slaw 0.6.0 → 0.6.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: e5ab33b04df46f9d2d087f7a22ce9c0e7d12278e
-  data.tar.gz: 276c7b704a30c435d0dfbce2e74c34cd2e902b51
+  metadata.gz: 75915642fffe0d65b3d4386745c12cd019e3d715
+  data.tar.gz: fb183f90a301b479fd4d422d7da479f8796e808d
 SHA512:
-  metadata.gz: 7632de6a68c70b2ce44a4854ec4bcb35867f269447b8c339286056301ef57d73548edabddd5a0ea963d4bc19c66b37d571beb1f8f635d7dc533ca32e1f57bd35
-  data.tar.gz: 6e4a1d0778b6616f6956fba51a1b604514a036f188d89232984d7f57dd5f75f3bad2b031507d9d2eeeb1f470d9b85fef063d055719e97853df7d8665f4715a1a
+  metadata.gz: acf9d6a07f1a927461394721228a8ebd6548fe79c5604e59f36463bdb4bd0a25df1bafa8aa391a937ed45756f1a70f85e9c48469d752c92817c04f392f62c169
+  data.tar.gz: 4ada4ce9e47e9d71bdcf55f1fc8c94c7977e03711b46d0ec86ceae17f8df5a2efcc7a23aa171adf967a0c24a3fd889d80f2aefdee3ebd6c6fb0f0893f14598d6

data/bin/slaw

@@ -12,9 +12,15 @@ class SlawCLI < Thor
   option :input, enum: ['text', 'pdf'], desc: "Type of input if it can't be determined automatically"
   option :pdftotext, desc: "Location of the pdftotext binary if not in PATH"
   option :definitions, type: :boolean, desc: "Find and link definitions (this can be slow). Default: false"
+  option :fragment, type: :string, desc: "Akoma Ntoso element name that the imported text represents. Support depends on the grammar."
+  option :id_prefix, type: :string, desc: "Prefix to be used when generating ID elements when parsing a fragment."
   def parse(name)
     logging
 
+    if options[:fragment] and options[:definitions]
+      raise Thor::Error.new("--definitions can't be used together with --fragment")
+    end
+
     Slaw::Extract::Extractor.pdftotext_path = options[:pdftotext] if options[:pdftotext]
     extractor = Slaw::Extract::Extractor.new
 
@@ -28,6 +34,20 @@ class SlawCLI < Thor
     end
 
     generator = Slaw::ActGenerator.new
+
+    if options[:fragment]
+      generator.document_class = Slaw::Fragment
+      fragment = options[:fragment]
+      fragment = 'act' if fragment.downcase == 'akomantoso'
+      generator.builder.parse_options[:root] = fragment
+
+      if options[:id_prefix]
+        prefix = options[:id_prefix]
+        prefix += "." unless prefix.end_with?('.')
+        generator.builder.fragment_id_prefix = prefix
+      end
+    end
+
     act = generator.generate_from_text(text)
 
     # definitions?

data/lib/slaw/act.rb

@@ -1,4 +1,31 @@
 module Slaw
+  class AknBase
+    include Slaw::Namespace
+
+    attr_accessor :doc
+
+    # Serialise the XML for this act, passing `args` to the Nokogiri serialiser.
+    # The most useful argument is usually `indent: 2` if you like your XML perdy.
+    #
+    # @return [String] serialized XML
+    def to_xml(*args)
+      @doc.to_xml(*args)
+    end
+    
+    # Parse the XML contained in the file-like or String object `io`
+    #
+    # @param io [String, file-like] io object or String with XML
+    def parse(io)
+      self.doc = Nokogiri::XML(io)
+    end
+  end
+
+  # A fragment is a part of a larger document and doesn't have the context associated
+  # with the document.
+  class Fragment < AknBase
+    alias_method :fragment, :doc
+  end
+
   # An Act wraps a single {http://www.akomantoso.org/ AkomaNtoso 2.0 XML} act document in the form of a
   # Nokogiri::XML::Document object.
   #
@@ -7,8 +34,7 @@ module Slaw
   # identifying whether it has been amended ({#amended?}), repealed
   # ({#repealed?}) or what chapters ({#chapters}), parts ({#parts}) and
   # sections ({#sections}) it contains.
-  class Act
-    include Slaw::Namespace
+  class Act < AknBase
 
     # Allow us to jump from the XML document for an act to the
     # Act instance itself
@@ -66,13 +92,6 @@ module Slaw
 
       File.open(filename) { |f| parse(f) }
     end
-    
-    # Parse the XML contained in the file-like or String object `io`
-    #
-    # @param io [String, file-like] io object or String with XML
-    def parse(io)
-      self.doc = Nokogiri::XML(io)
-    end
 
     # Set the XML document backing this bylaw.
     #
@@ -403,14 +422,6 @@ module Slaw
       validate.empty?
     end
 
-    # Serialise the XML for this act, passing `args` to the Nokogiri serialiser.
-    # The most useful argument is usually `indent: 2` if you like your XML perdy.
-    #
-    # @return [String] serialized XML
-    def to_xml(*args)
-      @doc.to_xml(*args)
-    end
-
     def inspect
       "<#{self.class.name} @id_uri=\"#{@id_uri}\">"
     end

data/lib/slaw/generator.rb

@@ -9,6 +9,9 @@ module Slaw
     # [Slaw::Parse::Builder] builder used by the generator
     attr_accessor :builder
 
