Class: Epuber::Compiler::XHTMLProcessor

Inherits:
Object
  • Object
show all
Defined in:
lib/epuber/compiler/xhtml_processor.rb

Defined Under Namespace

Classes: UnparseableLinkError

Class Method Summary collapse

Class Method Details

.add_missing_root_elements(xhtml_doc, title, epub_version) ⇒ Object

Method to add all missing items in XML root

Required items:

- html (with all namespaces and other attributes)
- body
  - head (with title)

Parameters:

  • xhtml_doc (Nokogiri::XML::Document) —

    input XML document to work with

  • title (String) —

    title of this document, since this is required by EPUB specification

  • epub_version (Epuber::Version) —

    version of result EPUB

Returns:

  • nil



65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
 
# File 'lib/epuber/compiler/xhtml_processor.rb', line 65

def self.add_missing_root_elements(xhtml_doc, title, epub_version)
  # add missing body element
  if xhtml_doc.at_css('body').nil?
    xhtml_doc.root.surround_with_element('body')
  end

  # add missing root html element
  if xhtml_doc.at_css('html').nil?
    attrs               = {}
    attrs['xmlns']      = 'http://www.w3.org/1999/xhtml'
    attrs['xmlns:epub'] = 'http://www.idpf.org/2007/ops' if epub_version >= 3
    xhtml_doc.root.surround_with_element('html', attrs)
  end

  # add missing head in html
  if xhtml_doc.at_css('html > head').nil?
    html = xhtml_doc.css('html').first
    head = xhtml_doc.create_element('head')
    head << xhtml_doc.create_element('title', title)
    head << xhtml_doc.create_element('meta', charset: 'utf-8') if epub_version >= 3.0

    html.children.first.before(head)
  end
end

.add_styles(xhtml_doc, styles) ⇒ Object

Method for adding style sheets with links, method will not add duplicate items

Parameters:

  • xhtml_doc (Nokogiri::XML::Document) —

    input XML document to work with

  • styles (Array<String>) —

    links to files

Returns:

  • nil



97
98
99
100
101
102
103
104
105
106
 
# File 'lib/epuber/compiler/xhtml_processor.rb', line 97

def self.add_styles(xhtml_doc, styles)
  head  = xhtml_doc.at_css('html > head')
  old_links = head.css('link[rel="stylesheet"]').map { |node| node['href'] }

  links_to_add = styles - old_links

  links_to_add.each do |path|
    head << xhtml_doc.create_element('link', href: path, rel: 'stylesheet', type: 'text/css')
  end
end

.add_viewport(xhtml_doc, viewport_size) ⇒ Object

Adds viewport meta tag to head of some document, but only if there is not some existing tag

Parameters:

  • xhtml_doc (Nokogiri::XML::Document)
  • viewport_size (Epuber::Size) 


113
114
115
116
117
118
119
 
# File 'lib/epuber/compiler/xhtml_processor.rb', line 113

def self.add_viewport(xhtml_doc, viewport_size)
  head = xhtml_doc.at_css('html > head')
  return unless head.at_css("meta[name='viewport']").nil?

  s = viewport_size
  head << xhtml_doc.create_element('meta', name: 'viewport', content: "width=#{s.width},height=#{s.height}")
end

.resolve_images(xhtml_doc, file_path, file_resolver) ⇒ Object

Returns nil.

Parameters:

  • xhtml_doc (Nokogiri::XML::Document)
  • file_path (String) —

    path of referring file

  • file_resolver (FileResolver)

Returns:

  • nil



222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
 
# File 'lib/epuber/compiler/xhtml_processor.rb', line 222

def self.resolve_images(xhtml_doc, file_path, file_resolver)
  dirname = File.dirname(file_path)

  xhtml_doc.css('img').each do |img|
    path = img['src']
    next if path.nil?

    begin
      new_path = file_resolver.dest_finder.find_file(path, groups: :image, context_path: dirname)

    rescue UnparseableLinkError, FileFinders::FileNotFoundError, FileFinders::MultipleFilesFoundError
      begin
        new_path = resolved_link_to_file(path, :image, dirname, file_resolver.source_finder).to_s
        pkg_abs_path = File.expand_path(new_path, dirname).unicode_normalize
        pkg_new_path = Pathname.new(pkg_abs_path).relative_path_from(Pathname.new(file_resolver.source_path)).to_s

        file = FileTypes::ImageFile.new(pkg_new_path)
        file.path_type = :manifest
        file_resolver.add_file(file)

      rescue UnparseableLinkError, FileFinders::FileNotFoundError, FileFinders::MultipleFilesFoundError => e
        UI.warning(e.to_s, location: img)

        next
      end
    end

    img['src'] = new_path
  end
end

.resolve_links(xhtml_doc, file_path, file_finder) ⇒ Array<URI>

Resolves all links to files in XHTML document and returns the valid and resolved versions

Parameters:

  • xhtml_doc (Nokogiri::XML::Document) —

    input XML document to work with

  • file_path (String) —

    path to file from which is searching for other file

  • file_finder (Epuber::Compiler::FileFinder) —

    finder for searching for files

Returns:

