aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorGareth Rees <gareth@mysociety.org>2014-04-02 12:11:32 +0100
committerGareth Rees <gareth@mysociety.org>2014-04-07 17:09:30 +0100
commite7d0f9a8b350ffe3c17451d6bb18051c7230ca61 (patch)
tree03f5ad3335ed677e49135d6617ccd7fc09d6877a
parent44eff43ee8024a03fe4c327638ac0dbc1b47f4fd (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
-rw-r--r--lib/attachment_to_html/adapters/pdf.rb62
-rw-r--r--spec/lib/attachment_to_html/adapters/pdf_spec.rb77
2 files changed, 37 insertions, 102 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
diff --git a/spec/lib/attachment_to_html/adapters/pdf_spec.rb b/spec/lib/attachment_to_html/adapters/pdf_spec.rb
index 65c376043..c02b157e4 100644
--- a/spec/lib/attachment_to_html/adapters/pdf_spec.rb
+++ b/spec/lib/attachment_to_html/adapters/pdf_spec.rb
@@ -3,94 +3,59 @@ require File.expand_path(File.dirname(__FILE__) + '/../../../spec_helper')
describe AttachmentToHTML::Adapters::PDF do
let(:attachment) { FactoryGirl.build(:pdf_attachment) }
- let(:pdf_adapter) { AttachmentToHTML::Adapters::PDF.new(attachment) }
-
- describe :wrapper do
-
- it 'defaults to wrapper' do
- pdf_adapter.wrapper.should == 'wrapper'
- end
-
- it 'accepts a wrapper option' do
- pdf_adapter = AttachmentToHTML::Adapters::PDF.new(attachment, :wrapper => 'wrap')
- pdf_adapter.wrapper.should == 'wrap'
- end
-
- end
+ let(:adapter) { AttachmentToHTML::Adapters::PDF.new(attachment) }
describe :tmpdir do
it 'defaults to the rails tmp directory' do
- pdf_adapter.tmpdir.should == Rails.root.join('tmp')
+ adapter.tmpdir.should == Rails.root.join('tmp')
end
it 'allows a tmpdir to be specified to store the converted document' do
- pdf_adapter = AttachmentToHTML::Adapters::PDF.new(attachment, :tmpdir => '/tmp')
- pdf_adapter.tmpdir.should == '/tmp'
+ adapter = AttachmentToHTML::Adapters::PDF.new(attachment, :tmpdir => '/tmp')
+ adapter.tmpdir.should == '/tmp'
end
end
- describe :to_html do
+ describe :title do
- it 'should be a valid html document' do
- parsed = Nokogiri::HTML.parse(pdf_adapter.to_html) do |config|
- config.strict
- end
- parsed.errors.any?.should be_false
- end
-
- it 'contains the attachment filename in the title tag' do
- parsed = Nokogiri::HTML.parse(pdf_adapter.to_html) do |config|
- config.strict
- end
- parsed.css('title').inner_html.should == attachment.display_filename
+ it 'uses the attachment filename for the title' do
+ adapter.title.should == attachment.display_filename
end
+
+ end
- it 'contains the wrapper div in the body tag' do
- pdf_adapter = AttachmentToHTML::Adapters::PDF.new(attachment, :wrapper => 'wrap')
- parsed = Nokogiri::HTML.parse(pdf_adapter.to_html) do |config|
- config.strict
- end
- parsed.css('body div').first.attributes['id'].value.should == 'wrap'
- end
+ describe :body do
- it 'contains the attachment body in the wrapper div' do
- pdf_adapter = AttachmentToHTML::Adapters::PDF.new(attachment, :wrapper => 'wrap')
- parsed = Nokogiri::HTML.parse(pdf_adapter.to_html) do |config|
- config.strict
- end
- parsed.css('div#wrap div#view-html-content').inner_html.should include('thisisthebody')
+ it 'extracts the body from the document' do
+ adapter.body.should include('thisisthebody')
end
it 'operates in the context of the supplied tmpdir' do
- pdf_adapter = AttachmentToHTML::Adapters::PDF.new(attachment, :tmpdir => '/tmp')
+ adapter = AttachmentToHTML::Adapters::PDF.new(attachment, :tmpdir => '/tmp')
Dir.should_receive(:chdir).with('/tmp').and_call_original
- pdf_adapter.to_html
+ adapter.body
end
end
+
describe :success? do
it 'is successful if the body has content excluding the tags' do
- pdf_adapter.to_html
- pdf_adapter.success?.should be_true
+ adapter.stub(:body).and_return('<p>some content</p>')
+ adapter.success?.should be_true
end
it 'is successful if the body contains images' do
- mocked_return = %Q(<!DOCTYPE html><html><head></head><body><img src="logo.png" /></body></html>)
- pdf_adapter = AttachmentToHTML::Adapters::PDF.new(attachment)
- pdf_adapter.stub(:to_html).and_return(mocked_return)
- pdf_adapter.success?.should be_true
+ adapter.stub(:body).and_return(%Q(<img src="logo.png" />))
+ adapter.success?.should be_true
end
it 'is not successful if the body has no content other than tags' do
- # TODO: Add and use spec/fixtures/files/empty.pdf
- attachment = FactoryGirl.build(:body_text, :body => '')
- pdf_adapter = AttachmentToHTML::Adapters::PDF.new(attachment)
- pdf_adapter.to_html
- pdf_adapter.success?.should be_false
+ adapter.stub(:body).and_return('<p></p>')
+ adapter.success?.should be_false
end
end
generated by cgit v1.2.3 (git 2.25.1) at 2025-07-19 07:14:47 +0000