slaw 0.17.2 → 1.0.0.alpha.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 60e83b6293761721e7a2213e9e5c668accc5873c
-  data.tar.gz: b8e31cc3f17512a32af5426a86f87c68c14a0b45
+  metadata.gz: e5c6f9929b92711f0a7c608387b49784c8c0198c
+  data.tar.gz: 7557e00919085931eae2ce9195bbe55342a50c9c
 SHA512:
-  metadata.gz: efad800e07b95ae255fe44ccf118f5dc43793b88394d5eec87b59ddc30ef0222aabbd2e513c7a5dd3c6b34a271e77dccb599f10e2adf8c2faa0f0a83c67f65bf
-  data.tar.gz: c5c44d83a2c736f62122a96301435517999ba008768f3d54642b828f9897163172da7202691bd32a11a5427613f90a1ae0e5863ae375f6a3bcdf0514616b4fad
+  metadata.gz: 386d5e7195f18838c00af784a46792f349913ac5fe07f7799e4e3055fd6720da3e1dcb6b2e10b93d14e2519758eecdd4b46fe31ceecf2bf10b707cf51f6e93dd
+  data.tar.gz: 73aa8a5060a8933a09bec8ebac589aeb835de216a76f7865f9bf5883eb89ab347117c648eda8e806eaa668d4f9c1ce7eefeb4bd9dd678abbcb1a200986d98063

data/README.md

@@ -8,7 +8,7 @@ Slaw allows you to:
 
 1. extract plain text from PDFs and clean up that text
 2. parse plain text and transform it into an Akoma Ntoso Act XML document
-3. render the XML document into HTML
+3. unparse Akoma Ntoso XML into text that can be parsed backed into Akoma Ntoso.
 
 Slaw is lightweight because it wraps around a Nokogiri XML representation of
 the parsed document. It provides some support methods for manipulating these
@@ -61,7 +61,7 @@ formats.
 
 The grammar cannot catch some subtleties of an act or by-law -- such as nested list numbering --
 so Slaw performs some post-processing on the XML produced by the parser. In particular,
-it nests lists correctly and looks for specially defined terms and their occurrences in the document.
+it nests lists correctly.
 
 ## Quick Start
 
@@ -218,6 +218,13 @@ Akoma Ntoso `component` elements at the end of the XML document, with a name of
 
 ## Changelog
 
+### 1.0.0
+
+* Improved support for other legal traditions / grammars.
+* Add Polish legal tradition grammar.
+* Slaw no longer does too much introspection of a parsed document, since that can be so tradition-dependent.
+* Remove definition linking, Slaw no longer supports it.
+
 ### 0.17.2
 
 * Match defined terms in 'definition' section.

data/bin/slaw

@@ -17,19 +17,15 @@ class SlawCLI < Thor
   desc "parse FILE", "Parse FILE into Akoma Ntoso XML"
   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."
   option :section_number_position, enum: ['before-title', 'after-title', 'guess'], desc: "Where do section titles come in relation to the section number? Default: before-title"
   option :reformat, type: :boolean, desc: "Reformat common formatting issues to make grammar matching better. Default: true for PDF files, false otherwise"
   option :crop, type: :string, desc: "Crop box for PDF files, as 'left,top,width,height'."
+  option :grammar, type: :string, desc: "Grammar name (usually a two-letter country code). Default is za."
   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
 
@@ -50,7 +46,7 @@ class SlawCLI < Thor
       text = extractor.extract_from_file(name)
     end
 
-    generator = Slaw::ActGenerator.new
+    generator = Slaw::ActGenerator.new(options[:grammar] || 'za')
 
     text = generator.reformat(text) if options[:reformat]
 
@@ -94,22 +90,9 @@ class SlawCLI < Thor
       exit 1
     end
 
-    # definitions?
-    generator.builder.link_definitions(act.doc) if options[:definitions]
-
     puts act.to_xml(indent: 2)
   end
 
-  desc "link-definitions FILE", "Find and link defined terms in FILE"
-  def link_definitions(name)
-    builder = Slaw::ActGenerator.new.builder
-
-    doc = File.open(name, 'r') { |f| doc = builder.parse_xml(f.read) }
-    builder.link_definitions(doc)
-
-    puts builder.to_xml(doc)
-  end
-
   desc "unparse FILE", "Unparse FILE from Akoma Ntoso XML back into text suitable for re-parsing"
   def unparse(name)
     generator = Slaw::ActGenerator.new

data/lib/slaw/generator.rb

@@ -1,8 +1,6 @@
 module Slaw
   # Base class for generating Act documents
   class ActGenerator
-    Treetop.load(File.dirname(__FILE__) + "/za/act.treetop")
-
     # [Treetop::Runtime::CompiledParser] compiled parser
     attr_accessor :parser
 
@@ -12,13 +10,31 @@ module Slaw
     # The type that will hold the generated document
     attr_accessor :document_class
 
