| 1 | #-*- coding: utf8 -*- |
|---|
| 2 | import os, re, urllib |
|---|
| 3 | |
|---|
| 4 | ## ----------------------------------------------------------------- |
|---|
| 5 | ## Purpose: make a nice publication page with an ADS database link |
|---|
| 6 | ## Author: Aymeric Spiga 19/05/2013 improvements 10-12/2013 counts 02/2014 |
|---|
| 7 | ## ----------------------------------------------------------------- |
|---|
| 8 | ## NB: uses BIBTEX2HTML https://www.lri.fr/~filliatr/bibtex2html/doc/manual.html |
|---|
| 9 | ## ... and of course NASA ADS http://adsabs.harvard.edu/ |
|---|
| 10 | ## ----------------------------------------------------------------- |
|---|
| 11 | |
|---|
| 12 | def printpubli(num): |
|---|
| 13 | if num == 1: char = "%.0f publication" % (num) |
|---|
| 14 | else: char = "%.0f publications" % (num) |
|---|
| 15 | return char |
|---|
| 16 | |
|---|
| 17 | def makepage(authorref, |
|---|
| 18 | bibstyle = "-s custom -nokeys", |
|---|
| 19 | listyear = [0], |
|---|
| 20 | customcond = None, |
|---|
| 21 | embedded = False, |
|---|
| 22 | linkads = None, |
|---|
| 23 | title = None, |
|---|
| 24 | retrieve = True, |
|---|
| 25 | addpdf = None, |
|---|
| 26 | addlink = None, |
|---|
| 27 | printnum = False, |
|---|
| 28 | verbose = True, |
|---|
| 29 | target = None): |
|---|
| 30 | |
|---|
| 31 | htmlcontent1 = "" |
|---|
| 32 | htmlcontent2 = "" |
|---|
| 33 | htmlcontent = "" |
|---|
| 34 | |
|---|
| 35 | ### HEADER |
|---|
| 36 | if embedded: |
|---|
| 37 | htmlfile = open('header.html','r') |
|---|
| 38 | htmlcontent1 = htmlfile.read() |
|---|
| 39 | htmlfile.close() |
|---|
| 40 | #else: |
|---|
| 41 | if title is None: |
|---|
| 42 | htmlcontent1 = htmlcontent1 + "<h2>"+authorref+"'s publications</h2>" |
|---|
| 43 | elif title == "": |
|---|
| 44 | pass |
|---|
| 45 | else: |
|---|
| 46 | htmlcontent1 = htmlcontent1 + title |
|---|
| 47 | |
|---|
| 48 | ### if linkads is None, we set it to "link.authorref" |
|---|
| 49 | if linkads is None: |
|---|
| 50 | linkads = authorref+'.link' |
|---|
| 51 | |
|---|
| 52 | ### GET INFO FROM ADS |
|---|
| 53 | if retrieve: |
|---|
| 54 | if verbose: print "retrieving info from ADS" |
|---|
| 55 | linkfile = open(linkads,'r') |
|---|
| 56 | url = linkfile.read() |
|---|
| 57 | linkfile.close() |
|---|
| 58 | html = urllib.urlopen(url).read() |
|---|
| 59 | ## fix problem with accents |
|---|
| 60 | find = re.compile(r"{\\'e}") |
|---|
| 61 | html = find.sub('é',html) |
|---|
| 62 | find = re.compile(r"{\\`e}") |
|---|
| 63 | html = find.sub('è',html) |
|---|
| 64 | ## |
|---|
| 65 | bibfile = open(linkads+'.bib','w') |
|---|
| 66 | print >> bibfile,html |
|---|
| 67 | bibfile.close() |
|---|
| 68 | |
|---|
| 69 | ### if only one year and no customcond, make it useful. ask for years >= this value |
|---|
| 70 | if len(listyear) == 1 and customcond is None: |
|---|
| 71 | customcond = "-c 'year>=%s'" % (listyear[0]) |
|---|
| 72 | listyear[0] = 99 |
|---|
| 73 | |
|---|
| 74 | ### YEAR LOOP |
|---|
| 75 | numpublitot = 0 ; nonzeroyears = [] |
|---|
| 76 | for year in listyear: |
|---|
| 77 | |
|---|
| 78 | author = authorref+str(year) |
|---|
| 79 | |
|---|
| 80 | # 0. define condition |
|---|
| 81 | # if not user-defined, make it simply year in each listyear instance |
|---|
| 82 | # if user-defined, then customcond will be the condition (possibly several) |
|---|
| 83 | if customcond is None and len(listyear) > 1: cond = "-c 'year=%s'" % (year) |
|---|
| 84 | elif len(listyear) > 1: cond = customcond + " -c 'year=%s'" % (year) |
|---|
| 85 | else: cond = customcond |
|---|
| 86 | |
|---|
| 87 | # 1. select items ARTICLE in the big bib file |
|---|
| 88 | # put those in a dedicated author.bib file |
|---|
| 89 | arg = \ |
|---|
| 90 | cond,\ |
|---|
| 91 | '"ARTICLE"',\ |
|---|
| 92 | author+'.txt',\ |
|---|
| 93 | author+'.bib',\ |
|---|
| 94 | linkads+'.bib' |
|---|
| 95 | cmd = "bib2bib --quiet %s -c '$type=%s' -oc %s -ob %s %s >> /dev/null 2>> /dev/null" % (arg) |
|---|
| 96 | os.system(cmd) |
|---|
| 97 | |
|---|
| 98 | # count number of publications (both per year and increment total) |
|---|
| 99 | numpubli = len(open(author+'.txt', 'r').readlines()) |
|---|
| 100 | if verbose: print "%s --> count: %.0f" % (author,numpubli) |
|---|
| 101 | numpublitot += numpubli |
|---|
| 102 | # record years with publications |
|---|
| 103 | if numpubli == 0: continue |
|---|
| 104 | else: nonzeroyears.append(year) |
|---|
| 105 | |
|---|
| 106 | # modify the bib file to insert pdf links |
|---|
| 107 | # the trick is to use the line adsurl and expect pdf to have the same name as ADS reference |
|---|
| 108 | # ... then besides this, it is necessary to link pdfs or rename those |
|---|
| 109 | # ... the online repository is indicated by addpdf |
|---|
| 110 | if retrieve: |
|---|
| 111 | if addpdf is not None: |
|---|
| 112 | bibcontent = '' |
|---|
| 113 | for line in open(linkads+'.bib'): |
|---|
| 114 | bibcontent += line |
|---|
| 115 | if 'adsurl' in line: |
|---|
| 116 | line = line.replace('adsurl','localpdf') |
|---|
| 117 | line = line.replace('http://cdsads.u-strasbg.fr/abs/',addpdf) |
|---|
| 118 | line = line.replace('},','.pdf},') |
|---|
| 119 | line = line.replace('%','_') |
|---|
| 120 | bibcontent += line |
|---|
| 121 | bibfile2 = open('temp','w') |
|---|
| 122 | print >> bibfile2,bibcontent |
|---|
| 123 | bibfile2.close() |
|---|
| 124 | os.system('mv temp '+linkads+'.bib') |
|---|
| 125 | |
|---|
| 126 | # 2. make the html page from the author.bib file |
|---|
| 127 | if customcond is None or len(listyear) > 1: |
|---|
| 128 | header = '<a name="%.0f"></a>' % (year) |
|---|
| 129 | header += "<h3>%.0f <a href=''>.</a> </h3>" % (year) |
|---|
| 130 | if printnum: header += "("+printpubli(numpubli)+")" |
|---|
| 131 | if embedded: header += '<br>' |
|---|
| 132 | else: |
|---|
| 133 | header = '' |
|---|
| 134 | if printnum: header += "("+printpubli(numpubli)+")" |
|---|
| 135 | |
|---|
| 136 | header = '"'+header+'"' |
|---|
| 137 | arg = \ |
|---|
| 138 | bibstyle,\ |
|---|
| 139 | header,\ |
|---|
| 140 | author+'.bib' |
|---|
| 141 | cmd = "bibtex2html -q \ |
|---|
| 142 | --both \ |
|---|
| 143 | -m ads.tex \ |
|---|
| 144 | %s \ |
|---|
| 145 | -nf adsurl 'ADS link' \ |
|---|
| 146 | -nf localpdf 'PDF version' \ |
|---|
| 147 | -r -d --revkeys \ |
|---|
| 148 | -nofooter --nodoc \ |
|---|
| 149 | --header %s -nokeywords \ |
|---|
| 150 | %s >> /dev/null 2>> /dev/null" % (arg) |
|---|
| 151 | os.system(cmd) |
|---|
| 152 | |
|---|
| 153 | # 3. load page content and delete intermediate HTML file |
|---|
| 154 | htmlfile = open(author+'.html','r') |
|---|
| 155 | htmlcontent = htmlcontent + htmlfile.read() |
|---|
| 156 | htmlfile.close() |
|---|
| 157 | os.system("rm -rf "+author+'.html') |
|---|
| 158 | |
|---|
| 159 | ## make a few corrections |
|---|
| 160 | ## bibcontent = open(author+'.bib','r').read() |
|---|
| 161 | ## bibcontent.replace('\grl','Yeah') |
|---|
| 162 | |
|---|
| 163 | find = re.compile(r'bib') |
|---|
| 164 | htmlcontent = find.sub('Bibtex entry',htmlcontent) |
|---|
| 165 | find = re.compile(r'Bibtex entry.html') |
|---|
| 166 | htmlcontent = find.sub('bib.html',htmlcontent) |
|---|
| 167 | |
|---|
| 168 | find = re.compile(r'DOI') |
|---|
| 169 | htmlcontent = find.sub('Journal website',htmlcontent) |
|---|
| 170 | |
|---|
| 171 | #find = re.compile(r'.pdf') |
|---|
| 172 | #htmlcontent = find.sub('PDF version',htmlcontent) |
|---|
| 173 | |
|---|
| 174 | find = re.compile(r'<table>') |
|---|
| 175 | htmlcontent = find.sub('<table border="0" cellspacing="15">',htmlcontent) |
|---|
| 176 | find = re.compile(r'<td align="right">') |
|---|
| 177 | htmlcontent = find.sub('<td align="center" width=17% style="font-size: 75%;">',htmlcontent) |
|---|
| 178 | |
|---|
| 179 | htmlcontent += '''<hr><p>Generated with |
|---|
| 180 | <a href='https://www.lri.fr/~filliatr/bibtex2html/doc/manual.html'>BibTeX2HTML</a> |
|---|
| 181 | and <a href='http://adsabs.harvard.edu/'>NASA ADS</a> |
|---|
| 182 | and a bit of <a href='http://www.python.org/'>Python</a></p>''' |
|---|
| 183 | if embedded: |
|---|
| 184 | htmlfile = open('footer.html','r') |
|---|
| 185 | htmlcontent += htmlfile.read() |
|---|
| 186 | htmlfile.close() |
|---|
| 187 | |
|---|
| 188 | ### TREAT HEADER PART DEPENDENT ON PREVIOUS LOOP |
|---|
| 189 | ### -- total publications + list of years |
|---|
| 190 | if customcond is None or len(listyear) > 1: |
|---|
| 191 | if len(nonzeroyears) == 0: |
|---|
| 192 | htmlcontent2 += "No publications found." |
|---|
| 193 | else: |
|---|
| 194 | htmlcontent2 += "Year: " |
|---|
| 195 | for year in nonzeroyears: |
|---|
| 196 | htmlcontent2 += "<a href='#"+str(year)+"'>"+str(year)+"</a>. " |
|---|
| 197 | if addlink is not None: htmlcontent2 += "<br>"+addlink |
|---|
| 198 | if printnum: htmlcontent2 += "<p>(Total: "+printpubli(numpublitot)+")</p>" |
|---|
| 199 | |
|---|
| 200 | ### PUT EVERYTHING TOGETHER AND CREATE A PAGE |
|---|
| 201 | htmlcontent = htmlcontent1 + htmlcontent2 + htmlcontent |
|---|
| 202 | htmlmain = open(authorref+'.html','w') |
|---|
| 203 | print >> htmlmain, htmlcontent |
|---|
| 204 | htmlmain.close() |
|---|
| 205 | |
|---|
| 206 | ## move results to target directory and remove txt files |
|---|
| 207 | if target is not None: |
|---|
| 208 | target=target+"/" |
|---|
| 209 | arg = target,\ |
|---|
| 210 | authorref+"*.html",\ |
|---|
| 211 | target,\ |
|---|
| 212 | authorref+"*.bib",\ |
|---|
| 213 | authorref+"*.txt",\ |
|---|
| 214 | target+linkads+".bib",\ |
|---|
| 215 | "*.css",\ |
|---|
| 216 | target |
|---|
| 217 | os.system( "mkdir -p %s ; mv %s %s ; rm -rf %s %s ; mv %s ./ 2> /dev/null ; cp %s %s 2> /dev/null" % (arg) ) |
|---|
