slaw 0.17.2 → 1.0.0.alpha.1

data/lib/slaw/grammars/schedules.treetop

@@ -0,0 +1,33 @@
+# encoding: UTF-8
+
+require 'slaw/grammars/schedules_nodes'
+
+module Slaw
+  module Grammars
+    grammar Schedules
+      rule schedules_container
+        schedules:schedules <ScheduleContainer>
+      end
+
+      rule schedules
+        children:schedule+ <GroupNode>
+      end
+
+      rule schedule
+        schedule_title
+        body:body?
+        <Schedule>
+      end
+
+      rule schedule_title
+        space? schedule_title_prefix space? "\""? num:alphanums? "\""? [ \t:.-]* title:(content)?
+        heading:(newline space? content)?
+        eol
+      end
+
+      rule schedule_title_prefix
+        'schedule'i 's'i?
+      end
+    end
+  end
+end

data/lib/slaw/grammars/schedules_nodes.rb

@@ -0,0 +1,107 @@
+require 'slaw/grammars/core_nodes'
+
+module Slaw
+  module Grammars
+    module Schedules
+      FRBR_URI = '/za/act/1980/01'
+      WORK_URI = FRBR_URI
+      EXPRESSION_URI = "#{FRBR_URI}/eng@"
+      MANIFESTATION_URI = EXPRESSION_URI
+
+      class ScheduleContainer < Treetop::Runtime::SyntaxNode
+        def to_xml(b)
+          b.components { |b| 
+            schedules.children.elements.each_with_index { |e, i|
+              e.to_xml(b, "", i+1)
+            }
+          }
+        end
+      end
+
+      class Schedule < Treetop::Runtime::SyntaxNode
+        def num
+          n = schedule_title.num.text_value
+          return (n && !n.empty?) ? n : nil
+        end
+
+        def alias
+          if not schedule_title.title.text_value.blank?
+            schedule_title.title.text_value
+          elsif num
+            "Schedule #{num}"
+          else
+            "Schedule"
+          end
+        end
+
+        def heading
+          if schedule_title.heading.respond_to? :content
+            schedule_title.heading.content.text_value
+          else
+            nil
+          end
+        end
+
+        def to_xml(b, idprefix=nil, i=1)
+          if num
+            n = num
+            component = "schedule#{n}"
+          else
+            n = i
+            # make a component name from the schedule title
+            component = self.alias.downcase().strip().gsub(/[^a-z0-9]/i, '').gsub(/ +/, '')
+          end
+
+          id = "#{idprefix}#{component}"
+
+          b.component(id: "component-#{id}") { |b|
+            b.doc_(name: component) { |b|
+              b.meta { |b|
+                b.identification(source: "#slaw") { |b|
+                  b.FRBRWork { |b|
+                    b.FRBRthis(value: "#{WORK_URI}/#{component}")
+                    b.FRBRuri(value: WORK_URI)
+                    b.FRBRalias(value: self.alias)
+                    b.FRBRdate(date: '1980-01-01', name: 'Generation')
+                    b.FRBRauthor(href: '#council')
+                    b.FRBRcountry(value: 'za')
+                  }
+                  b.FRBRExpression { |b|
+                    b.FRBRthis(value: "#{EXPRESSION_URI}/#{component}")
+                    b.FRBRuri(value: EXPRESSION_URI)
+                    b.FRBRdate(date: '1980-01-01', name: 'Generation')
+                    b.FRBRauthor(href: '#council')
+                    b.FRBRlanguage(language: 'eng')
+                  }
+                  b.FRBRManifestation { |b|
+                    b.FRBRthis(value: "#{MANIFESTATION_URI}/#{component}")
+                    b.FRBRuri(value: MANIFESTATION_URI)
+                    b.FRBRdate(date: Time.now.strftime('%Y-%m-%d'), name: 'Generation')
+                    b.FRBRauthor(href: '#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_with_index { |e| e.to_xml(b, idprefix, i) } if body.is_a? Body
+                }
+              }
+            }
+          }
+        end
+      end
+
+      class ScheduleStatement < Treetop::Runtime::SyntaxNode
+        def to_xml(b, idprefix)
+          b.p { |b| clauses.to_xml(b, idprefix) } if clauses
+        end
+      end
+    end
+  end
+end

data/lib/slaw/grammars/tables.treetop

@@ -0,0 +1,59 @@
+# encoding: UTF-8
+
+require 'slaw/grammars/terminals'
+require 'slaw/grammars/tables_nodes'
+
+module Slaw
+  module Grammars
+    grammar Tables
+      ##########
+      # wikimedia-style tables
+      #
+      # this grammar doesn't support inline table cells (eg: | col1 || col2 || col3)
+      # instead, the builder preprocesses tables to break inline cells onto their own
+      # lines, which we do support.
+
+      rule table
+        space? '{|' eol
+        table_body
+        '|}' eol
+        <Table>
+      end
+
+      rule table_body
+        (table_row / table_cell)*
+      end
+
+      rule table_row
+        '|-' space? eol
+      end
+
+      rule table_cell
+        # don't match end-of-table
+        !'|}'
+        [!|] attribs:table_attribs? space?
+        # first content line, then multiple lines
+        content:(line:table_line (![!|] space? line:table_line)*)
+        <TableCell>
+      end
+
+      rule table_line
+        clauses:clauses? eol
+        <TableLine>
+      end
+
+      rule table_attribs
+        space? attribs:(table_attrib+) '|'
+      end
+
+      rule table_attrib
+        name:([a-z_-]+) '=' value:(
+                                   ('"' (!'"' .)* '"') /
+                                   ("'" (!"'" .)* "'"))
+        space?
+      end
+
+      include Terminals
+    end
+  end
+end

data/lib/slaw/grammars/tables_nodes.rb

@@ -0,0 +1,74 @@
+module Slaw
+  module Grammars
+    module Tables
+      class Table < Treetop::Runtime::SyntaxNode
+        def to_xml(b, idprefix, i=0)
+          b.table(id: "#{idprefix}table#{i}") { |b|
+            # we'll gather cells into this row list
+            rows = []
+            cells = []
+
+            for child in table_body.elements
+              if child.is_a? TableCell
+                # cell
+                cells << child
+              else
+                # new row marker
+                rows << cells unless cells.empty?
+                cells = []
+              end
+            end
+            rows << cells unless cells.empty?
+
+            for row in rows
+              b.tr { |tr|
+                for cell in row
+                  cell.to_xml(tr, "")
+                end
+              }
+            end
+          }
+        end
+      end
+
+      class TableCell < Treetop::Runtime::SyntaxNode
+        def to_xml(b, idprefix)
+          tag = text_value[0] == '!' ? 'th' : 'td'
+
+          attrs = {}
+          if not attribs.empty?
+            for item in attribs.attribs.elements
+              # key=value (strip quotes around value)
+              attrs[item.name.text_value.strip] = item.value.text_value[1..-2]
+            end
+          end
+
+          b.send(tag.to_sym, attrs) { |b|
+            b.p { |b|
+              # first line, and the rest
+              lines = [content.line] + content.elements.last.elements.map(&:line)
+
+              lines.each_with_index do |line, i|
+                line.to_xml(b, i, i == lines.length-1)
+              end
+            }
+          }
+        end
+      end
+
+      class TableLine < Treetop::Runtime::SyntaxNode
+        # line of table content
+        def to_xml(b, i, tail)
+          clauses.to_xml(b) unless clauses.empty?
+
+          # add trailing newlines.
+          #   for the first line, eat whitespace at the start
+          #   for the last line, eat whitespace at the end
+          if not tail and (i > 0 or not clauses.empty?)
+            eol.text_value.count("\n").times { b.eol }
+          end
+        end
+      end
+    end
+  end
+end

data/lib/slaw/grammars/terminals.treetop

@@ -0,0 +1,84 @@
+# encoding: UTF-8
+
+module Slaw
+  module Grammars
+    grammar Terminals
+      #########
+      ## one line of basic content
+
+      rule content
+        # anything but a newline, followed by a
+        # newline or end of file (without consuming the newline)
+        [^\n]+ &eol
+      end
+
+      ##########
+      # terminals
+
+      # eg. 2, 2A, 2b
+      rule number_letter
+        number letter*
+      end
+
+      rule letter_ordinal
+        letter (letter / digit)*
+      end
+
+      rule dotted_number_3
+        number '.' number ('.' number)+
+      end
+
+      rule dotted_number_2
+        number '.' number
+      end
+
+      rule number
+        digit+
+      end
+
+      rule digit
+        [0-9]
+      end
+
+      rule letter
+        [a-zA-Z]
+      end
+
+      rule alphanums
+        [a-zA-Z0-9]+
+      end
+
+      rule quotes
+        ["“”]
+      end
+      
+      rule non_quotes
+        [^"“”]
+      end
+
+      ##########
+      # whitespace
+
+      rule space
+        [ \t]+
+      end
+
+      rule whitespace
+        [ \t\n]*
+      end
+
+      rule empty_line
+        space? newline
+      end
+
+      rule eol
+        newline
+        empty_line*
+      end
+
+      rule newline
+        "\n"
+      end
+    end
+  end
+end