  • (Array<URI>) —

    resolved links



201
202
203
204
205
206
 
# File 'lib/epuber/compiler/xhtml_processor.rb', line 201

def self.resolve_links(xhtml_doc, file_path, file_finder)
  [
    resolve_links_for(xhtml_doc, 'a', 'href', :text, file_path, file_finder),
    resolve_links_for(xhtml_doc, 'map > area', 'href', :text, file_path, file_finder),
  ].flatten
end

.resolve_links_for(xhtml_doc, tag_name, attribute_name, groups, file_path, file_finder) ⇒ Array<URI>

Resolves all links to files in XHTML document and returns the valid and resolved versions

Parameters:

  • xhtml_doc (Nokogiri::XML::Document) —

    input XML document to work with

  • tag_name (String) —

    CSS selector for tag

  • attribute_name (String) —

    name of attribute

  • groups (Symbol | Array<Symbol>) —

    groups of the searching file, could be for example :image when searching for file from tag <img>

  • file_path (String) —

    path to file from which is searching for other file

  • file_finder (Epuber::Compiler::FileFinder) —

    finder for searching for files

Returns:

  • (Array<URI>) —

    resolved links



168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
 
# File 'lib/epuber/compiler/xhtml_processor.rb', line 168

def self.resolve_links_for(xhtml_doc, tag_name, attribute_name, groups, file_path, file_finder)
  founded_links = []

  xhtml_doc.css("#{tag_name}[#{attribute_name}]").each do |node|
    begin
      src = node[attribute_name]
      # @type [String] src

      next if src.nil?

      target_file = resolved_link_to_file(src, groups, file_path, file_finder)
      founded_links << target_file

      node[attribute_name] = target_file.to_s
    rescue UnparseableLinkError, FileFinders::FileNotFoundError, FileFinders::MultipleFilesFoundError => e
      UI.warning(e.to_s, location: node)

      # skip not found files
      next
    end
  end

  founded_links
end

.resolved_link_to_file(path, groups, file_path, file_finder) ⇒ URI

Method which will resolve path to file from pattern

Parameters:

  • path (String) —

    pattern or path of the file

  • groups (Symbol | Array<Symbol>) —

    groups of the searching file, could be for example :image when searching for file from tag <img>

  • file_path (String) —

    path to file from which is searching for other file

  • file_finder (Epuber::Compiler::FileFinder) —

    finder for searching for files

Returns:

  • (URI) —

    resolved path to file or remote web page

Raises:

  • UnparseableLinkError, FileFinder::FileNotFoundError, FileFinder::MultipleFilesFoundError



132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
 
# File 'lib/epuber/compiler/xhtml_processor.rb', line 132

def self.resolved_link_to_file(path, groups, file_path, file_finder)
  raise FileFinders::FileNotFoundError.new(path, file_path) if path.empty?

  begin
    uri = URI(path)
  rescue URI::InvalidURIError
    begin
      uri = URI(URI::encode(path))
    rescue URI::InvalidURIError
      # skip not valid uri
      raise UnparseableLinkError, "Unparseable link `#{path}`"
    end
  end

  # skip uri with scheme (links to web pages)
  return uri unless uri.scheme.nil?

  # skip empty path
  return uri if uri.path.empty? && !uri.fragment.nil? && !uri.fragment.empty?

  uri.path = file_finder.find_file(uri.path, groups: groups, context_path: file_path)

  uri
end

.using_javascript?(xhtml_doc) ⇒ Bool

Parameters:

  • xhtml_doc (Nokogiri::XML::Document) —

    input XML document to work with

Returns:

  • (Bool) 


212
213
214
 
# File 'lib/epuber/compiler/xhtml_processor.rb', line 212

def self.using_javascript?(xhtml_doc)
  !xhtml_doc.at_css('script').nil?
end

.xml_document_from_string(text, file_path = nil) ⇒ Nokogiri::XML::Document

Method for parsing incomplete XML, supports multiple root elements

Parameters:

  • text (String) —

    input XHTML text

Returns:

  • (Nokogiri::XML::Document) —

    parsed document



22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
 
# File 'lib/epuber/compiler/xhtml_processor.rb', line 22

def self.xml_document_from_string(text, file_path = nil)
  if /\A[\n\r ]+(<\?xml)/ =~ text
    UI.warning('XML header must be at the beginning of document', location: UI::Location.new(file_path, 1))

    text = text.lstrip
  end

  doc = Nokogiri::XML(text)
  doc.encoding = 'UTF-8'
  doc.file_path = file_path

  fragment = Nokogiri::XML.fragment(text)
  root_elements = fragment.children.select { |el| el.element? }

  if root_elements.count == 1
    doc.root = root_elements.first
  elsif fragment.at_css('body').nil?
    doc.root = doc.create_element('body')

    fragment.children.select do |child|
      child.element? || child.comment? || child.text?
    end.each do |child|
      doc.root.add_child(child)
    end
  end

  doc
end