slaw 0.8.3 → 0.9.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 86252d91f002c7d21cc40838a6656943a705358a
-  data.tar.gz: c1f63cc236fc5dce297ffd3e08bbf2ea9eb4482c
+  metadata.gz: 08d012c5c08856227cf9c203f5c932d076115522
+  data.tar.gz: 742fa6d2ff279945b3ed3cdff468c62a9567cd17
 SHA512:
-  metadata.gz: e0191f62145209f2d0506366011519fe634cd27d8a0bbb1e62fe3415ea7a4967ed9f5a84960305546834622051bcc97674e10261e71b5da57b88b73bb2b4c163
-  data.tar.gz: fe67681172b29cc3ec971276cfb1124984f27999f80af064665c0cd402e03c4b941da9ec2e85745be453f751057038068db2f627c2392e09b7b07a9671cf2c8a
+  metadata.gz: 7f831cd0db6d4d05a80634645b248cfb8555122e85f59af07cc848cdb32330d3f97eab5d59b883802209eb4d9ebe3079a7d29ad80df20a73b2ec340cd1568d2d
+  data.tar.gz: 21285db937fb61a1d81ef7a8a8359219e2b10c066e36e9d01ec6091d107bdd3f2d8ccb5e5fb06f1cb394ff4b8cf52b78c20cedcaab36ef19b8b271623ef946c0

data/.travis.yml

@@ -3,3 +3,5 @@ rvm:
   - 1.9.3
   - 2.1.0
   - 2.1.1
+before_install:
+  - gem update bundler

data/README.md

@@ -216,6 +216,16 @@ Akoma Ntoso `component` elements at the end of the XML document, with a name of
 
 ## Changelog
 
+### 0.9.0
+
+* This release makes reasonably significant changes to generated XML, particularly
+  for sections without explicit subsections.
+* Blocklists with (aa) following (z) are using the same numbering format.
+* Change how blockList listIntroduction elements are created to be more generic
+* Support for sections that dive straight into lists without subsections
+* Simplify grammar
+* Fix elements with potentially duplicate ids
+
 ### 0.8.3
 
 * During cleanup, break lines on section titles that don't have a space after the number, eg: "New section title 4.(1) The content..."

data/lib/slaw/parse/blocklists.rb

@@ -36,7 +36,7 @@ module Slaw
       def self.nest_blocklists(doc)
         doc.xpath('//a:blockList', a: NS).each do |blocklist|
           items = blocklist.xpath('a:item', a: NS)
-          nest_blocklist_items(items.to_a, guess_number_format(items.first), nil, nil)
+          nest_blocklist_items(items.to_a, guess_number_format(items.first), nil, nil) unless items.empty?
         end
       end
 
@@ -47,11 +47,18 @@ module Slaw
         item = items.shift
 
         sublist_count = 0
+        number_format = our_number_format
 
         while item and item.name == 'item'
           number_format = guess_number_format(item, number_format)
           break unless number_format
 
+          # (aa) after (z) is same numbering type, pretend we've always
+          # been this format
+          if item.num == "(aa)" and item.previous_element and item.previous_element.num == "(z)"
+            our_number_format = number_format
+          end
+
           if number_format != our_number_format
             # new sublist, or back to the old list?
             if number_format < our_number_format
@@ -158,6 +165,17 @@ module Slaw
         end
       end
 
+      # Change p tags preceding a blocklist into listIntroductions within the blocklist
+      def self.fix_intros(doc)
+        doc.xpath('//a:blockList', a: NS).each do |blocklist|
+          prev = blocklist.previous
+          if prev and prev.name == 'p'
+            prev.name = 'listIntroduction'
+            blocklist.prepend_child(prev)
+          end
+        end
+      end
+
       class NumberingFormat
         include Comparable
 

data/lib/slaw/parse/builder.rb

@@ -152,7 +152,7 @@ module Slaw
       def postprocess(doc)
         normalise_headings(doc)
         find_short_title(doc)
-        nest_blocklists(doc)
+        adjust_blocklists(doc)
 
         doc
       end
@@ -353,16 +353,17 @@ module Slaw
         end
       end
 
-      # Correctly nest blocklists.
+      # Adjust blocklists:
       #
-      # The grammar gives us flat blocklists, we need to introspect the
-      # numbering of the lists to correctly nest them.
+      # - nest them correctly
+      # - change preceding p tags into listIntroductions
       #
       # @param doc [Nokogiri::XML::Document]
-      def nest_blocklists(doc)
-        logger.info("Nesting blocklists")
+      def adjust_blocklists(doc)
+        logger.info("Adjusting blocklists")
 
         Slaw::Parse::Blocklists.nest_blocklists(doc)
+        Slaw::Parse::Blocklists.fix_intros(doc)
       end
 
       protected

data/lib/slaw/version.rb

@@ -1,3 +1,3 @@
 module Slaw
-  VERSION = "0.8.3"
+  VERSION = "0.9.0"
 end

data/lib/slaw/za/act.treetop

@@ -22,38 +22,39 @@ module Slaw
       rule preface
         !'PREAMBLE'
         ('PREFACE'i space? eol)?
-        statements:(!'PREAMBLE' naked_statement)* <Preface>
+        statements:(!'PREAMBLE' pre_body_statement)* <Preface>
       end
 
       rule preamble
         'PREAMBLE'i space? eol
-        statements:naked_statement* <Preamble>
+        statements:pre_body_statement* <Preamble>
       end
 
       rule body
-        children:(chapter / part / section / block_paragraphs / subsection)+ <Body>
+        children:(chapter / part / section / subsection / block_paragraphs )+ <Body>
       end
 
       rule chapter
         heading:chapter_heading
-        children:(part / section / block_paragraphs / subsection)*
+        children:(part / section / subsection / block_paragraphs)*
         <Chapter>
       end
 
       rule part
         heading:part_heading
-        children:(section / block_paragraphs / subsection)*
+        children:(section / subsection / block_paragraphs)*
         <Part>
       end
 
       rule section
         section_title
-        subsections:subsection* <Section>
+        children:(subsection / block_paragraphs)* <Section>
       end
 
       rule subsection
-        statement:(numbered_statement / table / naked_statement)
-        blocklist:blocklist? <Subsection>
+        # TODO: do it make sense to allow an eol here?
+        space? subsection_prefix whitespace eol?
+        children:block_element* <Subsection>
       end
 
       rule schedules_container
@@ -70,14 +71,6 @@ module Slaw
         <Schedule>
       end
 
-      rule block_paragraphs
-        block_element+ <BlockParagraph>
-      end
-
-      rule block_element
-        (table / blocklist / naked_statement)
-      end
-
       ##########
       # group elements
       # 
@@ -132,7 +125,7 @@ module Slaw
       end
 
       rule section_title_content
-        space !numbered_statement_prefix content eol
+        space !subsection_prefix content eol
       end
 
       rule schedule_title
@@ -142,36 +135,29 @@ module Slaw
       end
 
       ##########
-      # statements
+      # blocks of content inside containers
 
-      rule numbered_statement
-        space? numbered_statement_prefix whitespace (!blocklist_item_prefix clauses eol)? <NumberedStatement>
+      rule block_paragraphs
+        block_element+ <BlockParagraph>
       end
 
-      rule naked_statement
-        space? !(chapter_heading / part_heading / section_title / schedule_title) clauses eol
-        <NakedStatement>
+      rule block_element
+        (table / blocklist / naked_statement)
       end
 
-      rule schedule_statement
-        space? (!schedule_title) clauses eol
-        <ScheduleStatement>
+      rule blocklist
+        blocklist_item+ <Blocklist>
       end
 
-      # one or more words, allowing inline elements
-      rule clauses
-        ((remark / [^ \t\n]+) [ \t]*)+
-        <Clauses>
+      rule blocklist_item
+        space? blocklist_item_prefix whitespace item_content:(!blocklist_item_prefix content eol)? eol?
+        <BlocklistItem>
       end
 
-      rule remark
-        '[[' content:(!']]' .)+ ']]'
-        <Remark>
+      rule blocklist_item_prefix
+        ('(' letter_ordinal ')') / dotted_number_3
       end
 
-      ##########
-      # tables
-
       rule table
         space? table_start eol? (!table_end content eol)* table_end eol
         <Table>
@@ -185,6 +171,33 @@ module Slaw
         space? '|}'
       end
 
