html2odt 0.3.2 → 0.3.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 4faafbc073e768bc0f1752ffe28e9227bb2a0ff7
-  data.tar.gz: fba8f90511c1205ec52d7cd466f3573ab2627f9e
+  metadata.gz: d6e392a2bb293ff1f5ddade2b2584369040903a6
+  data.tar.gz: a95aa2ca695d94b779ea9357cdc9a282871e1a52
 SHA512:
-  metadata.gz: 70d0313da825e9fc5e426716d570891f675e1c3ab772149d64c8668011eefd60de1915e2e736f2650726567f11c8eebbe97b48c6e698bfd314df6076611b166a
-  data.tar.gz: d2830075e1fe36f1be2e78056d3f582396c77b40e7282c503e7cfd899e8571967fef6f75fededef12f6744200ae050a8277e9a7d794c3c6be2c6f0ddbf4d8bd5
+  metadata.gz: 6bd769ca10d9cb05d365da4fc37dced0136b5509e94ff21e527ae50f920cb3c470cc4031e4b534856ff5d86ab96f402f6abe88f9df77da12ae07052c989f6ee0
+  data.tar.gz: 201dc464672f62141c80486f8a6984e89f56d793829936eedfe23c97b5e6eb86187bf56d2febbf0273c1eda938414a92b90c64ec59c7594e77709ed5e8634f46

data/CHANGELOG.md

@@ -1,3 +1,8 @@
+# v0.3.3 - 2016-06-07
+
+Properly handle HTTP errors on remote image handling. Improved handling of top
+level inline elements.
+
 # v0.3.2 - 2016-06-07
 
 Properly handle errors on remote image handling, bump nokogiri dependency to

data/Rakefile

@@ -5,6 +5,7 @@ Rake::TestTask.new(:test) do |t|
   t.libs << "test"
   t.libs << "lib"
   t.test_files = FileList['test/**/*_test.rb']
+  t.warning = true
 end
 
 task :default => :test

data/html2odt.gemspec

@@ -26,5 +26,5 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency "bundler", "~> 1.12"
   spec.add_development_dependency "rake", "~> 10.0"
   spec.add_development_dependency "minitest", "~> 5.0"
-  spec.add_development_dependency "byebug"
+  spec.add_development_dependency "fakeweb", "~> 1.3"
 end

data/lib/html2odt/document.rb

@@ -17,6 +17,7 @@ class Html2Odt::Document
   def initialize(template: Html2Odt::ODT_TEMPLATE, html: nil)
     @html     = html
     @template = template
+    @base_uri = nil
 
     read_xmls
   end
@@ -246,13 +247,8 @@ class Html2Odt::Document
   def fix_document_structure(html)
     doc = Nokogiri::HTML::DocumentFragment.parse(html)
 
-    # XHTML2ODT cannot handle <code> elements without parent block elements,
-    # i.e. a containing <pre> or <p>. Adding a <p> in that case
-    doc.css("code").select { |code| code.ancestors("p, pre").empty? }.each do |code|
-      p = create_node(doc, "p")
-      code.add_next_sibling(p)
-      p.add_child(code)
-    end
+    # Removing undesired elements
+    doc.css("script, object, embed, iframe, style, link, map, area").remove
 
     # XHTML2ODT cannot handle <br> within <pre> tags properly, replacing them
     # with new lines should have the same side effects.
@@ -260,9 +256,46 @@ class Html2Odt::Document
       br.replace("\n")
     end
 
+    # XHTML2ODT cannot handle inline nodes without containing block elements, so
+    # we're wrapping anything, that's a top-level inline element or text node
+    # into a newly created p tag, trying to join all sibling inline elements
+    # into a single paragraph.
+    children = doc.children.to_a
+
+    previous = nil
+    while !children.empty?
+      child = children.shift
+      if inline_node?(child)
+        if previous
+          previous.add_child(child)
+        elsif child.element? or child.text !~ /\A\s*\z/
+          p = create_node(doc, "p")
+          child.add_next_sibling(p)
+          p.add_child(child)
+
+          previous = p
+        end
+      else
+        previous = nil
+      end
+    end
+
+
     doc.to_xml(:save_with => Nokogiri::XML::Node::SaveOptions::AS_XML)
   end
 
+  def inline_node? node
+    return true if node.text?
+
+    # https://developer.mozilla.org/en-US/docs/Web/HTML/Inline_elements
+    [
+      "b", "big", "i", "small", "tt", "abbr", "acronym", "cite", "code", "dfn",
+      "em", "kbd", "strong", "samp", "time", "var", "a", "bdo", "br", "img",
+      "map", "object", "q", "script", "span", "sub", "sup", "button", "input",
+      "label", "select", "textarea"
+    ].include?(node.name)
+  end
+
   def fix_links(html)
     doc = Nokogiri::HTML::DocumentFragment.parse(html)
 
@@ -364,18 +397,19 @@ class Html2Odt::Document
   end
 
   def uri_to_file(uri)
-    file = Tempfile.new("html2odt")
-    file.binmode
-
     Net::HTTP.start(uri.host, uri.port, use_ssl: uri.scheme == "https") do |http|
       resp = http.get(uri.path)
 
+      return nil unless resp.is_a?(Net::HTTPSuccess)
+
+      file = Tempfile.new("html2odt")
+      file.binmode
+
       file.write(resp.body)
       file.flush
+
       file
     end
-
-    file
   rescue
     # Could not fetch remote image
     #

data/lib/html2odt/image.rb

@@ -1,8 +1,9 @@
 class Html2Odt::Image
-  attr_accessor :source
+  attr_reader :source
 
   def initialize(target_base)
     @target_base = target_base
+    @valid = nil
   end
 
   # Assign file instead of source, if you were creating tempfiles and need them

data/lib/html2odt/version.rb

@@ -1,3 +1,3 @@
 module Html2Odt
-  VERSION = "0.3.2"
+  VERSION = "0.3.3"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: html2odt
 version: !ruby/object:Gem::Version
-  version: 0.3.2
+  version: 0.3.3
 platform: ruby
 authors:
 - Gregor Schmidt (Planio)
@@ -95,19 +95,19 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '5.0'
 - !ruby/object:Gem::Dependency
-  name: byebug
+  name: fakeweb
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ~>
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '1.3'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ~>
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '1.3'
 description: html2odt generates ODT documents based on HTML fragments using xhtml2odt
 email:
 - gregor@plan.io