slaw 0.10.1 → 0.11.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 6e97aa37515b9c1ca126afd8a7ede5e1e614ffea
-  data.tar.gz: c2c85c2ac10c4da253edcde157f8a8b9ae3ad27e
+  metadata.gz: d9bf7237f83956f343675d24a55a77043dd4eaed
+  data.tar.gz: 4dbde167eb2194d5f8786c2a02b97204085b031d
 SHA512:
-  metadata.gz: 34332cfccb221e26870c8e1dfc29238b1471b8d8e7babc6fb7b66cf80faddf10a00b6d2d8b0dd1c001abf26e1940e7d15fafda2e90929630d9901d81562864e4
-  data.tar.gz: 57c9065e4f9031295ad7082ceed7d69201510939952b072f1dfa8baeadaa32432bc69777b389a651c58aea2b80865a785cea54d29ab3e41d02faf825ec08333b
+  metadata.gz: 3e204d080a241d971c49fcf06e4ff17fdd22418fcfaca3238f985c59a710895ce0d5ab61c32fde518f9885806d9c5bbf031b461b940e82c5acff85bd8daa6f3a
+  data.tar.gz: ce9e7b1b28a9799ad4d8451a0225be98707fd0ab144f10cae7a61cfb16a20a3720c975d0db08b9b5bfa36802ba874cff73f1d4fa41f46048e3f70453df99e61f

data/README.md

@@ -218,6 +218,11 @@ Akoma Ntoso `component` elements at the end of the XML document, with a name of
 
 ## Changelog
 
+### 0.11.0
+
+* Support newlines in table cells as EOL (or BR in HTML)
+* FIX unparsing of remarks, introduced in 0.10.0
+
 ### 0.10.1
 
 * Ensure backslash escaping handles listIntroductions and partial words correctly

data/lib/slaw/parse/cleanser.rb

@@ -18,13 +18,13 @@ module Slaw
         s = expand_tabs(s)
         s = chomp(s)
         s = enforce_newline(s)
-        s = remove_boilerplate(s)
       end
 
       # Run deeper introspections and reformat the text, such as
       # unwrapping/re-wrapping lines. These may not be safe to run
       # multiple times.
       def reformat(s)
+        s = remove_boilerplate(s)
         s = unbreak_lines(s)
         s = break_lines(s)
         s = strip_toc(s)
@@ -71,7 +71,7 @@ module Slaw
          .gsub(/^\s*\d+\s*$/, '')\
          .gsub(/^.*This gazette is also available.*$/, '')\
         # get rid of date lines
-         .gsub(/^\d+\s+\w+\s+\d+$/, '')\
+         .gsub(/^\d{1,2}\s+\w+\s+\d{4}$/, '')\
         # get rid of page number lines
          .gsub(/^\s*page \d+( of \d+)?\s*\n/i, '')\
          .gsub(/^\s*\d*\s*No\. \d+$/, '')\

data/lib/slaw/render/xsl/elements.xsl

@@ -113,4 +113,8 @@
     </xsl:element>
   </xsl:template>
 
+  <xsl:template match="a:eol">
+    <xsl:element name="br" />
+  </xsl:template>
+
 </xsl:stylesheet>

data/lib/slaw/version.rb

@@ -1,3 +1,3 @@
 module Slaw
-  VERSION = "0.10.1"
+  VERSION = "0.11.0"
 end

data/lib/slaw/za/act_nodes.rb

@@ -342,7 +342,9 @@ module Slaw
       class Table < Treetop::Runtime::SyntaxNode
         def to_xml(b, idprefix, i=0)
           # parse the table using wikicloth
-          html = WikiCloth::Parser.new({data: self.text_value}).to_html
+          # strip whitespace at the start of lines, to avoid wikicloth from treating it as PRE
+          text = self.text_value.strip.gsub(/^[ \t]+/, '')
+          html = WikiCloth::Parser.new({data: text}).to_html
 
           # we need to strip any surrounding p tags and add
           # an id to the table
@@ -355,12 +357,23 @@ module Slaw
             p = Nokogiri::XML::Node.new("p", html)
             p.children = cell.children
             p.parent = cell
+
+            # replace newlines with <eol>
+            p.search("text()").each do |text|
+              lines = text.content.strip.split(/\n/)
+              text.content = lines.shift
+
+              for line in lines
+                eol = text.add_next_sibling(Nokogiri::XML::Node.new("eol", html))
+                text = eol.add_next_sibling(Nokogiri::XML::Text.new(line, html))
+              end
+            end
           end
 
           table.xpath('//text()[1]').each{ |t|      t.content = t.content.lstrip }
           table.xpath('//text()[last()]').each{ |t| t.content = t.content.rstrip }
 
-          b << table.to_html
+          b.parent << table
         end
       end
 

data/lib/slaw/za/act_text.xsl

@@ -103,21 +103,13 @@
   </xsl:template>
 
   <xsl:template match="a:p">
-    <xsl:call-template name="escape">
-      <xsl:with-param name="value" select="." />
-    </xsl:call-template>
+    <xsl:apply-templates/>
     <!-- p tags must end with a newline -->
     <xsl:text>
 
 </xsl:text>
   </xsl:template>
 
-  <xsl:template match="a:listIntroduction|a:intro">
-    <xsl:call-template name="escape">
-      <xsl:with-param name="value" select="." />
-    </xsl:call-template>
-  </xsl:template>
-
   <xsl:template match="a:blockList">
     <xsl:if test="a:listIntroduction != ''">
       <xsl:apply-templates select="a:listIntroduction" />
@@ -144,6 +136,13 @@
     <xsl:apply-templates select="./*[not(self::a:intro)]" />
   </xsl:template>
 
+  <!-- first text nodes of these elems must be escaped if they have special chars -->
+  <xsl:template match="a:p/text()[1] | a:listIntroduction/text()[1] | a:intro/text()[1]">
+    <xsl:call-template name="escape">
+      <xsl:with-param name="value" select="." />
+    </xsl:call-template>
+  </xsl:template>
+
   <!-- components/schedules -->
   <xsl:template match="a:doc">
     <xsl:text>Schedule - </xsl:text>
@@ -231,6 +230,11 @@
     <xsl:text>]</xsl:text>
   </xsl:template>
 
+  <xsl:template match="a:eol">
+    <xsl:text>
+</xsl:text>
+  </xsl:template>
+
 
   <!-- for most nodes, just dump their text content -->
   <xsl:template match="*">

data/slaw.gemspec

@@ -29,4 +29,7 @@ Gem::Specification.new do |spec|
   spec.add_runtime_dependency "mimemagic", "~> 0.2.1"
   spec.add_runtime_dependency 'yomu', '~> 0.2.2'
   spec.add_runtime_dependency 'wikicloth', '~> 0.8.3'
+  # anchor twitter-text to avoid bug in 1.14.3
+  # https://github.com/twitter/twitter-text/issues/162
+  spec.add_runtime_dependency 'twitter-text', '~> 1.12.0'
 end

data/spec/generator_spec.rb

@@ -118,7 +118,7 @@ PREFACE not escaped
         <paragraph id="section-9.paragraph-0">
           <content>
             <blockList id="section-9.paragraph-0.list1">
-              <listIntroduction>(2) A special meeting:</listIntroduction>
+              <listIntroduction>(2) A special meeting <remark>[ foo ]</remark>:</listIntroduction>
               <item id="section-9.paragraph-0.list1.a">
                 <num>(a)</num>
                 <p>the chairperson so directs; or</p>
@@ -136,12 +136,75 @@ XML
       text = subject.text_from_act(doc)
       text.should == '1. Section
 
-\(2) A special meeting:
+\(2) A special meeting [[ foo ]]:
 
 (a) the chairperson so directs; or
 
 (b) a majority of the members
 
+'
+    end
+
+    it 'should unparse remarks correctly' do
+      doc = xml2doc(section(<<XML
+        <num>1.</num>
+        <paragraph id="section-19.paragraph-0">
+          <content>
+            <p>
+              <remark status="editorial">[ foo ]</remark>
+            </p>
+            <p>Section 1 <remark status="editorial">[ foo ]</remark></p>
+          </content>
+        </paragraph>
+XML
+      ))
+
+      text = subject.text_from_act(doc)
+      text.should == '1. 
+
+[[ foo ]]
+
+Section 1 [[ foo ]]
+
+'
+    end
+
+    it 'should replace eol with newlines in tables' do
+      doc = xml2doc(section(<<XML
+        <num>1.</num>
+        <table id="section-21.paragraph-0.table1">
+          <tr>
+            <td>
+              <p>foo<eol/>bar<eol/>baz</p>
+            </td>
+            <td>
+              <p>
+              one<eol/>two<eol/>three
+
+              </p>
+            </td>
+          </tr>
+        </table>'
+XML
+      ))
+
+      text = subject.text_from_act(doc)
+      text.should == '1. 
+
+{| 
+|-
+| foo
+bar
+baz
+| 
+              one
+two
+three
+
+              
+|-
+|}
+
 '
     end
   end

data/spec/za/act_spec.rb

@@ -1876,6 +1876,32 @@ EOS
 <td><p>r2c2</p></td></tr></table>'
     end
 
+    it 'should allow newlines in table cells' do
+      node = parse :table, <<EOS
+{|
+| foo
+bar
+
+baz
+|
+ one
+two
+
+ three
+|
+  four
+
+|-
+|}
+EOS
+
+      to_xml(node, "prefix.").should == '<table id="prefix.table0">
+  <tr><td><p>foo<eol/>bar<eol/><eol/>baz</p></td>
+<td><p>one<eol/>two<eol/><eol/>three</p></td>
+<td><p>four</p></td></tr>
+</table>'
+    end
+
     it 'should parse a table in a section' do
       node = parse :section, <<EOS
 10. A section title

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: slaw
 version: !ruby/object:Gem::Version
-  version: 0.10.1
+  version: 0.11.0
 platform: ruby
 authors:
 - Greg Kempe
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-12-18 00:00:00.000000000 Z
+date: 2017-01-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -150,6 +150,20 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: 0.8.3
+- !ruby/object:Gem::Dependency
+  name: twitter-text
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.12.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.12.0
 description: Slaw is a lightweight library for rendering and generating Akoma Ntoso
   acts from plain text and PDF documents.
 email: