slaw 10.2.0 → 10.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f76b66445a595e5130d1a3af8b153498d41007416bcd7c5ced418d4339030303
-  data.tar.gz: cd28f4839c8ecd4b430a6111a3c5ec3df764af84674b1e6205d54ac27b3c734f
+  metadata.gz: 85950871314a64110b980b50860cae66fce1bc9536374fca99204506150da04b
+  data.tar.gz: 8dc252e644f14502f652b0daa4e18bcaadaf9d310be23b9a8f7cbffe3b85123f
 SHA512:
-  metadata.gz: f04d34041c3dc21f552b81af8d331d4440cd7c36d94e38007d15be8b467cca15b2bedbc1b5d1fb7d652ebd2b263f695a02afd337ce20df5c0e5eca7b1ec02050
-  data.tar.gz: dae3ef8b12f13761543a90b1711b301af9a3368cbde2387c8c00adf717c6a297ddfcfe108b9ef53073b370178574877a978fb8968d63317ed736ca1c49417232
+  metadata.gz: d56cf86ff1e6502b82d6d79e76a4630e9259babcf57aaee54cc5ef2435b59a1a8ec5f6a60d58b81992cac392352f6db1acdce11aac18fc8ec461f2f8e1a92ef4
+  data.tar.gz: 19f716a0069a1d157a6f2407b1f2d9ef4610447b31b5c7113a4b7ec090e19b6e6981573c4f37baa08f2e60e8ad5864122827a3e0454b29df68f958a81500f8b9

data/README.md

@@ -86,6 +86,10 @@ You can create your own grammar by creating a gem that provides these files and
 
 ## Changelog
 
+### 10.3.1 (11 January 2021)
+
+* Strip ascii, unicode general and unicode supplemental punctuation from num elements when building eIds
+
 ### 10.2.0 (4 September 2020)
 
 * support inline superscript `^^text^^`

data/lib/slaw/grammars/counters.rb

@@ -24,20 +24,32 @@ module Slaw
       # Clean a <num> value for use in an eId
       # See https://docs.oasis-open.org/legaldocml/akn-nc/v1.0/os/akn-nc-v1.0-os.html#_Toc531692306
       #
-      # The number part of the identifiers of such elements corresponds to the
+      # "The number part of the identifiers of such elements corresponds to the
       # stripping of all final punctuation, meaningless separations as well as
       # redundant characters in the content of the <num> element. The
-      # representation is case-sensitive
+      # representation is case-sensitive."
+      #
+      # Our algorithm is:
+      # 1. strip all leading and trailing whitespace and punctuation (using the unicode punctuation blocks)
+      # 2. strip all whitespace
+      # 3. replace all remaining punctuation with a hyphen.
+      #
+      # The General Punctuation block is \u2000-\u206F, and the Supplemental Punctuation block is \u2E00-\u2E7F.
       #
       # (i) -> i
       # 1.2. -> 1-2
+      # “2.3“ -> 2-3
       # 3a bis -> 3abis
       def self.clean(num)
+        # leading whitespace and punctuation
+        num = num.gsub(/^[\s\u{2000}-\u{206f}\u{2e00}-\u{2e7f}!"#$%&'()*+,\-.\/:;<=>?@\[\]^_`{|}~]+/, '')
+        # trailing whitespace and punctuation
+        num.gsub!(/[\s\u{2000}-\u{206f}\u{2e00}-\u{2e7f}!"#$%&'()*+,\-.\/:;<=>?@\[\]^_`{|}~]+$/, '')
+        # whitespace
+        num.gsub!(/\s/, '')
+        # remaining punctuation to a hyphen
+        num.gsub!(/[\u{2000}-\u{206f}\u{2e00}-\u{2e7f}!"#$%&'()*+,\-.\/:;<=>?@\[\]^_`{|}~]+/, '-')
         num
-          .gsub(/[ ()\[\]]/, '')
-          .gsub(/\.+$/, '')
-          .gsub(/^\.+/, '')
-          .gsub(/\.+/, '-')
       end
     end
   end

data/lib/slaw/version.rb

@@ -1,3 +1,3 @@
 module Slaw
-  VERSION = "10.2.0"
+  VERSION = "10.3.1"
 end

data/spec/counters_spec.rb

@@ -0,0 +1,38 @@
+# encoding: UTF-8
+
+require 'spec_helper'
+
+require 'slaw'
+
+describe Slaw::Grammars::Counters do
+  describe '#clean' do
+    it 'should remove leading and trailing punctuation' do
+      described_class.clean("").should == ""
+      described_class.clean(" ").should == ""
+      described_class.clean("( )").should == ""
+      described_class.clean("(123.4-5)").should == "123-4-5"
+      described_class.clean("(312.32.7)").should == "312-32-7"
+      described_class.clean("(312_32_7)").should == "312-32-7"
+      described_class.clean("(6)").should == "6"
+      described_class.clean("[16]").should == "16"
+      described_class.clean("(i)").should == "i"
+      described_class.clean("[i]").should == "i"
+      described_class.clean("(2bis)").should == "2bis"
+      described_class.clean('"1.2.').should == "1-2"
+      described_class.clean("1.2.").should == "1-2"
+      described_class.clean("“2.3").should == "2-3"
+      described_class.clean("2,3").should == "2-3"
+      described_class.clean("2,3, 4,").should == "2-3-4"
+      described_class.clean("3a bis").should == "3abis"
+      described_class.clean("3é").should == "3é"
+      described_class.clean(" -3a--4,9").should == "3a-4-9"
+    end
+
+    it 'should handle non-arabic numerals' do
+      # hebrew aleph
+      described_class.clean("(א)").should == "א"
+      # chinese 3
+      described_class.clean("(三)").should == "三"
+    end
+  end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: slaw
 version: !ruby/object:Gem::Version
-  version: 10.2.0
+  version: 10.3.1
 platform: ruby
 authors:
 - Greg Kempe
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-09-04 00:00:00.000000000 Z
+date: 2021-01-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -155,6 +155,7 @@ files:
 - lib/slaw/version.rb
 - lib/slaw/xml_support.rb
 - slaw.gemspec
+- spec/counters_spec.rb
 - spec/extract/extractor_spec.rb
 - spec/fixtures/community-fire-safety.xml
 - spec/generator_spec.rb
@@ -192,6 +193,7 @@ signing_key:
 specification_version: 4
 summary: A lightweight library for using Akoma Ntoso acts in Ruby.
 test_files:
+- spec/counters_spec.rb
 - spec/extract/extractor_spec.rb
 - spec/fixtures/community-fire-safety.xml
 - spec/generator_spec.rb