diff options
| author | Gareth Rees <gareth@mysociety.org> | 2014-04-02 12:11:32 +0100 |
|---|---|---|
| committer | Gareth Rees <gareth@mysociety.org> | 2014-04-07 17:09:30 +0100 |
| commit | e7d0f9a8b350ffe3c17451d6bb18051c7230ca61 (patch) | |
| tree | 03f5ad3335ed677e49135d6617ccd7fc09d6877a /lib/attachment_to_html | |
| parent | 44eff43ee8024a03fe4c327638ac0dbc1b47f4fd (diff) | |
Simpler AttachmentToHTML::Adapters::PDF interface
TODO: We really should be testing the full output of PDF#body, but
inconsistencies between pdftohtml prevent sensible means of doing this.
For example:
adapter.body.should == %Q(\n<A name=1></a>thisisthebody<br/>\n<hr>\n)
Fails because some versions (correctly!) use lower case tag names.
Diffstat (limited to 'lib/attachment_to_html')
| -rw-r--r-- | lib/attachment_to_html/adapters/pdf.rb | 62 |
1 files changed, 16 insertions, 46 deletions
diff --git a/lib/attachment_to_html/adapters/pdf.rb b/lib/attachment_to_html/adapters/pdf.rb index cc1bf06bc..1fca2f201 100644 --- a/lib/attachment_to_html/adapters/pdf.rb +++ b/lib/attachment_to_html/adapters/pdf.rb @@ -3,27 +3,31 @@ module AttachmentToHTML # Convert application/pdf documents in to HTML class PDF - attr_reader :attachment, :wrapper, :tmpdir + attr_reader :attachment, :tmpdir # Public: Initialize a PDF converter # # attachment - the FoiAttachment to convert to HTML # opts - a Hash of options (default: {}): - # :wrapper - String id of the div that wraps the - # attachment body # :tmpdir - String name of directory to store the # converted document def initialize(attachment, opts = {}) @attachment = attachment - @wrapper = opts.fetch(:wrapper, 'wrapper') @tmpdir = opts.fetch(:tmpdir, ::Rails.root.join('tmp')) end - # Public: Convert the attachment to HTML + # Public: The title to use in the <title> tag # # Returns a String - def to_html - @html ||= generate_html + def title + @title ||= attachment.display_filename + end + + # Public: The contents of the extracted html <body> tag + # + # Returns a String + def body + @body ||= parse_body end # Public: Was the document conversion successful? @@ -35,51 +39,17 @@ module AttachmentToHTML private - def generate_html - html = "<!DOCTYPE html>" - html += "<html>" - html += "<head>" - html += "<title>#{ title }</title>" - html += "</head>" - html += "<body>" - html += "<div id=\"#{ wrapper }\">" - html += "<div id=\"view-html-content\">" - html += body - html += "</div>" - html += "</div>" - html += "</body>" - html += "</html>" - end - - def title - @title ||= attachment.display_filename + def parse_body + match = convert.match(/<body[^>]*>(.*?)<\/body>/mi) + match ? match[1] : '' end - def body - parsed_body - end - - # Parse the output of the converted attachment so that we can pluck - # the parts we need and insert in to our own sensible template - # - # Returns a Nokogiri::HTML::Document - def parsed - @parsed ||= Nokogiri::HTML.parse(convert) - end - - def parsed_body - parsed.css('body').inner_html - end - - # Does the body element have any content, excluding HTML tags? - # - # Returns a Boolean def has_content? - !parsed.css('body').inner_text.empty? + !body.gsub(/\s+/,"").gsub(/\<[^\>]*\>/, "").empty? end def contains_images? - parsed.css('body img').any? + body.match(/<img[^>]*>/mi) ? true : false end def convert |