+    # The type that will hold the generated document
+    attr_accessor :document_class
+
     def initialize
       @parser = Slaw::ZA::ActParser.new
       @builder = Slaw::Parse::Builder.new(parser: @parser)

data/lib/slaw/parse/builder.rb

@@ -26,6 +26,12 @@ module Slaw
 
       @@parsers = {}
 
+      # Additional hash of options to be provided to the parser when parsing.
+      attr_accessor :parse_options
+
+      # Prefix to use when generating IDs for fragments
+      attr_accessor :fragment_id_prefix
+
       # Create a new builder.
       #
       # Specify either `:parser` or `:grammar_file` and `:grammar_class`.
@@ -49,6 +55,8 @@ module Slaw
         else
           raise ArgumentError.new("Specify either :parser or :grammar_file and :grammar_class")
         end
+
+        @parse_options = {}
       end
 
       # Do all the work necessary to parse text into a well-formed XML document.
@@ -65,7 +73,7 @@ module Slaw
       # resulting XML to normalise it.
       #
       # @param text [String] the text to parse
-      # @param parse_options [Hash] options to parse to the parser
+      # @param parse_options [Hash] options to pass to the parser
       #
       # @return [String] an XML string
       def parse_text(text, parse_options={})
@@ -76,11 +84,12 @@ module Slaw
       # Parse plain text into a syntax tree.
       #
       # @param text [String] the text to parse
-      # @param parse_options [Hash] options to parse to the parser
+      # @param parse_options [Hash] options to pass to the parser
       #
       # @return [Object] the root of the resulting parse tree, usually a Treetop::Runtime::SyntaxNode object
       def text_to_syntax_tree(text, parse_options={})
         logger.info("Parsing...")
+        parse_options = @parse_options.dup.update(parse_options)
         tree = @parser.parse(text, parse_options)
         logger.info("Parsed!")
 
@@ -107,7 +116,14 @@ module Slaw
         builder.akomaNtoso("xmlns:xsi"=> "http://www.w3.org/2001/XMLSchema-instance", 
                            "xsi:schemaLocation" => "http://www.akomantoso.org/2.0 akomantoso20.xsd",
                            "xmlns" => NS) { |b|
-          tree.to_xml(b)
+          args = [b]
+
+          # should we provide an id prefix?
+          arity = tree.method('to_xml').arity 
+          arity = arity.abs-1 if arity < 0
+          args << (fragment_id_prefix || "") if arity > 1
+
+          tree.to_xml(*args)
         }
 
         s

data/lib/slaw/version.rb

@@ -1,3 +1,3 @@
 module Slaw
-  VERSION = "0.6.0"
+  VERSION = "0.6.2"
 end

data/lib/slaw/za/act.treetop

@@ -82,7 +82,7 @@ module Slaw
         #
         # Additionally, the section title is optional.
         !{ |s| options[:section_number_after_title] }
-        section_title_prefix section_title:section_title_content? eol?
+        space? section_title_prefix section_title:section_title_content? eol?
         <SectionTitleType2>
       end
 

data/lib/slaw/za/act_nodes.rb

@@ -7,7 +7,7 @@ module Slaw
         EXPRESSION_URI = "#{FRBR_URI}/eng@"
         MANIFESTATION_URI = EXPRESSION_URI
 
-        def to_xml(b)
+        def to_xml(b, idprefix)
           b.act(contains: "originalVersion") { |b|
             write_meta(b)
             write_preamble(b)
@@ -192,7 +192,7 @@ module Slaw
 
             idprefix = "#{id}."
 
-            subsections.elements.each_with_index { |e, i| e.to_xml(b, i, idprefix) }
+            subsections.elements.each_with_index { |e, i| e.to_xml(b, idprefix, i) }
           }
         end
       end
@@ -231,7 +231,7 @@ module Slaw
       end
 
       class Subsection < Treetop::Runtime::SyntaxNode
-        def to_xml(b, i, idprefix)
+        def to_xml(b, idprefix, i=0)
           if statement.is_a?(NumberedStatement)
             attribs = {id: idprefix + statement.num.gsub(/[()]/, '')}
           else
@@ -246,9 +246,9 @@ module Slaw
             b.content { |b| 
               if blocklist and blocklist.is_a?(Blocklist)
                 if statement.content
-                  blocklist.to_xml(b, i, idprefix) { |b| b << statement.content.text_value }
+                  blocklist.to_xml(b, idprefix, i) { |b| b << statement.content.text_value }
                 else
-                  blocklist.to_xml(b, i, idprefix)
+                  blocklist.to_xml(b, idprefix, i)
                 end
               else
                 # raw content
@@ -283,7 +283,7 @@ module Slaw
       class Blocklist < Treetop::Runtime::SyntaxNode
         # Render a block list to xml. If a block is given,
         # yield to it a builder to insert a listIntroduction node
-        def to_xml(b, i, idprefix, &block)
+        def to_xml(b, idprefix, i=0, &block)
           id = idprefix + "list#{i}"
           idprefix = id + '.'
 
@@ -382,7 +382,7 @@ module Slaw
           end
         end
 
-        def to_xml(b, i)
+        def to_xml(b, idprefix, i=0)
           n = num.nil? ? i : num
           id = "schedule-#{n}"
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: slaw
 version: !ruby/object:Gem::Version
-  version: 0.6.0
+  version: 0.6.2
 platform: ruby
 authors:
 - Greg Kempe
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-04-28 00:00:00.000000000 Z
+date: 2015-04-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler