alexandria-book-collection-manager 0.7.5 → 0.7.9

data/lib/alexandria/book_providers/pseudomarc.rb

@@ -41,9 +41,9 @@ module Alexandria
       notes: ["520", "a"]
     }.freeze
 
-    def self.get_fields(data, type, stripping, m = USMARC_MAPPINGS)
+    def self.get_fields(data, type, stripping, mappings = USMARC_MAPPINGS)
       field = ""
-      m[type][1..m[type].length - 1].each do |part|
+      mappings[type][1..mappings[type].length - 1].each do |part|
         if data.first[part]
           part_data = data.first[part].strip
           if part_data =~ stripping
@@ -58,69 +58,68 @@ module Alexandria
       field
     end
 
-    def self.marc_text_to_book(marc, m = USMARC_MAPPINGS)
+    def self.marc_text_to_book(marc, mappings = USMARC_MAPPINGS)
       details = marc_text_to_details(marc)
-      unless details.empty?
-        title = nil
-        title_data = details[m[:title][0]]
-        if title_data
-          title_data_all = get_fields(title_data, :title, %r{(.*)[/:]$}, m)
-          title = title_data_all if title_data_all
-        end
+      return if details.empty?
 
-        authors = []
-        author_data = details[m[:authors][0]]
-        author_data&.each do |ad|
-          author = ad[m[:authors][1]]
-          if author
-            author = author.strip
-            author = Regexp.last_match[1] if author =~ /(.*),$/
-            authors << author
-          end
-        end
+      title = nil
+      title_data = details[mappings[:title][0]]
+      if title_data
+        title_data_all = get_fields(title_data, :title, %r{(.*)[/:]$}, mappings)
+        title = title_data_all if title_data_all
+      end
 
-        isbn = nil
-        binding = nil
-        isbn_data = details[m[:isbn][0]]
-        if isbn_data
-          isbn = Regexp.last_match[1] if isbn_data.first[m[:isbn][1]] =~ /([-0-9xX]+)/
+      authors = []
+      author_data = details[mappings[:authors][0]]
+      author_data&.each do |ad|
+        author = ad[mappings[:authors][1]]
+        if author
+          author = author.strip
+          author = Regexp.last_match[1] if author =~ /(.*),$/
+          authors << author
         end
+      end
 
-        binding_data = details[m[:binding][0]]
-        if binding_data
-          if binding_data.first[m[:binding][1]] =~ /([a-zA-Z][a-z\s]+[a-z])/
-            binding = Regexp.last_match[1]
-          end
-        end
+      isbn = nil
+      binding = nil
+      isbn_data = details[mappings[:isbn][0]]
+      if isbn_data && isbn_data.first[mappings[:isbn][1]] =~ /([-0-9xX]+)/
+        isbn = Regexp.last_match[1]
+      end
 
-        publisher = nil
-        publisher_data = details[m[:publisher][0]]
-        publisher = publisher_data.first[m[:publisher][1]] if publisher_data
+      binding_data = details[mappings[:binding][0]]
+      if binding_data &&
+          binding_data.first[mappings[:binding][1]] =~ /([a-zA-Z][a-z\s]+[a-z])/
+        binding = Regexp.last_match[1]
+      end
 
-        year = nil
-        publication_data = details[m[:year][0]]
-        if publication_data
-          year = publication_data.first[m[:year][1]]
-          year = Regexp.last_match[1].to_i if year =~ /(\d+)/
-        end
+      publisher = nil
+      publisher_data = details[mappings[:publisher][0]]
+      publisher = publisher_data.first[mappings[:publisher][1]] if publisher_data
 
-        notes = ""
-        notes_data = details[m[:notes][0]]
-        notes_data&.each do |note|
-          txt = note[m[:notes][1]]
-          notes += txt if txt
-        end
+      year = nil
+      publication_data = details[mappings[:year][0]]
+      if publication_data
+        year = publication_data.first[mappings[:year][1]]
+        year = Regexp.last_match[1].to_i if year =~ /(\d+)/
+      end
 
