = "http://27.126.156.175/drreposed/sprite_res/execute_sprite.php"
		response=requests.post(urll,{'get':args}, allow_redirects=True)
		rposttext=response.text.encode('utf-8')
		if len(rposttext)>300:
			rpost_link=re.findall(r"http://27.126.156.175/drreposed/sprite_res/result_sprite.php[?]id[=].*",rposttext)[0].split("<")[0]
			rpost_result_file=re.findall(r"http://27.126.156.175/drreposed/sprite_res/result/.*/results.txt",rposttext)[0]
			return rpost_link, rpost_result_file
		else: return rposttext
	else: return "Error! PDB ID must be a four-letter code."

if __name__ == '__main__': # 13- PARSE COMMAND LINE ARGUMENTS
	usage = "--USAGE: python template_drreposer.py -i interface (first,second,third) -a args -o output file (optional)\n\n" \
"1) First interface requires two arguments joined by comma. e.g. 'PDB_ID,1eqc' or 'HETATM_RESIDUES,cts'\nSearch for PDB ID: 'PDB_ID,$query' || Search for Ligand ID: 'HETATM_RESIDUES,$query'\n\n\n" \
"2) Second interface requires two or three arguments joined by comma. e.g. pdbid,1mxd' or 'ligand,ACR' or 'drug,acarbose' or 'keywords,indication,diabetes'\n" \
"Search for PDB ID: 'pdb,$query'\nSearch for Ligand ID: 'ligand,$query'\nSearch for Drug: 'drug,$query'\nSearch for Keywords:\n" \
"	Drug indication-->'keywords,indication,$query'\n	Organism-->'keywords,organism,$query'\n	Macromolecule-->'keywords,macromoleculename,$query'\n	Pfam annotation-->'keywords,pfamannotation,$query'\n\n\n" \
"3) Third interface requires a PDB ID in four-letter code. e.g. '11gs'"
	parser = optparse.OptionParser(usage=usage)
	parser.add_option("-i", "--interface", dest="interface", type=str, help="Interface")
	parser.add_option("-a", "--args", dest="args", type=str, help="Arguments")
	parser.add_option("-o", "--outputfile", dest="output", type=str, help="Output text file")
	(options, args) = parser.parse_args()

	if (options.interface and options.args):
		if (options.output):
			output = open(options.output,"wb")
			if options.interface=="first":
				results=FIRST_INTERFACE(options.args)
				if isinstance(results, list):
					for res in results: print>>output, res
				else: print results
			if options.interface=="second":
				results=SECOND_INTERFACE(options.args)
				if isinstance(results, list):
					for res in results: print>>output, res
				else: print results
			if options.interface=="third":
				results=THIRD_INTERFACE(options.args)
				if isinstance(results, list):
					for res in results: print>>output, res
				else: print results

		if not (options.output):
			if options.interface=="first":
				results=FIRST_INTERFACE(options.args)
				if isinstance(results, list):
					for res in results: print res
				else: print results
			if options.interface=="second":
				results=SECOND_INTERFACE(options.args)
				if isinstance(results, list):
					for res in results: print res
				else: print results
			if options.interface=="third":
				results=THIRD_INTERFACE(options.args)
				if isinstance(results, list):
					for res in results: print res
				else: print results
	else:
		print "Not enough arguments supplied"
		print usage
		quit()

#print FIRST_INTERFACE('PDB_ID,1eqc')
#print FIRST_INTERFACE('HETATM_RESIDUES,CTS')
#print SECOND_INTERFACE('pdb,1mxd')
#print SECOND_INTERFACE('ligand,ACR')
#print SECOND_INTERFACE('drug,acarbose')
#print SECOND_INTERFACE('keywords,indication,parkinson')
#print SECOND_INTERFACE('keywords,organism,candida albicans')
#print SECOND_INTERFACE('keywords,macromoleculename,phosphodiesterase')
#print SECOND_INTERFACE('keywords,pfamannotation,PDEase_I')
#print THIRD_INTERFACE('11gs')