RubyGems - htmltoword - Versions diffs - 0.4.4 → 0.5 - Mend

htmltoword 0.4.4 → 0.5

Files changed (9) hide show

checksums.yaml +4 -4
data/lib/htmltoword/document.rb +11 -1
data/lib/htmltoword/version.rb +1 -1
data/lib/htmltoword/xslt/base.xslt +70 -21
data/lib/htmltoword/xslt/inline_elements.xslt +39 -0
data/lib/htmltoword/xslt/links.xslt +35 -0
data/lib/htmltoword/xslt/numbering.xslt +29 -34
data/lib/htmltoword/xslt/relations.xslt +26 -0
metadata +5 -2

checksums.yaml CHANGED

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: befba72eb576d39090834848e2d36f81a89eb55f
-  data.tar.gz: c10496fa4f4a07f6a0ef22ded7f3fede7a5be334
+  metadata.gz: e6be38e580ebe86fd332a24f97b1c75a9f5f2ae7
+  data.tar.gz: e19adf814c82af95c2ed51ea67145c4b622173be
 SHA512:
-  metadata.gz: d859816964b95338f9396118e02316406efaa921a50c522314ddb836af6b0ff29f15fa3eda2c46649bca055f3d6afcffaabf5281236cd7851fe2a424315b0f7f
-  data.tar.gz: e7e58429b762efb32c24ab52fbcbe0148ab78f339c5d350d1feeef2154cfa5aa35526e4ecd1446ce15e7e05c9807b038b3496827ff58461a355e98d05afaca6a
+  metadata.gz: 0ab229591820f8846076f63bcf6dbe4eedac8f537247f8fe8d7dd7621862f9461f50d3687d0cc4ea0cb16fa1a6a04449fadb566c1df32cd65aa1aa9676004392
+  data.tar.gz: 03f99d75484650577e4328a059431eba8e15ca71063fd47d1e492b8a60dccdce44a8c00383b6135c1ea0335f19025b771efba56fb9b56e8f2e3723159355e777

data/lib/htmltoword/document.rb CHANGED

@@ -36,10 +36,18 @@ module Htmltoword
         'word/numbering.xml'
       end
+      def relations_xml_file
+        'word/_rels/document.xml.rels'
+      end
       def numbering_xslt
         File.join(Htmltoword.config.default_xslt_path, 'numbering.xslt')
       end
+      def relations_xslt
+        File.join(Htmltoword.config.default_xslt_path, 'relations.xslt')
+      end
       def xslt_template(extras = false)
         File.join(Htmltoword.config.default_xslt_path, (extras ? 'htmltoword.xslt' : 'base.xslt'))
       end
@@ -76,7 +84,9 @@ module Htmltoword
       html = html.presence || '<body></body>'
       source = Nokogiri::HTML(html.gsub(/>\s+</, '><'))
       transform_and_replace(source, Document.numbering_xslt, Document.numbering_xml_file)
-      transform_and_replace(source, Document.xslt_template(extras), file_name, extras)
+      transform_and_replace(source, Document.relations_xslt, Document.relations_xml_file)
+      cleaned_source = Nokogiri::XSLT(File.open(File.join(Htmltoword.config.default_xslt_path, 'inline_elements.xslt'))).transform(source)
+      transform_and_replace(cleaned_source, Document.xslt_template(extras), file_name, extras)
     end
     private

data/lib/htmltoword/version.rb CHANGED

@@ -1,3 +1,3 @@
 module Htmltoword
-  VERSION = "0.4.4"
+  VERSION = "0.5"
 end

data/lib/htmltoword/xslt/base.xslt CHANGED

@@ -18,6 +18,7 @@
   <xsl:output method="xml" encoding="utf-8" omit-xml-declaration="yes" indent="yes" />
   <xsl:include href="./functions.xslt"/>
   <xsl:include href="./tables.xslt"/>
+  <xsl:include href="./links.xslt"/>
   <xsl:template match="/">
     <xsl:apply-templates />
@@ -50,14 +51,14 @@
   </xsl:template>
-  <xsl:template match="br[not(ancestor::p) and not(ancestor::div) and not(name(..)='td') and not(name(..)='li') or
+  <xsl:template match="br[not(ancestor::p) and not(ancestor::div) and not(ancestor::td|ancestor::li) or
                           (preceding-sibling::div or following-sibling::div or preceding-sibling::p or following-sibling::p)]">
     <w:p>
       <w:r></w:r>
     </w:p>
   </xsl:template>
-  <xsl:template match="br[(name(..)='li' or name(..)='td') and
+  <xsl:template match="br[(ancestor::li or ancestor::td) and
                           (preceding-sibling::div or following-sibling::div or preceding-sibling::p or following-sibling::p)]">
     <w:r>
       <w:br />
@@ -109,31 +110,72 @@
   </xsl:template>
   <xsl:template match="ol|ul">
-    <xsl:param name="local_level" select="0"/>
     <xsl:param name="global_level" select="count(preceding::ol[not(ancestor::ol or ancestor::ul)]) + count(preceding::ul[not(ancestor::ol or ancestor::ul)]) + 1"/>
     <xsl:apply-templates>
-      <xsl:with-param name="local_level" select="$local_level + 1" />
       <xsl:with-param name="global_level" select="$global_level" />
     </xsl:apply-templates>
   </xsl:template>
-  <xsl:template match="li">
-    <xsl:param name="local_level" />
+  <xsl:template name="listItem" match="li">
     <xsl:param name="global_level" />
-    <w:p>
-      <w:pPr>
-        <w:pStyle w:val="ListParagraph"></w:pStyle>
-        <w:numPr>
-          <w:ilvl w:val="{$local_level - 1}"/>
-          <w:numId w:val="{$global_level}"/>
-        </w:numPr>
-      </w:pPr>
-      <xsl:apply-templates select="*[not(name()='ol' or name()='ul')]|text()"/>
-    </w:p>
-    <xsl:apply-templates select="./ol|./ul">
-      <xsl:with-param name="local_level" select="$local_level" />
-      <xsl:with-param name="global_level" select="$global_level" />
-    </xsl:apply-templates>
+    <xsl:param name="preceding-siblings" select="0"/>
+    <xsl:for-each select="node()">
+      <xsl:choose>
+        <xsl:when test="self::br">
+          <w:p>
+            <w:pPr>
+              <w:pStyle w:val="ListParagraph"></w:pStyle>
+              <w:numPr>
+                <w:ilvl w:val="0"/>
+                <w:numId w:val="0"/>
+              </w:numPr>
+            </w:pPr>
+            <w:r></w:r>
+          </w:p>
+        </xsl:when>
+        <xsl:when test="self::ol|self::ul">
+          <xsl:apply-templates>
+            <xsl:with-param name="global_level" select="$global_level" />
+          </xsl:apply-templates>
+        </xsl:when>
+        <xsl:when test="not(descendant::li)"> <!-- simpler div, p, headings, etc... -->
+          <xsl:variable name="ilvl" select="count(ancestor::ol) + count(ancestor::ul) - 1"></xsl:variable>
+          <xsl:choose>
+            <xsl:when test="$preceding-siblings + count(preceding-sibling::*) > 0">
+              <w:p>
+                <w:pPr>
+                  <w:pStyle w:val="ListParagraph"></w:pStyle>
+                  <w:numPr>
+                    <w:ilvl w:val="0"/>
+                    <w:numId w:val="0"/>
+                  </w:numPr>
+                  <w:ind w:left="{720 * ($ilvl + 1)}"/>
+                </w:pPr>
+                <xsl:apply-templates/>
+              </w:p>
+            </xsl:when>
+            <xsl:otherwise>
+              <w:p>
+                <w:pPr>
+                  <w:pStyle w:val="ListParagraph"></w:pStyle>
+                  <w:numPr>
+                    <w:ilvl w:val="{$ilvl}"/>
+                    <w:numId w:val="{$global_level}"/>
+                  </w:numPr>
+                </w:pPr>
+                <xsl:apply-templates/>
+              </w:p>
+            </xsl:otherwise>
+          </xsl:choose>
+        </xsl:when>
+        <xsl:otherwise> <!-- mixed things div having list and stuff content... -->
+          <xsl:call-template name="listItem">
+            <xsl:with-param name="global_level" select="$global_level" />
+            <xsl:with-param name="preceding-siblings" select="$preceding-siblings + count(preceding-sibling::*)" />
+          </xsl:call-template>
+        </xsl:otherwise>
+      </xsl:choose>
+    </xsl:for-each>
   </xsl:template>
   <xsl:template match="span[not(ancestor::td) and not(ancestor::li) and (preceding-sibling::h1 or preceding-sibling::h2 or preceding-sibling::h3 or preceding-sibling::h4 or preceding-sibling::h5 or preceding-sibling::h6 or preceding-sibling::table or preceding-sibling::p or preceding-sibling::ol or preceding-sibling::ul or preceding-sibling::div or following-sibling::h1 or following-sibling::h2 or following-sibling::h3 or following-sibling::h4 or following-sibling::h5 or following-sibling::h6 or following-sibling::table or following-sibling::p or following-sibling::ol or following-sibling::ul or following-sibling::div)]
@@ -158,7 +200,14 @@
         The div template will not create a w:p because the div contains a h2. Therefore we need to wrap the inline elements span|a|small in a p here.
       </xsl:comment>
     <w:p>
-      <xsl:apply-templates />
+      <xsl:choose>
+        <xsl:when test="self::a[starts-with(@href, 'http://') or starts-with(@href, 'https://')]">
+          <xsl:call-template name="link" />
+        </xsl:when>
+        <xsl:otherwise>
+          <xsl:apply-templates />
+        </xsl:otherwise>
+      </xsl:choose>
     </w:p>
   </xsl:template>

data/lib/htmltoword/xslt/inline_elements.xslt ADDED

@@ -0,0 +1,39 @@
+<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+  <xsl:output encoding="utf-8" omit-xml-declaration="yes" indent="yes" />
+  <xsl:strip-space elements="*"/>
+  <xsl:template match="node()|@*">
+    <xsl:copy>
+      <xsl:apply-templates select="@*|node()[1]"/>
+    </xsl:copy>
+    <xsl:apply-templates select="following-sibling::node()[1]"/>
+  </xsl:template>
+  <!-- get first inline element of a sequence or text having block element siblings... -->
+  <xsl:template match="node()[self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()][parent::div|parent::li|parent::td]">
+    <div>
+      <xsl:attribute name="class"><xsl:value-of select="../@class"/></xsl:attribute>
+      <xsl:attribute name="style"><xsl:value-of select="../@style"/></xsl:attribute>
+      <xsl:call-template name="inlineElement"/>
+    </div>
+    <xsl:apply-templates select="following-sibling::node()[not((self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text())[parent::div|parent::li|parent::td])][1]"/>
+  </xsl:template>
+  <!-- get following inline elements... -->
+  <xsl:template match="a[preceding-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]]
+    |b[preceding-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]]
+    |em[preceding-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]]
+    |i[preceding-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]]
+    |small[preceding-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]]
+    |span[preceding-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]]
+    |strong[preceding-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]]
+    |u[preceding-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]]
+    |text()[preceding-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]]"
+    name="inlineElement">
+    <xsl:copy>
+      <xsl:apply-templates select="@*|node()[1]"/>
+    </xsl:copy>
+    <xsl:apply-templates select="following-sibling::node()[1][self::a|self::b|self::em|self::i|self::small|self::span|self::strong|self::u|self::text()]"/>
+  </xsl:template>
+</xsl:stylesheet>

data/lib/htmltoword/xslt/links.xslt ADDED

@@ -0,0 +1,35 @@
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"
+                xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"
+                xmlns:o="urn:schemas-microsoft-com:office:office"
+                xmlns:v="urn:schemas-microsoft-com:vml"
+                xmlns:WX="http://schemas.microsoft.com/office/word/2003/auxHint"
+                xmlns:aml="http://schemas.microsoft.com/aml/2001/core"
+                xmlns:w10="urn:schemas-microsoft-com:office:word"
+                xmlns:pkg="http://schemas.microsoft.com/office/2006/xmlPackage"
+                xmlns:msxsl="urn:schemas-microsoft-com:xslt"
+                xmlns:ext="http://www.xmllab.net/wordml2html/ext"
+                xmlns:java="http://xml.apache.org/xalan/java"
+                xmlns:str="http://exslt.org/strings"
+                xmlns:func="http://exslt.org/functions"
+                xmlns:fn="http://www.w3.org/2005/xpath-functions"
+                version="1.0"
+                exclude-result-prefixes="java msxsl ext w o v WX aml w10"
+                extension-element-prefixes="func">
+  <xsl:template match="a[starts-with(@href, 'http://') or starts-with(@href, 'https://')]" name="link">
+    <w:hyperlink>
+      <xsl:attribute name="r:id">rId<xsl:value-of select="count(preceding::a[starts-with(@href, 'http://') or starts-with(@href, 'https://')]) + 8" /></xsl:attribute>
+      <w:r>
+        <w:rPr>
+          <w:rStyle w:val="Hyperlink"/>
+          <w:color w:val="000080"/>
+          <w:u w:val="single"/>
+        </w:rPr>
+        <w:t xml:space="preserve">
+          <xsl:value-of select="."/>
+        </w:t>
+      </w:r>
+    </w:hyperlink>
+  </xsl:template>
+</xsl:stylesheet>

data/lib/htmltoword/xslt/numbering.xslt CHANGED

@@ -72,26 +72,22 @@
     </w:numbering>
   </xsl:template>
-  <xsl:template match="ol|ul">
-    <xsl:param name="local_level" select="0"/>
-    <xsl:param name="global_level" select="count(preceding::ol[not(ancestor::ol or ancestor::ul)]) + count(preceding::ul[not(ancestor::ol or ancestor::ul)]) + 1"/>
+  <xsl:template name="container" match="ol|ul">
+    <xsl:variable name="global_level" select="count(preceding::ol[not(ancestor::ol or ancestor::ul)]) + count(preceding::ul[not(ancestor::ol or ancestor::ul)]) + 1"/>
     <xsl:variable name="style" select="func:list-type(name(.), concat(' ', @style, ' '), concat(' ', @class, ' '))"/>
     <xsl:choose>
-      <xsl:when test="$local_level = 0">
+      <xsl:when test="not(ancestor::ol or ancestor::ul)">
         <w:abstractNum w:abstractNumId="{$global_level - 1}">
           <w:nsid w:val="{concat('099A08C', $global_level)}"/>
           <w:multiLevelType w:val="hybridMultilevel"/>
           <xsl:call-template name="numbering_level">
-            <xsl:with-param name="ilvl" select="$local_level"/>
+            <xsl:with-param name="ilvl" select="0"/>
             <xsl:with-param name="style" select="$style"/>
           </xsl:call-template>
-          <xsl:apply-templates>
-            <xsl:with-param name="local_level" select="$local_level + 1" />
-            <xsl:with-param name="global_level" select="$global_level" />
-          </xsl:apply-templates>
-          <xsl:if test="count(.//ol) + count(.//ul) = 0">
+          <xsl:call-template name="item"/>
+          <xsl:if test="count(.//ol|.//ul) &lt; 6">
             <xsl:call-template name="autocomplete">
-              <xsl:with-param name="ilvl" select="$local_level + 1"/>
+              <xsl:with-param name="ilvl" select="count(.//ol) + count(.//ul)"/>
               <xsl:with-param name="style" select="$style"/>
             </xsl:call-template>
           </xsl:if>
@@ -99,31 +95,25 @@
       </xsl:when>
       <xsl:otherwise>
         <xsl:call-template name="numbering_level">
-          <xsl:with-param name="ilvl" select="$local_level"/>
+          <xsl:with-param name="ilvl" select="count(ancestor::ol) + count(ancestor::ul)"/>
           <xsl:with-param name="style" select="$style"/>
         </xsl:call-template>
-        <xsl:apply-templates>
-          <xsl:with-param name="local_level" select="$local_level + 1" />
-          <xsl:with-param name="global_level" select="$global_level" />
-        </xsl:apply-templates>
-        <xsl:if test="count(.//ol) + count(.//ul) = 0">
-          <xsl:call-template name="autocomplete">
-            <xsl:with-param name="ilvl" select="$local_level + 1"/>
-            <xsl:with-param name="style" select="$style"/>
-          </xsl:call-template>
-        </xsl:if>
+        <xsl:call-template name="item"/>
       </xsl:otherwise>
     </xsl:choose>
   </xsl:template>
-  <xsl:template match="li">
-    <xsl:param name="local_level" />
-    <xsl:param name="global_level" />
-    <xsl:apply-templates select="./ol|./ul">
-      <xsl:with-param name="local_level" select="$local_level" />
-      <xsl:with-param name="global_level" select="$global_level" />
-    </xsl:apply-templates>
+  <xsl:template name="item">
+    <xsl:for-each select="node()">
+      <xsl:choose>
+        <xsl:when test="self::ol|self::ul">
+          <xsl:call-template name="container"/>
+        </xsl:when>
+        <xsl:otherwise>
+          <xsl:call-template name="item"/>
+        </xsl:otherwise>
+      </xsl:choose>
+    </xsl:for-each>
   </xsl:template>
   <xsl:template match="*">
@@ -165,13 +155,19 @@
   <xsl:template name="autocomplete">
     <xsl:param name="ilvl"/>
     <xsl:param name="style" />
-    <xsl:if test="$ilvl &lt; 6">
+    <xsl:variable name="current_level">
+      <xsl:choose>
+        <xsl:when test="$ilvl &lt; 1">1</xsl:when>
+        <xsl:otherwise><xsl:value-of select="$ilvl"/></xsl:otherwise>
+      </xsl:choose>
+    </xsl:variable>
+    <xsl:if test="$current_level &lt; 6">
       <xsl:call-template name="numbering_level">
-        <xsl:with-param name="ilvl" select="$ilvl"/>
+        <xsl:with-param name="ilvl" select="$current_level"/>
         <xsl:with-param name="style" select="$style"/>
       </xsl:call-template>
       <xsl:call-template name="autocomplete">
-        <xsl:with-param name="ilvl" select="$ilvl + 1"/>
+        <xsl:with-param name="ilvl" select="$current_level + 1"/>
         <xsl:with-param name="style" select="$style"/>
       </xsl:call-template>
     </xsl:if>
@@ -190,5 +186,4 @@
       </xsl:call-template>
     </xsl:if>
   </xsl:template>
 </xsl:stylesheet>

data/lib/htmltoword/xslt/relations.xslt ADDED

@@ -0,0 +1,26 @@
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                xmlns="http://schemas.openxmlformats.org/package/2006/relationships"
+                version="1.0">
+  <xsl:output method="xml" encoding="utf-8" omit-xml-declaration="yes" indent="yes" />
+  <xsl:template match="a[starts-with(@href, 'http://') or starts-with(@href, 'https://')]" priority="1">
+    <Relationship Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/hyperlink" Target="{@href}" TargetMode="External">
+      <xsl:attribute name="Id">rId<xsl:value-of select="count(preceding::a[starts-with(@href, 'http://') or starts-with(@href, 'https://')]) + 8"/></xsl:attribute>
+    </Relationship>
+  </xsl:template>
+  <xsl:template match="/">
+    <Relationships>
+      <Relationship Id="rId3" Type="http://schemas.microsoft.com/office/2007/relationships/stylesWithEffects" Target="stylesWithEffects.xml"/>
+      <Relationship Id="rId4" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/settings" Target="settings.xml"/>
+      <Relationship Id="rId5" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/webSettings" Target="webSettings.xml"/>
+      <Relationship Id="rId6" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/fontTable" Target="fontTable.xml"/>
+      <Relationship Id="rId7" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/theme" Target="theme/theme1.xml"/>
+      <Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/numbering" Target="numbering.xml"/>
+      <Relationship Id="rId2" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles" Target="styles.xml"/>
+      <xsl:apply-templates select="*"/>
+    </Relationships>
+  </xsl:template>
+  <xsl:template match="text()|@*"/>
+</xsl:stylesheet>

metadata CHANGED

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: htmltoword
 version: !ruby/object:Gem::Version
-  version: 0.4.4
+  version: '0.5'
 platform: ruby
 authors:
 - Nicholas Frandsen, Cristina Matonte
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2015-09-16 00:00:00.000000000 Z
+date: 2015-11-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: actionpack
@@ -144,7 +144,10 @@ files:
 - lib/htmltoword/xslt/extras.xslt
 - lib/htmltoword/xslt/functions.xslt
 - lib/htmltoword/xslt/htmltoword.xslt
+- lib/htmltoword/xslt/inline_elements.xslt
+- lib/htmltoword/xslt/links.xslt
 - lib/htmltoword/xslt/numbering.xslt
+- lib/htmltoword/xslt/relations.xslt
 - lib/htmltoword/xslt/style2.xslt
 - lib/htmltoword/xslt/tables.xslt
 homepage: http://github.com/karnov/htmltoword