-        if title.nil? && isbn.nil?
-          # probably didn't undertand the MARC dialect
-          return nil
-        end
+      notes = ""
+      notes_data = details[mappings[:notes][0]]
+      notes_data&.each do |note|
+        txt = note[mappings[:notes][1]]
+        notes += txt if txt
+      end
 
-        book = Alexandria::Book.new(title, authors, isbn,
-                                    publisher, year, binding)
-        book.notes = notes unless notes.empty?
-        book
+      if title.nil? && isbn.nil?
+        # probably didn't undertand the MARC dialect
+        return nil
       end
+
+      book = Alexandria::Book.new(title, authors, isbn,
+                                  publisher, year, binding)
+      book.notes = notes unless notes.empty?
+      book
     end
 
     def self.marc_text_to_details(marc)
@@ -132,31 +131,22 @@ module Alexandria
 
           this_line_data = {}
 
-          # puts code
-          # puts data
           d_idx = 0
           while d_idx < data.size
-            d_str = data[d_idx..-1]
-            # puts d_str
-            if (idx = d_str =~ /\$([a-z]) ([^\$]+)/)
-              # puts idx
-              sub_code = Regexp.last_match[1]
-              sub_data = Regexp.last_match[2]
-              this_line_data[sub_code] = sub_data
-              # puts "  " + $1
-              # puts "    " + $2
-              # puts idx
-              d_idx += idx + 2 # (2 extra to push beyond this '$a' etc.)
-            else
-              break
-            end
+            d_str = data[d_idx..]
+            idx = d_str =~ /\$([a-z]) ([^$]+)/
+            break unless idx
+
+            sub_code = Regexp.last_match[1]
+            sub_data = Regexp.last_match[2]
+            this_line_data[sub_code] = sub_data
+            d_idx += idx + 2 # (2 extra to push beyond this '$a' etc.)
           end
 
           unless this_line_data.empty?
             details[code] = [] unless details.key?(code)
             details[code] << this_line_data
           end
-
         end
       end
       details

data/lib/alexandria/book_providers/sbn_provider.rb

@@ -0,0 +1,108 @@
+# frozen_string_literal: true
+
+# This file is part of Alexandria.
+#
+# See the file README.md for authorship and licensing information.
+
+require "alexandria/book_providers/z3950_provider"
+
+module Alexandria
+  class BookProviders
+    class SBNProvider < Z3950Provider
+      # http://sbnonline.sbn.it/
+      # http://it.wikipedia.org/wiki/ICCU
+      unabstract
+
+      include GetText
+      GetText.bindtextdomain(Alexandria::TEXTDOMAIN, charset: "UTF-8")
+
+      def initialize
+        super("SBN", "Servizio Bibliotecario Nazionale (Italy)")
+        prefs.variable_named("hostname").default_value = "opac.sbn.it"
+        prefs.variable_named("port").default_value = 3950
+        prefs.variable_named("database").default_value = "nopac"
+        # supported 'USMARC', 'UNIMARC' , 'SUTRS'
+        prefs.variable_named("record_syntax").default_value = "USMARC"
+        prefs.variable_named("charset").default_value = "ISO-8859-1"
+        prefs.read
+      end
+
+      def url(book)
+        "http://sbnonline.sbn.it/cgi-bin/zgw/BRIEF.pl?displayquery=" \
+        "%253CB%253E%253Cfont%2520color%253D%2523000064%253E" \
+        "Codice%2520ISBN%253C%2FB%253E%253C%2Ffont%253E%2520" \
+        "contiene%2520%2522%2520%253CFONT%2520COLOR%253Dred%253E" +
+          canonicalise_isbn_with_dashes(book.isbn) +
+          "%253C%2FFONT%253E%2522&session=&zurl=opac" \
+          "&zquery=%281%3D7+4%3D2+2%3D3+5%3D100+6%3D1+3%3D3+%22" +
+          canonicalise_isbn_with_dashes(book.isbn) +
+          "%22%29&language=it&maxentries=10&target=0&position=1"
+      rescue StandardError => ex
+        log.warn { "Cannot create url for book #{book}; #{ex.message}" }
+        nil
+      end
+
+      private
+
+      def canonicalise_criterion(criterion, _type)
+        canonicalise_isbn_with_dashes(criterion)
+      end
+
+      def request_count(_type)
+        0
+      end
+
+      def canonicalise_isbn_with_dashes(isbn)
+        # The reference for the position of the dashes is
+        # http://www.isbn-international.org/converter/ranges.htm
+
+        isbn = Alexandria::Library.canonicalise_isbn(isbn)
+
+        if isbn[0..1] == "88"
+          # Italian speaking area
+          if isbn > "8895000" && (isbn <= "8899999996")
+            isbn[0..1] + "-" + isbn[2..6] + "-" + isbn[7..8] + "-" + isbn[9..9]
+          elsif isbn > "88900000"
+            isbn[0..1] + "-" + isbn[2..7] + "-" + isbn[8..8] + "-" + isbn[9..9]
+          elsif isbn > "8885000"
+            isbn[0..1] + "-" + isbn[2..6] + "-" + isbn[7..8] + "-" + isbn[9..9]
+          elsif isbn > "886000"
+            isbn[0..1] + "-" + isbn[2..5] + "-" + isbn[6..8] + "-" + isbn[9..9]
+          elsif isbn > "88200"
+            isbn[0..1] + "-" + isbn[2..4] + "-" + isbn[5..8] + "-" + isbn[9..9]
+          elsif isbn > "8800"
+            isbn[0..1] + "-" + isbn[2..3] + "-" + isbn[4..8] + "-" + isbn[9..9]
+          else
+            raise _("Invalid ISBN")
+          end
+
+        else
+          isbn
+        end
+      end
+      #
+      # Remarks about SBN
+      #
+      # This provider requires that value of conn.count is 0.
+      # It's a Yaz option "Number of records to be retrieved".
+      # This provider requires to specify the value of conn.element_set_name = 'F'.
+      # It's a Yaz option "Element-Set name of records".
+      # See http://www.indexdata.dk/yaz/doc/zoom.resultsets.tkl
+      #
+      # Dashes:
+      # this database requires that Italian books are searched with dashes :(
+      # However, they have also books with dashes in wrong positions, for
+      # instance 88-061-4934-2
+      #
+      # References:
+      # http://opac.internetculturale.it/cgi-bin/main.cgi?type=field
+      # http://www.internetculturale.it/
+      # http://sbnonline.sbn.it/zgw/homeit.html
+      # http://www.iccu.sbn.it/genera.jsp?id=124
+      # with link at http://www.iccu.sbn.it/upload/documenti/cartecsbn.pdf
+      # http://www.loc.gov/cgi-bin/zgstart?ACTION=INIT&FORM_HOST_PORT=/prod/www/data/z3950/iccu.html,opac.sbn.it,2100
+      # http://gwz.cilea.it/cgi-bin/reportOpac.cgi
+      #
+    end
+  end
+end

data/lib/alexandria/book_providers/{thalia.rb → thalia_provider.rb}

@@ -1,27 +1,13 @@
 # frozen_string_literal: true
 
-# Copyright (C) 2009 Cathal Mc Ginley
-# Copyright (C) 2014 Matijs van Zuijlen
+# This file is part of Alexandria.
 #
-# Alexandria 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 2 of the
-# License, or (at your option) any later version.
-#
-# Alexandria 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 Alexandria; see the file COPYING.  If not,
-# write to the Free Software Foundation, Inc., 51 Franklin Street,
-# Fifth Floor, Boston, MA 02110-1301 USA.
+# See the file README.md for authorship and licensing information.
 
 # http://de.wikipedia.org/wiki/Thalia_%28Buchhandel%29
 # Thalia.de bought the Austrian book trade chain Amadeus
 
-# New Tlalia provider, taken from Palatina MetaDataSource and modified
+# New Thalia provider, taken from Palatina MetaDataSource and modified
 # for Alexandria. (21 Dec 2009)
 
 require "net/http"
@@ -31,9 +17,9 @@ require "alexandria/book_providers/web"
 module Alexandria
   class BookProviders
     class ThaliaProvider < WebsiteBasedProvider
-      include Alexandria::Logging
+      include Logging
 
-      SITE = "http://www.thalia.de"
+      SITE = "https://www.thalia.de"
       BASE_SEARCH_URL = "#{SITE}/shop/bde_bu_hg_startseite/suche/?%s=%s" # type,term
 
       def initialize
@@ -48,7 +34,7 @@ module Alexandria
 
       def search(criterion, type)
         req = create_search_uri(type, criterion)
-        puts req if $DEBUG
+        log.debug { req }
         html_data = transport.get_response(URI.parse(req))
         if type == SEARCH_BY_ISBN
           parse_result_data(html_data.body, criterion)
@@ -80,40 +66,36 @@ module Alexandria
       def parse_search_result_data(html)
         doc = html_to_doc(html)
         book_search_results = []
-        results_divs = doc / "div.articlePresentationSearchCH"
-        results_divs.each do |div|
+
+        results_items = doc / "ul.weitere-formate li.format"
+
+        results_items.each do |item|
           result = {}
-          title_link = div % "div.articleText/h2/a"
-          result[:title] = title_link.inner_html
-          result[:lookup_url] = title_link["href"]
+          item_link = item % "a"
+          result[:lookup_url] = "#{SITE}#{item_link['href']}"
           book_search_results << result
         end
         book_search_results
       end
 
       def data_from_label(node, label_text)
-        label_node = node % "strong[text()*='#{label_text}']"
-        if (item_node = label_node.parent)
-          data = ""
-          item_node.children.each do |n|
-            data += n.to_html if n.text?
-          end
-          data.strip
-        else
-          ""
-        end
+        label_node = node % "th[text()*='#{label_text}']"
+        return "" unless label_node
+
+        item_node = label_node.parent % "td"
+        item_node.inner_text.strip
       end
 
       def get_book_from_search_result(result)
         log.debug { "Fetching book from #{result[:lookup_url]}" }
         html_data = transport.get_response(URI.parse(result[:lookup_url]))
-        parse_result_data(html_data.body, "noisbn", true)
+        parse_result_data(html_data.body, "noisbn", recursing: true)
       end
 
-      def parse_result_data(html, isbn, recursing = false)
+      def parse_result_data(html, isbn, recursing: false)
         doc = html_to_doc(html)
 
-        results_divs = doc / "div.articlePresentationSearchCH"
+        results_divs = doc / "ul.weitere-formate"
         unless results_divs.empty?
           if recursing
             # already recursing, avoid doing so endlessly second time
@@ -122,73 +104,54 @@ module Alexandria
             return
           end
 
-          # ISBN-lookup results in multiple results (trying to be
-          # useful, such as for new editions e.g. 9780974514055
-          # "Programming Ruby" )
+          # ISBN-lookup results in multiple results
           results = parse_search_result_data(html)
-          isbn10 = Library.canonicalise_isbn(isbn)
-          # e.g. .../dave_thomas/ISBN0-9745140-5-5/ID6017044.html
           chosen = results.first # fallback!
-          results.each do |rslt|
-            if rslt[:lookup_url] =~ %r{/ISBN(\d+[\d-]*)/}
-              if Regexp.last_match[1].delete("-") == isbn10
-                chosen = rslt
-                break
-              end
-            end
-          end
           html_data = transport.get_response(URI.parse(chosen[:lookup_url]))
-          return parse_result_data(html_data.body, isbn, true)
+          return parse_result_data(html_data.body, isbn, recursing: true)
         end
 
         begin
-          if (div = doc % "div#contentFull")
-            title_img = ((div % :h2) / :img).first
-            title = title_img["alt"]
-
-            # note, the following img also has alt="von Author, Author..."
+          if (div = doc % "section#sbe-product-details")
+            title = div["data-titel"]
 
-            if (author_h = doc % 'h3[text()*="Mehr von"]') # "More from..." links
+            if (author_p = doc % "p.aim-author")
               authors = []
-              author_links = author_h.parent / :a
+              author_links = author_p / :a
               author_links.each do |a|
-                if a["href"] =~ %r{BUCH/sa}
-                  # 'sa' means search author, there may also be 'ssw' (search keyword) links
-                  authors << a.inner_text[0..-2].strip
-                  # NOTE stripping the little >> character here...
-                end
+                authors << a.inner_text.strip
               end
             end
 
-            item_details = doc % "ul.itemDataList"
+            item_details = doc % "section.artikeldetails"
             isbns = []
             isbns << data_from_label(item_details, "EAN")
             isbns << data_from_label(item_details, "ISBN")
+            isbns.reject!(&:empty?)
 
             year = nil
-            date = data_from_label(item_details, "Erschienen:")
-            year = Regexp.last_match[1].to_i if date =~ /([\d]{4})/
+            date = data_from_label(item_details, "Erscheinungsdatum")
+            year = Regexp.last_match[1].to_i if date =~ /(\d{4})/
 
-            binding = data_from_label(item_details, "Einband")
+            book_binding = data_from_label(item_details, "Einband")
 
-            publisher = data_from_label(item_details, "Erschienen bei:")
+            publisher = data_from_label(item_details, "Verlag")
 
             book = Book.new(title, authors, isbns.first,
-                            publisher, year, binding)
+                            publisher, year, book_binding)
 
             image_url = nil
-            if (image_link = doc % "a[@id=itemPicStart]")
-              image_url = image_link["href"]
+            if (image = doc % "section.imagesPreview img")
+              image_url = image["src"]
             end
 
             [book, image_url]
-
           end
         rescue StandardError => ex
           trace = ex.backtrace.join("\n> ")
           log.warn do
             "Failed parsing search results for Thalia " \
-            "#{ex.message} #{trace}"
+              "#{ex.message} #{trace}"
           end
           raise NoResultsError
         end

data/lib/alexandria/book_providers/web.rb

@@ -4,7 +4,7 @@
 #
 # See the file README.md for authorship and licensing information.
 
-require "hpricot"
+require "nokogiri"
 require "htmlentities"
 
 module Alexandria
@@ -19,7 +19,7 @@ module Alexandria
         html.force_encoding source_data_charset
         utf8_html = html.encode("utf-8")
         normalized_html = @htmlentities.decode(utf8_html)
-        Hpricot(normalized_html)
+        Nokogiri.parse(normalized_html)
       end
 
       ## from Palatina

data/lib/alexandria/book_providers/worldcat.rb

@@ -22,7 +22,7 @@ require "alexandria/book_providers/web"
 module Alexandria
   class BookProviders
     class WorldCatProvider < WebsiteBasedProvider
-      include Alexandria::Logging
+      include Logging
 
       SITE = "https://www.worldcat.org"
       BASE_SEARCH_URL = "#{SITE}/search?q=%s%s&qt=advanced" # type, term
@@ -76,11 +76,11 @@ module Alexandria
         doc = html_to_doc(html, "UTF-8")
         book_search_results = []
         begin
-          result_cells = doc / "td.result/div.name/.."
-          # puts result_cells.length
-          result_cells.each do |td|
-            type_icon = (td % "div.type/img.icn")
-            next unless type_icon && type_icon["src"] =~ /icon-bks/
+          result_divs = doc / "td.result/div.name"
+          result_divs.each do |div|
+            td = div.parent
+            type_icon = td % "div.type/img.icn"
+            next unless type_icon && type_icon["src"].include?("icon-bks")
 
             name_div = td % "div.name"
             title = name_div.inner_text
@@ -97,7 +97,7 @@ module Alexandria
           trace = ex.backtrace.join("\n> ")
           log.warn do
             "Failed parsing search results for WorldCat " \
-                    "#{ex.message} #{trace}"
+              "#{ex.message} #{trace}"
           end
         end
         book_search_results
@@ -129,22 +129,20 @@ module Alexandria
               html2 = rslt2.body
 
               book, cover_url = parse_result_data(html2, search_isbn, true)
-              first_result = [book, cover_url] if first_result.nil?
 
               log.debug { "got book #{book}" }
 
-              if search_isbn
-                search_isbn_canon = Library.canonicalise_ean(search_isbn)
-                rslt_isbn_canon = Library.canonicalise_ean(book.isbn)
-                if search_isbn_canon == rslt_isbn_canon
-                  log.info { "book #{book} is a match" }
-                  return [book, cover_url]
-                end
-                log.debug { "not a match, checking next" }
-              else
-                # no constraint to match isbn, just return first result
+              return [book, cover_url] unless search_isbn
+
+              first_result = [book, cover_url] if first_result.nil?
+
+              search_isbn_canon = Library.canonicalise_ean(search_isbn)
+              rslt_isbn_canon = Library.canonicalise_ean(book.isbn)
+              if search_isbn_canon == rslt_isbn_canon
+                log.info { "book #{book} is a match" }
                 return [book, cover_url]
               end
+              log.debug { "not a match, checking next" }
             end
 
             # gone through all and no ISBN match, so just return first result
@@ -152,7 +150,6 @@ module Alexandria
               "no more results to check. Returning first result, just an approximation"
             end
             return first_result
-
           end
 
           title_header = doc % "h1.title"
@@ -174,15 +171,16 @@ module Alexandria
           # can we do better? get the City name?? or multiple publishers?
           bibdata = doc % "div#bibdata"
           bibdata_table = bibdata % :table
-          publisher_row = bibdata_table % "th[text()*=Publisher]/.."
+          publisher_header = bibdata_table % "th[text()*=Publisher]"
 
-          if publisher_row
+          if publisher_header
+            publisher_row = publisher_header.parent
             publication_info = (publisher_row / "td").last.inner_text
 
             publication_info =~ if publication_info.index(";")
-                                  /;[\s]*([^\d]+)[\s]*[\d]*/
+                                  /;\s*([^\d]+)\s*\d*/
                                 elsif publication_info.index(":")
-                                  /:[\s]*([^;:,]+)/
+                                  /:\s*([^;:,]+)/
                                 else
                                   /([^;,]+)/
                                 end
@@ -195,20 +193,18 @@ module Alexandria
             year = nil
           end
 
-          isbn = search_isbn
-          unless isbn
-            isbn_row = doc % "tr#details-standardno"
-            if isbn_row
-              isbns = (isbn_row / "td").last.inner_text.split
-              isbn = Library.canonicalise_isbn(isbns.first)
-            else
-              log.warn { "No ISBN found on page" }
-            end
+          isbn_row = doc % "tr#details-standardno"
+          if isbn_row
+            isbns = (isbn_row / "td").last.inner_text.split
+            isbn = Library.canonicalise_isbn(isbns.first)
+          else
+            log.warn { "No ISBN found on page" }
+            isbn = search_isbn
           end
 
-          binding = "" # not given on WorldCat website (as far as I can tell)
+          book_binding = "" # not given on WorldCat website (as far as I can tell)
 
-          book = Book.new(title, authors, isbn, publisher, year, binding)
+          book = Book.new(title, authors, isbn, publisher, year, book_binding)
 
           image_url = nil # hm, it's on the website, but uses JavaScript...
 
@@ -219,11 +215,11 @@ module Alexandria
           trace = ex.backtrace.join("\n> ")
           log.warn do
             "Failed parsing search results for WorldCat " \
-                    "#{ex.message} #{trace}"
+              "#{ex.message} #{trace}"
           end
           raise NoResultsError
         end
       end
-    end # class WorldCatProvider
-  end # class BookProviders
-end # module Alexandria
+    end
+  end
+end