-    def initialize
-      @parser = Slaw::ZA::ActParser.new
+    @@parsers = {}
+
+    def initialize(grammar)
+      @grammar = grammar
+
+      @parser = build_parser
       @builder = Slaw::Parse::Builder.new(parser: @parser)
+      @parser = @builder.parser
       @cleanser = Slaw::Parse::Cleanser.new
       @document_class = Slaw::Act
     end
 
+    def build_parser
+      unless @@parsers[@grammar]
+        # load the grammar
+        grammar_file = File.dirname(__FILE__) + "/grammars/#{@grammar}/act.treetop"
+        Treetop.load(grammar_file)
+
+        grammar_class = "Slaw::Grammars::#{@grammar.upcase}::ActParser"
+        @@parsers[@grammar] = eval(grammar_class)
+      end
+
+      @parser = @@parsers[@grammar].new
+    end
+
     # Generate a Slaw::Act instance from plain text.
     #
     # @param text [String] plain text
@@ -66,8 +82,7 @@ module Slaw
     # Transform an Akoma Ntoso XML document back into a plain-text version
     # suitable for re-parsing back into XML with no loss of structure.
     def text_from_act(doc)
-      here = File.dirname(__FILE__)
-      xslt = Nokogiri::XSLT(File.read(File.join([here, 'za/act_text.xsl'])))
+      xslt = Nokogiri::XSLT(File.read(File.join([File.dirname(__FILE__), "grammars/#{@grammar}/act_text.xsl"])))
       xslt.transform(doc).child.to_xml
     end
   end

data/lib/slaw/grammars/core_nodes.rb

@@ -0,0 +1,17 @@
+module Slaw
+  module Grammars
+    class GroupNode < Treetop::Runtime::SyntaxNode
+      def to_xml(b, *args)
+        children.elements.each { |e| e.to_xml(b, *args) }
+      end
+    end
+
+    class Body < Treetop::Runtime::SyntaxNode
+      def to_xml(b)
+        b.body { |b|
+          children.elements.each_with_index { |e, i| e.to_xml(b, '', i) }
+        }
+      end
+    end
+  end
+end

data/lib/slaw/grammars/inlines.treetop

@@ -0,0 +1,45 @@
+# encoding: UTF-8
+
+require 'slaw/grammars/terminals'
+require 'slaw/grammars/inlines_nodes'
+
+module Slaw
+  module Grammars
+    grammar Inlines
+      ##########
+      # inline content
+
+      rule inline_statement
+        space? '\\'? clauses eol
+        <NakedStatement>
+      end
+
+      # one or more words, allowing inline elements
+      rule clauses
+        (remark / image / ref / [^\n])+
+        <Clauses>
+      end
+
+      rule remark
+        '[[' content:(ref / (!']]' .))+ ']]'
+        <Remark>
+      end
+
+      rule image
+        # images like markdown
+        # eg. ![title text](image url)
+        #
+        # the title text is optional, but the enclosing square brackets aren't
+        '![' content:(!'](' [^\n])* '](' href:([^)\n]+) ')'
+        <Image>
+      end
+
+      rule ref
+        # links like markdown
+        # eg. [link text](link url)
+        '[' content:(!'](' [^\n])+ '](' href:([^)\n]+) ')'
+        <Ref>
+      end
+    end
+  end
+end

data/lib/slaw/grammars/inlines_nodes.rb

@@ -0,0 +1,58 @@
+module Slaw
+  module Grammars
+    module Inlines
+      class NakedStatement < Treetop::Runtime::SyntaxNode
+        def to_xml(b, idprefix, i=0)
+          b.p { |b| clauses.to_xml(b, idprefix) } if clauses
+        end
+
+        def content
+          clauses
+        end
+      end
+
+      class Clauses < Treetop::Runtime::SyntaxNode
+        def to_xml(b, idprefix=nil)
+          for e in elements
+            if e.respond_to? :to_xml
+              e.to_xml(b, idprefix)
+            else
+              b << e.text_value
+            end
+          end
+        end
+      end
+
+      class Remark < Treetop::Runtime::SyntaxNode
+        def to_xml(b, idprefix)
+          b.remark(status: 'editorial') do |b|
+            b << '['
+            for e in content.elements
+              if e.respond_to? :to_xml
+                e.to_xml(b, idprefix)
+              else
+                b << e.text_value
+              end
+            end
+            b << ']'
+          end
+        end
+      end
+
+      class Image < Treetop::Runtime::SyntaxNode
+        def to_xml(b, idprefix)
+          attrs = {src: href.text_value}
+          attrs[:alt] = content.text_value unless content.text_value.empty?
+          b.img(attrs)
+        end
+      end
+
+      class Ref < Treetop::Runtime::SyntaxNode
+        def to_xml(b, idprefix)
+          b.ref(content.text_value, href: href.text_value)
+        end
+      end
+
+    end
+  end
+end