data/lib/slaw/grammars/za/act.treetop

@@ -0,0 +1,222 @@
+# encoding: UTF-8
+
+require 'slaw/parse/grammar_helpers'
+require 'slaw/grammars/za/act_nodes'
+
+require 'slaw/grammars/terminals'
+require 'slaw/grammars/tables'
+require 'slaw/grammars/schedules'
+require 'slaw/grammars/inlines'
+
+module Slaw
+  module Grammars
+    module ZA
+      grammar Act
+        include Slaw::Parse::GrammarHelpers
+
+        ########
+        # major containers
+
+        rule act
+          empty_line*
+          preface:preface?
+          preamble:preamble?
+          body
+          schedules:schedules_container? <Act>
+        end
+
+        rule preface
+          !'PREAMBLE'
+          ('PREFACE'i space? eol)?
+          statements:(!'PREAMBLE' pre_body_statement)* <Preface>
+        end
+
+        rule preamble
+          'PREAMBLE'i space? eol
+          statements:pre_body_statement* <Preamble>
+        end
+
+        rule body
+          children:(chapter / part / section / subsection / block_paragraphs)+ <Body>
+        end
+
+        rule chapter
+          heading:chapter_heading
+          children:(part / section / subsection / block_paragraphs)*
+          <Chapter>
+        end
+
+        rule part
+          heading:part_heading
+          children:(section / subsection / block_paragraphs)*
+          <Part>
+        end
+
+        rule section
+          section_title
+          children:(subsection / block_paragraphs)* <Section>
+        end
+
+        rule subsection
+          space? subsection_prefix space?
+          # eg: (2) (a) foo
+          first_child:inline_block_element?
+          # eg: (2)
+          eol?
+          children:block_element* <Subsection>
+        end
+
+        ##########
+        # group elements
+        # 
+        # these are used externally and provide support when parsing just
+        # a particular portion of a document
+
+        rule chapters
+          children:chapter+ <GroupNode>
+        end
+
+        rule parts
+          children:part+ <GroupNode>
+        end
+
+        rule sections
+          children:section+ <GroupNode>
+        end
+
+        ##########
+        # headings
+
+        rule chapter_heading
+          space? chapter_heading_prefix heading:(newline? content)? eol
+          <ChapterHeading>
+        end
+
+        rule part_heading
+          space? part_heading_prefix heading:(newline? content)? eol
+          <PartHeading>
+        end
+
+        rule section_title
+          section_title_1 / section_1_title
+        end
+
+        rule section_title_1
+          &{ |s| options[:section_number_after_title] }
+          # Section title
+          # 1. Section content
+          content eol
+          section_title_prefix whitespace <SectionTitleType1>
+        end
+
+        rule section_1_title
+          # 1. Section title
+          # Section content
+          #
+          # Additionally, the section title is optional.
+          !{ |s| options[:section_number_after_title] }
+          space? section_title_prefix section_title:section_title_content? eol?
+          <SectionTitleType2>
+        end
+
+        rule section_title_content
+          # don't match subsections, eg.
+          #
+          # 10. (1) subsection content...
+          space !subsection_prefix content eol
+        end
+
+        ##########
+        # blocks of content inside containers
+
+        rule block_paragraphs
+          block_element+ <BlockParagraph>
+        end
+
+        rule block_element
+          (table / blocklist / naked_statement)
+        end
+
+        # Block elements that don't have to appear at the start of a line.
+        # ie. we don't need to guard against the start of a chapter, section, etc.
+        rule inline_block_element
+          (table / blocklist / inline_statement)
+        end
+
+        rule blocklist
+          blocklist_item+ <Blocklist>
+        end
+
+        rule blocklist_item
+          # TODO: this whitespace should probably be space, to allow empty blocklist items followed by plain text
+          space? blocklist_item_prefix whitespace item_content:(!blocklist_item_prefix clauses:clauses? eol)? eol?
+          <BlocklistItem>
+        end
+
+        rule blocklist_item_prefix
+          ('(' letter_ordinal ')') / dotted_number_3
+        end
+
+        ##########
+        # statements - single lines of content
+        #
+        # If a statement starts with a backslash, it's considered to have escaped the subsequent word,
+        # and is ignored. This allows escaping of section headings, etc.
+
+        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
+
+        ##########
+        # prefixes
+
+        rule part_heading_prefix
+          'part'i space alphanums [ :-]*
+        end
+
+        rule chapter_heading_prefix
+          'chapter'i space alphanums [ :-]*
+        end
+
+        rule section_title_prefix
+          number_letter '.'?
+        end
+
+        rule subsection_prefix
+          # there are two subsection handling syntaxes:
+          # 
+          # (1) foo
+          # (2A) foo
+          #
+          # and
+          #
+          # 8.2 for
+          # 8.3 bar
+          #
+          # The second is less common, but this allows us to handle it.
+          # Note that it is usually accompanied by a similar list number format:
+          #
+          # 8.2.1 item 1
+          # 8.2.2 item 2
+          #
+          # which aren't subsections, but lists, so force the space at the end
+          # of the number to catch this case.
+          num:('(' number_letter ')')
+          /
+          num:dotted_number_2 '.'? space
+        end
+
+        include Slaw::Grammars::Inlines
+        include Slaw::Grammars::Tables
+        include Slaw::Grammars::Schedules
+        include Slaw::Grammars::Terminals
+      end
+    end
+  end
+end