oddb2xml 2.0.5 → 2.0.6

data/spec/spec_helper.rb

@@ -13,7 +13,12 @@ Bundler.require
 
 require 'rspec'
 require 'webmock/rspec'
+require 'pp'
 
+begin # load pry if is available
+require 'pry'
+rescue LoadError
+end
 
 module Oddb2xml
   # we override here a few directories to make input/output when running specs to
@@ -24,6 +29,9 @@ module Oddb2xml
   SpecCompressor = File.join(Oddb2xml::SpecData, 'compressor')
 end
 
+ExcipiensIs_a_Substance = false #  might change later
+AllCompositionLines = File.expand_path("#{__FILE__}/../data/compositions.txt")
+
 require 'oddb2xml'
 
 module ServerMockHelper

metadata

@@ -1,195 +1,209 @@
 --- !ruby/object:Gem::Specification
 name: oddb2xml
 version: !ruby/object:Gem::Version
-  version: 2.0.5
+  version: 2.0.6
 platform: ruby
 authors:
 - Yasuhiro Asaka, Zeno R.R. Davatz
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-04-01 00:00:00.000000000 Z
+date: 2015-04-28 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rubyzip
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.1.3
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.1.3
 - !ruby/object:Gem::Dependency
   name: archive-tar-minitar
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 0.5.2
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 0.5.2
 - !ruby/object:Gem::Dependency
   name: mechanize
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 2.5.1
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 2.5.1
 - !ruby/object:Gem::Dependency
   name: nokogiri
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.5.10
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.5.10
 - !ruby/object:Gem::Dependency
   name: savon
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 2.4.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 2.4.0
 - !ruby/object:Gem::Dependency
   name: spreadsheet
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.0.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.0.0
 - !ruby/object:Gem::Dependency
   name: rubyXL
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 3.3.1
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 3.3.1
 - !ruby/object:Gem::Dependency
   name: sax-machine
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 0.1.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 0.1.0
+- !ruby/object:Gem::Dependency
+  name: parslet
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.7.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.7.0
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
   name: webmock
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
   name: rdoc
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
 description: oddb2xml creates xml files using swissINDEX, BAG-XML and Swissmedic.
@@ -199,9 +213,9 @@ executables:
 extensions: []
 extra_rdoc_files: []
 files:
-- ".gitignore"
-- ".rspec"
-- ".travis.yml"
+- .gitignore
+- .rspec
+- .travis.yml
 - Gemfile
 - Gemfile.lock
 - History.txt
@@ -213,15 +227,17 @@ files:
 - bin/oddb2xml
 - data/gal_forms.yaml
 - data/gal_groups.yaml
+- dokumentation_calc.textile
 - lib/oddb2xml.rb
 - lib/oddb2xml/builder.rb
 - lib/oddb2xml/calc.rb
 - lib/oddb2xml/cli.rb
+- lib/oddb2xml/compositions_syntax.rb
 - lib/oddb2xml/compressor.rb
 - lib/oddb2xml/downloader.rb
 - lib/oddb2xml/extractor.rb
 - lib/oddb2xml/options.rb
-- lib/oddb2xml/parse_compositions.rb
+- lib/oddb2xml/parslet_compositions.rb
 - lib/oddb2xml/util.rb
 - lib/oddb2xml/version.rb
 - lib/oddb2xml/xml_definitions.rb
@@ -230,6 +246,7 @@ files:
 - spec/builder_spec.rb
 - spec/calc_spec.rb
 - spec/cli_spec.rb
+- spec/composition_syntax_spec.rb
 - spec/compressor_spec.rb
 - spec/data/GL_Diff_SB.xml
 - spec/data/ItCodes.xml
@@ -238,6 +255,7 @@ files:
 - spec/data/Preparation.xml
 - spec/data/Preparations.xml
 - spec/data/XMLPublications.zip
+- spec/data/compositions.txt
 - spec/data/compressor/oddb2xml_files_bm_update.txt
 - spec/data/compressor/oddb2xml_files_lppv.txt
 - spec/data/compressor/oddb2xml_files_nonpharma.xls
@@ -276,6 +294,7 @@ files:
 - spec/downloader_spec.rb
 - spec/extractor_spec.rb
 - spec/options_spec.rb
+- spec/parslet_spec.rb
 - spec/spec_helper.rb
 - test_options.rb
 - tools/cacert.pem
@@ -291,12 +310,12 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
@@ -309,6 +328,7 @@ test_files:
 - spec/builder_spec.rb
 - spec/calc_spec.rb
 - spec/cli_spec.rb
+- spec/composition_syntax_spec.rb
 - spec/compressor_spec.rb
 - spec/data/GL_Diff_SB.xml
 - spec/data/ItCodes.xml
@@ -317,6 +337,7 @@ test_files:
 - spec/data/Preparation.xml
 - spec/data/Preparations.xml
 - spec/data/XMLPublications.zip
+- spec/data/compositions.txt
 - spec/data/compressor/oddb2xml_files_bm_update.txt
 - spec/data/compressor/oddb2xml_files_lppv.txt
 - spec/data/compressor/oddb2xml_files_nonpharma.xls
@@ -355,4 +376,5 @@ test_files:
 - spec/downloader_spec.rb
 - spec/extractor_spec.rb
 - spec/options_spec.rb
+- spec/parslet_spec.rb
 - spec/spec_helper.rb

data/lib/oddb2xml/parse_compositions.rb

@@ -1,106 +0,0 @@
-# encoding: utf-8
-
-# This file is shared since oddb2xml 2.0.0 (lib/oddb2xml/parse_compositions.rb)
-# with oddb.org src/plugin/parse_compositions.rb
-#
-# It allows an easy parsing of the column P Zusammensetzung of the swissmedic packages.xlsx file
-#
-
-module ParseUtil
-  SCALE_P = %r{pro\s+(?<scale>(?<qty>[\d.,]+)\s*(?<unit>[kcmuµn]?[glh]))}u
-  ParseComposition   = Struct.new("ParseComposition",  :source, :label, :label_description, :substances, :galenic_form, :route_of_administration)
-  ParseSubstance     = Struct.new("ParseSubstance",    :name, :qty, :unit, :chemical_substance, :chemical_qty, :chemical_unit, :is_active_agent, :dose, :cdose)
-  def ParseUtil.capitalize(string)
-    string.split(/\s+/u).collect { |word| word.capitalize }.join(' ')
-  end
-
-  def ParseUtil.dose_to_qty_unit(string, filler=nil)
-    return nil unless string
-    dose = string.split(/\b\s*(?![.,\d\-]|Mio\.?)/u, 2)
-    if dose && (scale = SCALE_P.match(filler)) && dose[1] && !dose[1].include?('/')
-      unit = dose[1] << '/'
-      num = scale[:qty].to_f
-      if num <= 1
-        unit << scale[:unit]
-      else
-        unit << scale[:scale]
-      end
-      dose[1] = unit
-    elsif dose and dose.size == 2
-      unit = dose[1]
-    end
-    dose
-  end
-
-  def ParseUtil.parse_compositions(composition, active_agents_string = '')
-    rep_1 = '----';   to_1 = '('
-    rep_2 = '-----';  to_2 = ')'
-    rep_3 = '------'; to_3 = ','
-    active_agents = active_agents_string ? active_agents_string.downcase.split(/,\s+/) : []
-    comps = []
-    label_pattern = /^(?<label>A|I|B|II|C|III|D|IV|E|V|F|VI)(\):|\))\s*(?<description>[^:]+)/
-    composition_text = composition.gsub(/\r\n?/u, "\n")
-    puts "composition_text for #{name}: #{composition_text}" if composition_text.split(/\n/u).size > 1 and $VERBOSE
-    lines = composition_text.split(/\n/u)
-    idx = 0
-    compositions = lines.select do |line|
-      if match = label_pattern.match(line)
-        label = match[:label]
-        label_description = match[:description]
-      else
-        label = nil
-        label_description = nil
-      end
-      idx += 1
-      next if idx > 1 and /^(?<label>A|I|B|II|C|III|D|IV|E|V|F|VI)[)]\s*(et)/.match(line) # avoid lines like 'I) et II)'
-      next if idx > 1 and /^Corresp\./i.match(line) # avoid lines like 'Corresp. mineralia: '
-      substances = []
-      filler = line.split(',')[-1].sub(/\.$/, '')
-      filler_match = /^(?<name>[^,\d]+)\s*(?<dose>[\d\-.]+(\s*(?:(Mio\.?\s*)?(U\.\s*Ph\.\s*Eur\.|[^\s,]+))))/.match(filler)
-      components = line.gsub(/(\d),(\d+)/, "\\1.\\2").split(/([^\(]+\([^)]+\)[^,]+|),/).each {
-        |component|
-        next unless component.size > 0
-        next if /^ratio:/i.match(component.strip)
-        to_consider = component.strip.split(':')[-1].gsub(to_1, rep_1).gsub(to_2, rep_2).gsub(to_3, rep_3) # remove label
-        # very ugly hack to ignore ,()
-        ptrn1 = /^(?<name>.+)(\s+|$)(?<dose>[\d\-.]+(\s*(?:(Mio\.?\s*)?(U\.\s*Ph\.\s*Eur\.|[^\s,]+))))/
-        m =  /^(?<name>.+)(\s+|$)/.match(to_consider)
-        m_with_dose = ptrn1.match(to_consider)
-        m = m_with_dose if m_with_dose
-        if m2 = /^(|[^:]+:\s)(E\s+\d+)$/.match(component.strip)
-          to_add = ParseSubstance.new(m2[2], '', nil, nil, nil, nil, active_agents.index(m2[2].downcase) ? true : false)
-          substances << to_add
-        elsif m
-          ptrn = /(?<name>.+)\s+(?<dose>[\d\-.]+(\s*(?:(Mio\.?\s*)?(U\.\s*Ph\.\s*Eur\.|[^\s,]+))))(\s*(?:ut|corresp\.?)\s+(?<chemical>[^\d,]+)\s*(?<cdose>[\d\-.]+(\s*(?:(Mio\.?\s*)?(U\.\s*Ph\.\s*Eur\.|[^\s,]+))(\s*[mv]\/[mv])?))?)/
-          m_with_chemical = ptrn.match(to_consider)
-          m = m_with_chemical if m_with_chemical
-          name      = m[:name].strip
-          chemical  = m_with_chemical ? m[:chemical] : nil
-          cdose     = m_with_chemical ? m[:cdose] : nil
-          dose      = m_with_dose     ? m[:dose]  : nil
-          if m_with_chemical and active_agents.index(m_with_chemical[:chemical].strip)
-            is_active_agent = true
-            name            = m[:chemical].strip
-            dose            = m[:cdose]
-            chemical        = m[:name].strip
-            cdose           = m[:dose]
-          else
-            is_active_agent =  active_agents.index(m[:name].strip) != nil
-          end
-          unit = nil
-          name = name.gsub(rep_3, to_3).gsub(rep_2, to_2).gsub(rep_1, to_1)
-          emulsion_pattern = /\s+pro($|\s+)|emulsion|solution/i
-          next if emulsion_pattern.match(name)
-          name = name.split(/\s/).collect{ |x| x.capitalize }.join(' ').strip
-          chemical = chemical.split(/\s/).collect{ |x| x.capitalize }.join(' ').strip if chemical
-          qty,  unit  = ParseUtil.dose_to_qty_unit(dose, filler)
-          cqty, cunit = ParseUtil.dose_to_qty_unit(cdose, filler)
-          dose = "#{qty} #{unit}" if unit and unit.match(/\//)
-          substances << ParseSubstance.new(name, qty, unit, chemical, cqty, cunit, is_active_agent, dose, cdose)
-        end
-      }
-      comps << ParseComposition.new(line, label, label_description, substances) if substances.size > 0
-    end
-    comps
-  end
-end