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

@@ -0,0 +1,246 @@
+# encoding: UTF-8
+
+require 'slaw/parse/grammar_helpers'
+require 'slaw/grammars/pl/act_nodes'
+
+require 'slaw/grammars/terminals'
+require 'slaw/grammars/tables'
+require 'slaw/grammars/schedules'
+require 'slaw/grammars/inlines'
+
+module Slaw
+  module Grammars
+    module PL
+      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:(division / subdivision / chapter / article / section / paragraph / point / litera / block_paragraphs)+ <Body>
+        end
+
+        rule division
+          heading:division_heading
+          children:(subdivision / chapter / article / section / paragraph / point / litera / block_paragraphs)*
+          <Division>
+        end
+
+        rule subdivision
+          heading:subdivision_heading
+          children:(chapter / article / section / paragraph / point / litera / block_paragraphs)*
+          <Subdivision>
+        end
+
+        rule chapter
+          heading:chapter_heading
+          children:(article / section / paragraph / point / litera / block_paragraphs)*
+          <Chapter>
+        end
+
+        rule article
+          # Art. 55. 1. something
+          article_prefix whitespace
+          intro:block_element? eol?
+          children:(section / paragraph / point / litera / block_paragraphs)* <Article>
+        end
+
+        rule section
+          # § 55. foo
+          section_prefix whitespace
+          intro:block_element? eol?
+          children:(paragraph / point / litera / block_paragraphs)* <Section>
+        end
+
+        rule paragraph
+          # ustęp:
+          #  34. ...
+          paragraph_prefix space?
+          intro:block_element? eol?
+          children:(point / litera / block_paragraphs)* <Paragraph>
+        end
+
+        rule point
+          # 12) aoeuaoeu
+          # 12a) aoeuaoeu
+          point_prefix whitespace
+          intro:block_element? eol?
+          children:(litera / block_paragraphs)* <Point>
+        end
+
+        rule litera
+          # a) aoeuaoeu
+          litera_prefix whitespace
+          intro:block_element? eol?
+          children:block_paragraphs* <Litera>
+        end
+
+        ##########
+        # group elements
+        # 
+        # these are used externally and provide support when parsing just
+        # a particular portion of a document
+
+        rule divisions
+          children:division+ <GroupNode>
+        end
+
+        rule subdivisions
+          children:subdivision+ <GroupNode>
+        end
+
+        rule chapters
+          children:chapter+ <GroupNode>
+        end
+
+        rule articles
+          children:article+ <GroupNode>
+        end
+
+        rule sections
+          children:section+ <GroupNode>
+        end
+
+        rule paragraphs
+          children:paragraph+ <GroupNode>
+        end
+
+        rule points
+          children:point+ <GroupNode>
+        end
+
+        ##########
+        # headings
+
+        rule division_heading
+          space? prefix:division_heading_prefix heading:(newline? content)? eol
+          <GenericHeading>
+        end
+
+        rule subdivision_heading
+          space? prefix:subdivision_heading_prefix heading:(newline? content)? eol
+          <GenericHeading>
+        end
+
+        rule chapter_heading
+          space? prefix:chapter_heading_prefix heading:(newline? content)? eol
+          <GenericHeading>
+        end
+
+        ##########
+        # blocks of content inside containers
+
+        rule block_paragraphs
+          block_element+ <BlockParagraph>
+        end
+
+        rule block_element
+          # XXX: blocklist
+          (table / 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
+          # XXX: blocklist
+          (table / 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? !(division_heading / subdivision_heading / chapter_heading / article_prefix / section_prefix / schedule_title / paragraph_prefix / point_prefix / litera_prefix) '\\'? clauses eol
+          <NakedStatement>
+        end
+
+        rule pre_body_statement
+          space? !(division_heading / subdivision_heading / chapter_heading / article_prefix / section_prefix / schedule_title) '\\'? clauses eol
+          <NakedStatement>
+        end
+
+        ##########
+        # prefixes
+
+        rule division_heading_prefix
+          'dzia'i ('ł'/'Ł') space alphanums [ :-]*
+        end
+
+        rule subdivision_heading_prefix
+          'oddzia'i ('ł'/'Ł') space alphanums [ :.-]*
+        end
+
+        rule chapter_heading_prefix
+          'rozdzia'i ('ł'/'Ł') space alphanums [ :.-]*
+        end
+
+        rule article_prefix
+          ('Art.'i / ('Artyku'i 'ł'/'Ł')) space number_letter '.'?
+        end
+
+        rule section_prefix
+          '§' space alphanums '.'?
+        end
+
+        rule paragraph_prefix
+          number_letter '.'
+        end
+
+        rule point_prefix
+          # 1) foo
+          # 2A) foo
+          number_letter ')'
+        end
+
+        rule litera_prefix
+          # a) foo
+          # bb) foo
+          letters:letter+ ')'
+        end
+
+        include Slaw::Grammars::Inlines
+        include Slaw::Grammars::Tables
+        include Slaw::Grammars::Schedules
+        include Slaw::Grammars::Terminals
+      end
+    end
+  end
+end
+