## this script is made by Alex Travesset
## Date: July 2012

import re
import argparse

parser = argparse.ArgumentParser(description='Converts a bibtex file to html, which should be later formatted by css.',epilog='The current script only works for the article item, other entries will be added in the future. The output is intentionally ugly. The way this script is intended to work is to cut and paste the output file into your html page and format it using css styles by defining the classes for the div and span tags, see output file for the names of the different classes. If you have questions or comments contact the author: Alex Travesset. Publications in his web page: http://www.public.iastate.edu/~trvsst/Publications/publications.html where done using this script.')

parser.add_argument('file_input',metavar='file.bib',help='bibtex file to be converted to html')

parser.add_argument('-f',default='Biblio_css.html',metavar='file.html',dest='file_out',help='html output file (default = Biblio_css.html)')

parser.add_argument('-O',metavar='n',dest='format_list',type=int,nargs=6,default=range(0,6),help='Order of fields in html file, where 0=title 1=author 2=journal 3=year 4=volume 5=pages (default: 0 1 2 3 4 5)')

parser.add_argument('-S',default=3,metavar='n',dest='n_sort',type=int,choices=range(6),help='sorting criteria, where  0=title 1=author 2=journal 3=year 4=volume 5=pages (default: 3)')

parser.add_argument('-noreverse',action='store_const',default=True,const=False,metavar='',dest='rev',help='Sorting is done in reverse by defaut, -noreverse overrides the default')

args = parser.parse_args()

#define the article fields and the numpages field
article_fields=['title','author','journal','year','volume','pages','url','doi']
art_numpages='numpages'

#sort according to the specified criteria
art_field=list(article_fields)

#note that the last two fields 6,7 (url and doi remain intact)
for n in range(6):
	art_field[n]=article_fields[args.format_list[n]]

#define the sorting criteria
sort_field=article_fields[args.n_sort]

#temporal variables
art_trans=list(art_field)
add_pages=art_numpages

#define an empty list that will contain all articles 
List_art=[];

#open the bibtex file
f=open(args.file_input,'r')
#read the entire list
biblio=f.read()
f.close()

#tag to identify a reference
p_article=re.compile('[@]ARTICLE[{]')

#split by references
refs=p_article.split(biblio)

#define
p_white=re.compile('[\n]')

for ref in refs:
	for strn in art_field:
		ini_field=strn+'\s[=]'+'\s[{]'
		p_prev_stuff=re.compile(ini_field)
		if(p_prev_stuff.search(ref)):
			p_elim=re.compile('(?<='+ini_field+').+?(?=[}])',re.DOTALL)
			m=p_elim.search(ref)
			#eliminate return characters
			line=p_white.split(m.group())
			f_line=''
			for li in line:
				f_line+=li
				art_trans[art_field.index(strn)]=m.group()

	#some journals have the annoying habit of using number of pages, not final page	
	ini_field=art_numpages+'\s[=]'+'\s[{]'
	p_prev_stuff=re.compile(ini_field)
	if(p_prev_stuff.search(ref)):
		p_elim=re.compile('(?<='+ini_field+').+?(?=[}])',re.DOTALL)
		m=p_elim.search(ref)
		add_pages=m.group()
	if(add_pages!=art_numpages):
		#correct the number of pages initial page, final page
		n_str=art_trans[art_field.index('pages')]
		art_trans[art_field.index('pages')]=n_str+'-'+str(int(n_str)+int(add_pages)-1)
		add_pages=art_numpages
	#add the new bibliography to the entry
	List_art.append(list(art_trans))

#now we will fix the authors with specified format, which is Lastname, Initials;.. and Lastname, 
def Sort_author(z):
	auths=z.split('and')
	auth_tem=''
	p_coma=re.compile('[,]')
	for auth in auths:
	#if there is a coma (,) Lastname appears first, find if there is a coma
		if(p_coma.search(auth)):
			if(auths.index(auth)==0):
				name=p_coma.split(auth)
				auth_tem=name[0]+', '+re.search('\w',name[1]).group()
			elif(auths.index(auth)==len(auths)-1):
				name=p_coma.split(auth)
				auth_tem+=' and '+re.search('\w',name[1]).group()+'. '+name[0]+','
			else:
				name=p_coma.split(auth)
				auth_tem+='; '+re.search('\w',name[1]).group()+'. '+name[0]
	#if there is no coma then First Name appears first	
		else:
			name=auth.split()
			if(auths.index(auth)==0):
				auth_tem=name[len(name)-1]+', '+re.search('\w',name[0]).group()
			elif(auths.index(auth)==len(auths)-1):
                                name=auth.split()
                                auth_tem+=' and '+re.search('\w',name[0]).group()+'. '+name[len(name)-1]+','
                        else:
                                name=auth.split()
                                auth_tem+='; '+re.search('\w',name[0]).group()+'. '+name[len(name)-1]
	return auth_tem

for jj in List_art:
	jj[art_field.index(article_fields[1])]=Sort_author(jj[art_field.index(article_fields[1])])

#now we will sort according to specified key 
def SortMy(z):
	return z[art_field.index(sort_field)]

List_art.sort(key=SortMy,reverse=args.rev)


#create html file
f=open(args.file_out,'w')

str_div_op='<div class="References">'
str_div_cl='</div>'

span_list=list(art_field)
for strn in art_field:
        span_list[art_field.index(strn)]='<span class="'+strn+'">'


str_spn_cl='</span>'

str_ref='<a href="'
str_ref_end='</a>'

for ll in List_art:
	#place year in parenthesis
       	strn_d=ll[art_field.index(article_fields[3])]
	ll[art_field.index(article_fields[3])]='('+strn_d+')'
	#done with that now built add the div tag pointing to article 
	f.write(str_div_op)
	f.write('\n')
	str_new=''
        for i in range(len(art_field)-2):
		str_new+=span_list[i]+ll[i]+str_spn_cl
        f.write(str_new)
	#done with the required fields, now add the url of the paper if there isone
	if(ll[len(art_field)-2]!=art_field[len(art_field)-2]):
        	str_new=str_ref+ll[len(art_field)-2]+'">(link)'+str_ref_end
        	f.write(str_new)
	#add the end tag for article
       	f.write('\n')
       	f.write(str_div_cl)
        f.write('\n')

##program ends here