+      ##########
+      # statements - single lines of content
+
+      rule naked_statement
+        space? !(chapter_heading / part_heading / section_title / schedule_title / subsection_prefix) clauses eol
+        <NakedStatement>
+      end
+
+      rule pre_body_statement
+        space? !(chapter_heading / part_heading / section_title / schedule_title) clauses eol
+        <NakedStatement>
+      end
+
+      ##########
+      # inline content
+
+      # one or more words, allowing inline elements
+      rule clauses
+        ((remark / [^ \t\n]+) [ \t]*)+
+        <Clauses>
+      end
+
+      rule remark
+        '[[' content:(!']]' .)+ ']]'
+        <Remark>
+      end
+
       ##########
       # prefixes
 
@@ -204,7 +217,7 @@ module Slaw
         number_letter '.'?
       end
 
-      rule numbered_statement_prefix
+      rule subsection_prefix
         # there are two subsection handling syntaxes:
         # 
         # (1) foo
@@ -228,26 +241,6 @@ module Slaw
         num:dotted_number_2 '.'? space
       end
 
-      ##########
-      # blocklists
-
-      rule blocklist
-        blocklist_item+ <Blocklist>
-      end
-
-      rule blocklist_item
-        space? blocklist_item_prefix whitespace item_content:(!blocklist_item_prefix content eol)?
-        <BlocklistItem>
-      end
-
-      rule blocklist_item_prefix
-        ('(' letter_ordinal ')') / dotted_number_3
-      end
-
-      rule letter_ordinal
-        letter (letter / digit)*
-      end
-
       #########
       ## one line of basic content
 
@@ -265,6 +258,10 @@ module Slaw
         number letter*
       end
 
+      rule letter_ordinal
+        letter (letter / digit)*
+      end
+
       rule dotted_number_3
         number '.' number ('.' number)+
       end

data/lib/slaw/za/act_nodes.rb

@@ -79,7 +79,7 @@ module Slaw
       class Body < Treetop::Runtime::SyntaxNode
         def to_xml(b)
           b.body { |b|
-            children.elements.each { |e| e.to_xml(b, '') }
+            children.elements.each_with_index { |e, i| e.to_xml(b, '', i) }
           }
         end
       end
@@ -207,7 +207,7 @@ module Slaw
 
             idprefix = "#{id}."
 
-            subsections.elements.each_with_index { |e, i| e.to_xml(b, idprefix, i) }
+            children.elements.each_with_index { |e, i| e.to_xml(b, idprefix, i) }
           }
         end
       end
@@ -245,70 +245,39 @@ module Slaw
         end
       end
 
-      class Subsection < Treetop::Runtime::SyntaxNode
-        def to_xml(b, idprefix, i=0)
-          if statement.is_a?(NumberedStatement)
-            attribs = {id: idprefix + statement.num.gsub(/[()]/, '')}
-          else
-            attribs = {id: idprefix + "subsection-#{i}"}
-          end
-
-          idprefix = attribs[:id] + "."
-
-          b.subsection(attribs) { |b|
-            b.num(statement.num) if statement.is_a?(NumberedStatement)
-            
-            b.content { |b| 
-              if blocklist and blocklist.is_a?(Blocklist)
-                if statement.content
-                  # provide the statement as the list introduction to the block list
-                  blocklist.to_xml(b, idprefix, i) { |b| statement.content.to_xml(b, idprefix) }
-                else
-                  blocklist.to_xml(b, idprefix, i)
-                end
-              else
-                # raw content
-                statement.to_xml(b, idprefix)
-              end
-            }
-          }
-        end
-      end
-
       class BlockParagraph < Treetop::Runtime::SyntaxNode
         def to_xml(b, idprefix='', i=0)
-          b.paragraph(id: "#{idprefix}paragraph-0") { |b|
+          id = "#{idprefix}paragraph-0"
+          idprefix = "#{id}."
+
+          b.paragraph(id: id) { |b|
             b.content { |b|
-              elements.each_with_index { |e, i| e.to_xml(b, idprefix) }
+              elements.each_with_index { |e, i| e.to_xml(b, idprefix, i) }
             }
           }
         end
       end
 
-      class NumberedStatement < Treetop::Runtime::SyntaxNode
+      class Subsection < Treetop::Runtime::SyntaxNode
         def num
-          numbered_statement_prefix.num.text_value
-        end
-
-        def parentheses?
-          !numbered_statement_prefix.respond_to? :dotted_number_2
+          subsection_prefix.num.text_value
         end
 
-        def content
-          if elements[3].text_value == ""
-            nil
-          else
-            elements[3].clauses
-          end
-        end
+        def to_xml(b, idprefix, i)
+          id = idprefix + num.gsub(/[()]/, '')
+          idprefix = id + "."
 
-        def to_xml(b, idprefix)
-          b.p { |b| content.to_xml(b, idprefix) } if content
+          b.subsection(id: id) { |b|
+            b.num(num)
+            b.content { |b|
+              children.elements.each_with_index { |e, i| e.to_xml(b, idprefix, i) }
+            }
+          }
         end
       end
 
       class NakedStatement < Treetop::Runtime::SyntaxNode
-        def to_xml(b, idprefix)
+        def to_xml(b, idprefix, i=0)
           b.p { |b| clauses.to_xml(b, idprefix) } if clauses
         end
 
@@ -371,7 +340,7 @@ module Slaw
       end
 
       class Table < Treetop::Runtime::SyntaxNode
-        def to_xml(b, idprefix)
+        def to_xml(b, idprefix, i=0)
           # parse the table using wikicloth
           html = WikiCloth::Parser.new({data: self.text_value}).to_html
 
@@ -379,7 +348,7 @@ module Slaw
           # an id to the table
           html = Nokogiri::HTML(html)
           table = html.css("table").first
-          table['id'] = "#{idprefix}table0"
+          table['id'] = "#{idprefix}table#{i}"
 
           # wrap td and th content in p tags
           table.css("td, th").each do |cell|
@@ -470,11 +439,13 @@ module Slaw
               }
 
               b.mainBody { |b| 
+                idprefix = "#{id}."
+
                 # there is no good AKN hierarchy container for schedules, so we
                 # just use article because we don't use it anywhere else.
                 b.article(id: id) { |b|
                   b.heading(heading) if heading
-                  body.children.elements.each { |e| e.to_xml(b) } if body.is_a? Body
+                  body.children.elements.each_with_index { |e| e.to_xml(b, idprefix, i) } if body.is_a? Body
                 }
               }
             }

data/spec/parse/builder_spec.rb

@@ -7,7 +7,7 @@ describe Slaw::Parse::Builder do
   let(:parser) { double("parser") }
   subject { Slaw::Parse::Builder.new(parser: parser) }
 
-  describe '#nest_blocklists' do
+  describe '#adjust_blocklists' do
     it 'should nest simple blocks' do
       doc = xml2doc(subsection(<<XML
             <blockList id="section-10.1.lst0">
@@ -39,7 +39,7 @@ describe Slaw::Parse::Builder do
 XML
       ))
 
-      subject.nest_blocklists(doc)
+      subject.adjust_blocklists(doc)
       doc.to_s.should == subsection(<<XML
             <blockList id="section-10.1.lst0">
               <item id="section-10.1.lst0.a">
@@ -100,7 +100,7 @@ XML
 XML
       ))
 
-      subject.nest_blocklists(doc)
+      subject.adjust_blocklists(doc)
       doc.to_s.should == subsection(<<XML
             <blockList id="section-10.1.lst0">
               <item id="section-10.1.lst0.a">
@@ -147,7 +147,7 @@ XML
 XML
       ))
 
-      subject.nest_blocklists(doc)
+      subject.adjust_blocklists(doc)
       doc.to_s.should == subsection(<<XML
             <blockList id="section-10.1.lst0">
               <item id="section-10.1.lst0.h">
@@ -196,7 +196,7 @@ XML
 XML
       ))
 
-      subject.nest_blocklists(doc)
+      subject.adjust_blocklists(doc)
       doc.to_s.should == subsection(<<XML
             <blockList id="section-10.1.lst0">
               <item id="section-10.1.lst0.t">
@@ -262,7 +262,7 @@ XML
 XML
       ))
 
-      subject.nest_blocklists(doc)
+      subject.adjust_blocklists(doc)
       doc.to_s.should == subsection(<<XML
             <blockList id="section-28.3.list2">
               <item id="section-28.3.list2.g">
@@ -344,7 +344,7 @@ XML
 XML
       ))
 
-      subject.nest_blocklists(doc)
+      subject.adjust_blocklists(doc)
       doc.to_s.should == subsection(<<XML
             <blockList id="section-28.3.list2">
               <item id="section-28.3.list2.g">
@@ -394,6 +394,106 @@ XML
 
     # -------------------------------------------------------------------------
 
+    it 'should treat (aa) after (z) as siblings' do
+      doc = xml2doc(subsection(<<XML
+        <blockList id="list0">
+          <item id="list0.y">
+            <num>(y)</num>
+            <p>foo</p>
+          </item>
+          <item id="list0.z">
+            <num>(z)</num>
+            <p>item-z</p>
+          </item>
+          <item id="list0.aa">
+            <num>(aa)</num>
+            <p>item-aa</p>
+          </item>
+          <item id="list0.bb">
+            <num>(bb)</num>
+            <p>item-bb</p>
+          </item>
+        </blockList>
+XML
+    ))
+
+      subject.adjust_blocklists(doc)
+      doc.to_s.should == subsection(<<XML
+            <blockList id="list0">
+              <item id="list0.y">
+                <num>(y)</num>
+                <p>foo</p>
+              </item>
+              <item id="list0.z">
+                <num>(z)</num>
+                <p>item-z</p>
+              </item>
+              <item id="list0.aa">
+                <num>(aa)</num>
+                <p>item-aa</p>
+              </item>
+              <item id="list0.bb">
+                <num>(bb)</num>
+                <p>item-bb</p>
+              </item>
+            </blockList>
+XML
+      )
+    end
+
+    # -------------------------------------------------------------------------
+
+    it 'should treat (AA) after (z) a sublist' do
+      doc = xml2doc(subsection(<<XML
+        <blockList id="list0">
+          <item id="list0.y">
+            <num>(y)</num>
+            <p>foo</p>
+          </item>
+          <item id="list0.z">
+            <num>(z)</num>
+            <p>item-z</p>
+          </item>
+          <item id="list0.AA">
+            <num>(AA)</num>
+            <p>item-AA</p>
+          </item>
+          <item id="list0.BB">
+            <num>(BB)</num>
+            <p>item-BB</p>
+          </item>
+        </blockList>
+XML
+    ))
+
+      subject.adjust_blocklists(doc)
+      doc.to_s.should == subsection(<<XML
+            <blockList id="list0">
+              <item id="list0.y">
+                <num>(y)</num>
+                <p>foo</p>
+              </item>
+              <item id="list0.z">
+                <num>(z)</num>
+                <blockList id="list0.z.list0">
+                  <listIntroduction>item-z</listIntroduction>
+                  <item id="list0.z.list0.AA">
+                    <num>(AA)</num>
+                    <p>item-AA</p>
+                  </item>
+                  <item id="list0.z.list0.BB">
+                    <num>(BB)</num>
+                    <p>item-BB</p>
+                  </item>
+                </blockList>
+              </item>
+            </blockList>
+XML
+      )
+    end
+
+    # -------------------------------------------------------------------------
+
     it 'should handle deeply nested lists' do
       doc = xml2doc(subsection(<<XML
         <blockList id="list0">
@@ -441,7 +541,7 @@ XML
 XML
     ))
 
-      subject.nest_blocklists(doc)
+      subject.adjust_blocklists(doc)
       doc.to_s.should == subsection(<<XML
             <blockList id="list0">
               <item id="list0.a">
@@ -520,7 +620,7 @@ XML
 XML
       ))
 
-      subject.nest_blocklists(doc)
+      subject.adjust_blocklists(doc)
       doc.to_s.should == subsection(<<XML
             <blockList id="section-10.1.lst0">
               <item id="section-10.1.lst0.h">
@@ -567,7 +667,7 @@ XML
 XML
       ))
 
-      subject.nest_blocklists(doc)
+      subject.adjust_blocklists(doc)
       doc.to_s.should == subsection(<<XML
             <blockList id="section-9.subsection-2.list2">
               <item id="section-9.subsection-2.list2.9.2.1">
@@ -585,6 +685,31 @@ XML
                 </blockList>
               </item>
             </blockList>
+XML
+      )
+    end
+
+    it 'should handle p tags just before' do
+      doc = xml2doc(subsection(<<XML
+        <p>intro</p>
+        <blockList id="section-10.1.lst0">
+          <item id="section-10.1.lst0.a">
+            <num>(a)</num>
+            <p>foo</p>
+          </item>
+        </blockList>
+XML
+        ))
+
+      subject.adjust_blocklists(doc)
+      doc.to_s.should == subsection(<<XML
+            <blockList id="section-10.1.lst0">
+              <listIntroduction>intro</listIntroduction>
+              <item id="section-10.1.lst0.a">
+                <num>(a)</num>
+                <p>foo</p>
+              </item>
+            </blockList>
 XML
       )
     end