Module: Extcite

Defined in:
lib/extcite.rb,
lib/extcite/version.rb

Constant Summary collapse

VERSION = 
"0.4.1"

Class Method Summary collapse

Class Method Details

.cont_neg(ids:) ⇒ Object

Get citation(s) using Crossref content negotation

Return: an string of bib data

Examples:

require 'extcite'
Extcite.cont_neg(ids: "10.1016/j.dendro.2014.01.004")

Parameters:



305
306
307
308
 
# File 'lib/extcite.rb', line 305

def self.cont_neg(ids:)
  out = Serrano.content_negotiation(ids: ids)
  return out
end

.extract(path:, file: "out.bib", output: "bib") ⇒ Object

Extract DOIs from one or more PDFs

Return: writes bib files to a .bib file or an array if file is nil

When writing to a file, `extract` by default appends to the end
of the file so you can build up your bibtex file with your
citations

Examples:

require 'extcite'
require 'faraday'# get a paper in pdf format

path = '2068.pdf'
res = Faraday.new(:url => "https://peerj.com/articles/2068.pdf").get;
f = File.new(path, "wb");
f.write(res.body);
f.close()# extract doi from the pdf

Extcite.extract(path: path)
Extcite.extract(path: path, file: nil)

Parameters:

  • path (String) —

    Path to a pdf file, or a folder of PDF files

  • file (String) (defaults to: "out.bib") —

    File name to write data to - or nil to stdout

  • output (String) (defaults to: "bib") —

    Typeo of output. only bibtex for now



38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
 
# File 'lib/extcite.rb', line 38

def self.extract(path:, file: "out.bib", output: "bib")
  path = make_paths(path)
  path.each do |x|
    # try PDF metadata first
    ids = nil
    rr = PDF::Reader.new(x);
    pdfmeta = rr.metadata
    if !pdfmeta.nil?
      xml = Oga.parse_xml(pdfmeta);
      begin
        tt = xml.xpath('//rdf:Description')        # try dc:identifier attribute

        ss = tt.attr('dc:identifier')[0]
        if !ss.nil?
          ids = ss.text.sub(/doi:/, '')
        else
          # try prism:doi node
          pdoi = xml.xpath('//rdf:Description//prism:doi')
          if pdoi.length == 1
            ids = pdoi.text
          else
            # try pdf:WPS-ARTICLEDOI node
            wpsdoi = xml.xpath('//rdf:Description//pdf:WPS-ARTICLEDOI')
            if wpsdoi.length == 1
              ids = wpsdoi.text
            else
              # try pdfx:WPS-ARTICLEDOI node
              pdfxwpsdoi = xml.xpath('//rdf:Description//pdfx:WPS-ARTICLEDOI')
              if pdfxwpsdoi.length == 1
                ids = pdfxwpsdoi.text
              else
                ids = nil
              end
            end
          end
        end
      rescue
        ids = nil
      end
    end

    # if not found, try regexing for DOI
    if ids.nil?
      ids = Extcite.get_ids(txt: Extcite.extract_text_one(x))
    end

    if ids.length == 0
      puts "no DOI found in " + x
    else
      if !ids.match(/arxiv/i).nil? && ids.length < 200
        conn = Faraday.new(:url => 'http://export.arxiv.org/api/query?id_list=' + ids.gsub(/arxiv:/i, '')).get
        bibs = conn.body.make_bib_arxiv(ids.gsub(/arxiv:/i, ''))
      else
        bibs = Extcite.cont_neg(ids: ids)
      end

      # if an error or not found, skip
      bibstest = nil
      if bibs.class == Array
        bibstest = bibs[0]
      else
        bibstest = bibs
      end

      if !bibstest.nil?
        if !bibstest.match(/error|not found/i).nil? || !bibstest.match(/<\/html>/i).nil?
          puts "DOI found: " + ids + " ; but citation not found via content negotation - passing"          # do something else?

        else
          if file.nil?
            return bibstest
          else
            puts "writing " + ids + " to " + file
            bibs.write_bib(file)
          end
        end
      end
    end
  end
end

.extract_dois(path:) ⇒ Object

Extract DOIs from one or more PDFs after extracting text

Examples:

require 'extcite'
require 'faraday'# get a paper in pdf format

path = '2068.pdf'
res = Faraday.new(:url => "https://peerj.com/articles/2068.pdf").get
f = File.new(path, "wb")
f.write(res.body)
f.close()# extract doi from the pdf

Extcite.extract_dois(path: path)

Parameters:

  • path (String) —

    Path to a pdf file, or a folder of PDF files



219
220
221
222
223
 
# File 'lib/extcite.rb', line 219

def self.extract_dois(path:)
  txt = Extcite.extract_text(path: path)  # return txt.map { |z| z.match("[0-9]+\\.[0-9]+/.+").to_s.gsub(/\s.+/, '') }

  return Extcite.get_ids(txt: txt)
end

.extract_from_metadata(path:, try_regex: true) ⇒ Object

Try to extract DOIs from one or more PDF metadata sections

Return: DOI string

Examples:

require 'extcite'
require 'faraday'# get a paper in pdf format

path = '2068.pdf'
res = Faraday.new(:url => "https://peerj.com/articles/2068.pdf").get;
f = File.new(path, "wb");
f.write(res.body)
f.close()# extract doi from the pdf

Extcite.extract_from_metadata(path: path)

Parameters:

  • path (String) —

    Path to a pdf file, or a folder of PDF files. required

  • try_regex (true, false) (defaults to: true) —

    If no PDF metadata, try extracting using regex on the text. Uses `Extcite.extract_text_one` to extract text, then `Extcite.get_ids` to extract a DOI if any found



140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
 
# File 'lib/extcite.rb', line 140

def self.extract_from_metadata(path:, try_regex: true)
  path = make_paths(path)
  path.each do |x|
    # try PDF metadata first
    ids = nil
    rr = PDF::Reader.new(x);
    pdfmeta = rr.metadata
    if !pdfmeta.nil?
      begin
        xml = Oga.parse_xml(pdfmeta);
      rescue Exception => e
        xml = nil
      end

      if !xml.nil?
        begin
          tt = xml.xpath('//rdf:Description')          # try dc:identifier attribute

          ss = tt.attr('dc:identifier')[0]
          if !ss.nil?
            ids = ss.text.sub(/doi:/, '')
          else
            # try prism:doi node
            pdoi = xml.xpath('//rdf:Description//prism:doi')
            if pdoi.length == 1
              ids = pdoi.text
            else
              # try pdf:WPS-ARTICLEDOI node
              wpsdoi = xml.xpath('//rdf:Description//pdf:WPS-ARTICLEDOI')
              if wpsdoi.length == 1
                ids = wpsdoi.text
              else
                # try pdfx:WPS-ARTICLEDOI node
                pdfxwpsdoi = xml.xpath('//rdf:Description//pdfx:WPS-ARTICLEDOI')
                if pdfxwpsdoi.length == 1
                  ids = pdfxwpsdoi.text
                else
                  # try dc:source
                  dc_source = xml.xpath('//dc:source')
                  if dc_source.length == 1
                    ids = dc_source.text.match(/10\..+/).to_s
                  else
                    ids = nil
                  end
                end
              end
            end
          end
        rescue
          ids = nil
        end
      end
    end

    # if not found, try regexing for DOI
    if ids.nil? and try_regex
      ids = Extcite.get_ids(txt: Extcite.extract_text_one(x))
    end

    return ids
  end
end

.extract_text(path:) ⇒ Object

Extract text from a pdf, or many pdfs

This method is used internally within fetch to parse PDFs.

Examples:

require 'extcite'
require 'faraday'# get a paper in pdf format

path = '2068.pdf'
res = Faraday.new(:url => "https://peerj.com/articles/2068.pdf").get
f = File.new(path, "wb")
f.write(res.body)
f.close()# extract doi from the pdf

Extcite.extract_text(path: path)

Parameters:

  • path (String) —

    Path to a pdf file, or a folder of PDF files



272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
 
# File 'lib/extcite.rb', line 272

def self.extract_text(path:)
  path = Array(path)
  if path.length == 1
    if File.directory?(path[0])      # keep only files with .pdf extension

      path = dir_files(path[0]).keep_if { |z| !!z.match(/.pdf/) }
    end
  end

  out = []
  path.each do |x|
    begin
      rr = PDF::Reader.new(x)
      txt = rr.pages.map { |page| page.text }.join("\n")
    rescue Exception => e
      warn e
      txt = ""
    end
    out << txt
  end
  return out
end

.get_ids(txt:) ⇒ Object

Get DOIs from a String or Array of String's

Return: Array of DOIs

Examples:

require 'extcite'
Extcite.get_ids(txt: '10.1016/j.dendro.2014.01.004 adfasdf asd fas df asdfsd')

Parameters:

  • txt (String) —

    String or Array of String's



235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
 
# File 'lib/extcite.rb', line 235

def self.get_ids(txt:)
  # see if there's

  return Array(txt).map { |z|
    # detect if is an arxiv paper
    if !z.match(/arxiv:[0-9]+\.[0-9A-Za-z]+/i).nil?      # if so, return arxiv id for later extraction of arxiv citation via their API

      z = z.match(/arxiv:[0-9]+\.[0-9A-Za-z]+/i).to_s
    else
      doi_pattern = '(10[.][0-9]{4,}(?:[.][0-9]+)*/(?:(?![%"#? ])\\S)+)'
      z = z.match(doi_pattern).to_s.gsub(/\s.+/, '')      # z = z.match("10\\.[0-9]+/.+").to_s.gsub(/\s.+/, '')

    end    # clean up doi

    z = z.gsub(/\.$|\.;$|\.\]$|\.\}$|\.\)$|,$/, '')
    return z.gsub(/;$|\]$|\}$|\)$/, '')
  }[0]
end