//
// A simple bibtex duplicates checking tool - just for @articles 
// Input has to be pre-parsed by the bibparse program from the BibTooL package http://cgi.di.uoa.gr/~charnik/oss/bibtool/#downloads
//
// Usage: bibparse file1.bib [file2.bib ...] |tr '{}' '  '| bdc
//
//TODO: would be good to treat arXiv entries - but different people define them differently

/*      Version 0.3 Copyright (C) 2015 Jiri Pittner <jiri@pittnerovi.com>
 *     
 *      This program is free software: you can redistribute it and/or modify
 *      it under the terms of the GNU General Public License as published by
 *      the Free Software Foundation, either version 3 of the License, or
 *      (at your option) any later version.
 *    
 *      This program is distributed in the hope that it will be useful,
 *      but WITHOUT ANY WARRANTY; without even the implied warranty of
 *      MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *      GNU General Public License for more details.
 *      You should have received a copy of the GNU General Public License
 *      along with this program.  If not, see <http://www.gnu.org/licenses/>.
 *      
 */

#include <iostream>
#include <iomanip>
#include <list>
#include <string>
#include <vector>
#include <string>
#include <algorithm>
#include <iterator>
#include <sstream>
#include <sys/types.h>
#include <unistd.h>
#include <stdlib.h>
#include <string.h>
//#include <boost/foreach.hpp>
//#include <boost/tokenizer.hpp>
#include <boost/algorithm/string.hpp>


using namespace std;

class ITEM
	{
public:
	bool deleted;
	string label;
	string author;
	string title;
	};

class ARTICLE : public ITEM
	{
public:
	string journal;	
	int volume;
	int page;
	int year;
	void clear(void) {deleted=0; volume=0; page=0; year=0; author=""; journal="";}
	};

istream& operator>>(istream &s, ARTICLE &a)
{
string line;
a.clear();
while(getline(s,line))
        {
        if(s.eof()) exit(1);
	vector<string> fields;
	boost::split(fields, line, boost::is_any_of("="), boost::token_compress_off);
	if(fields.size() != 2) break;
	if(fields[0] == "volume") a.volume=atoi(fields[1].c_str());
	if(fields[0] == "pages") a.page=atoi(fields[1].c_str());
	if(fields[0] == "year") a.year=atoi(fields[1].c_str());
	if(fields[0] == "journal") a.journal=fields[1];
	if(fields[0] == "author") a.author=fields[1];
	if(fields[0] == "title") a.title=fields[1];
	}

return s;
}

ostream& operator<<(std::ostream &s, const ARTICLE &a)
{
s <<a.label<<": "<<a.author<<" "<<a.journal<<" "<< " "<<a.volume<<" "<<a.page<<" "<<a.year;
return s;
}


int main(int argc, char **argv)
{
//read all items from standard input
list<ARTICLE> articles;

string line;
while(getline(cin,line))
	{
	if(cin.eof()) exit(0);		
	vector<string> fields;
	boost::split(fields, line, boost::is_any_of(" "), boost::token_compress_off);
	//cout << fields.size()<<" "<<fields[0]<<" "<<line<<endl;
	if(fields[0] == "@article" && fields.size() == 2) 
		{
		ARTICLE tmp;
		tmp.label=fields[1];
		//cout <<tmp.label<<endl;
		cin >> tmp;
		//cout <<tmp;
		articles.push_front(tmp);
		}
	}


//check pairwise among all articles (not deleted yet)
list<ARTICLE>:: iterator i,j;
for(i=articles.begin(); i != articles.end(); ++i)
	{
	for(j=articles.begin(); j != i; ++j)
	   if(!j->deleted)
		{
		if(i->label == j->label ) cout << "DUPLICATE LABEL::"<<i->label<<endl;
		if(i->year == j->year && i->page == j->page && i->volume == j->volume && i->year!=0 && i->page!=0 && i->volume!=0 )
			{
			string sure= j->journal == i->journal ? "" : "SUSPECT";
			cerr<< sure<<"DUPLICATE @ARTICLE:: " << *i << " & "<<*j<<endl;
			list<ARTICLE>:: iterator ii,jj;
			ii=i; jj=j; //normally keep the first one and delete later ones
			if(j->title!="" && i->title=="") {ii=j; jj=i;} //but delete preferably the entry with less information (missing title)
			cout << "\\bibalias{"<<jj->label<<"}{"<<ii->label<<"} %remove " <<jj->label<<endl;
			jj->deleted=1; //indicate deletion to reasonably handle triplicates and higher
			}
		}
	}

}