diff options
Diffstat (limited to 'lib/attachment_to_html/adapters')
| -rw-r--r-- | lib/attachment_to_html/adapters/could_not_convert.rb | 49 | ||||
| -rw-r--r-- | lib/attachment_to_html/adapters/google_docs_viewer.rb | 56 | ||||
| -rw-r--r-- | lib/attachment_to_html/adapters/pdf.rb | 97 | ||||
| -rw-r--r-- | lib/attachment_to_html/adapters/rtf.rb | 107 | ||||
| -rw-r--r-- | lib/attachment_to_html/adapters/text.rb | 63 |
5 files changed, 372 insertions, 0 deletions
diff --git a/lib/attachment_to_html/adapters/could_not_convert.rb b/lib/attachment_to_html/adapters/could_not_convert.rb new file mode 100644 index 000000000..8e4bf39dc --- /dev/null +++ b/lib/attachment_to_html/adapters/could_not_convert.rb @@ -0,0 +1,49 @@ +module AttachmentToHTML + module Adapters + class CouldNotConvert + + attr_reader :attachment + + # Public: Initialize a PDF converter + # + # attachment - the FoiAttachment to convert to HTML + # opts - a Hash of options (default: {}): + # No options currently accepted + def initialize(attachment, opts = {}) + @attachment = attachment + end + + # Public: The title to use in the <title> tag + # + # Returns a String + 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? + # As this is a fallback option and not doing anything dynamic + # we're assuming this is successful whatever the case + # + # Returns true + def success? + true + end + + private + + def parse_body + "<p>Sorry, we were unable to convert this file to HTML. " \ + "Please use the download link at the top right.</p>" + end + + end + end +end
\ No newline at end of file diff --git a/lib/attachment_to_html/adapters/google_docs_viewer.rb b/lib/attachment_to_html/adapters/google_docs_viewer.rb new file mode 100644 index 000000000..991fbb757 --- /dev/null +++ b/lib/attachment_to_html/adapters/google_docs_viewer.rb @@ -0,0 +1,56 @@ +module AttachmentToHTML + module Adapters + # Renders the attachment in a Google Docs Viewer + class GoogleDocsViewer + + attr_reader :attachment, :attachment_url + + # Public: Initialize a GoogleDocsViewer converter + # + # attachment - the FoiAttachment to convert to HTML + # opts - a Hash of options (default: {}): + # :attachment_url - a String url to the attachment for + # Google to render (default: nil) + def initialize(attachment, opts = {}) + @attachment = attachment + @attachment_url = opts.fetch(:attachment_url, nil) + end + + # Public: The title to use in the <title> tag + # + # Returns a String + 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? + # We can't really tell whether the document conversion has been + # successful as such; We're assuming that given a correctly + # constructed iframe (which is tested) that Google will make this + # Just Work. + # + # Returns true + def success? + true + end + + private + + def parse_body + %Q(<iframe src="#{ protocol }://docs.google.com/viewer?url=#{ attachment_url }&embedded=true" width="100%" height="100%" style="border: none;"></iframe>) + end + + def protocol + AlaveteliConfiguration.force_ssl ? 'https' : 'http' + end + + end + end +end diff --git a/lib/attachment_to_html/adapters/pdf.rb b/lib/attachment_to_html/adapters/pdf.rb new file mode 100644 index 000000000..1fca2f201 --- /dev/null +++ b/lib/attachment_to_html/adapters/pdf.rb @@ -0,0 +1,97 @@ +module AttachmentToHTML + module Adapters + # Convert application/pdf documents in to HTML + class PDF + + attr_reader :attachment, :tmpdir + + # Public: Initialize a PDF converter + # + # attachment - the FoiAttachment to convert to HTML + # opts - a Hash of options (default: {}): + # :tmpdir - String name of directory to store the + # converted document + def initialize(attachment, opts = {}) + @attachment = attachment + @tmpdir = opts.fetch(:tmpdir, ::Rails.root.join('tmp')) + end + + # Public: The title to use in the <title> tag + # + # Returns a String + 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? + # + # Returns a Boolean + def success? + has_content? || contains_images? + end + + private + + def parse_body + match = convert.match(/<body[^>]*>(.*?)<\/body>/mi) + match ? match[1] : '' + end + + def has_content? + !body.gsub(/\s+/,"").gsub(/\<[^\>]*\>/, "").empty? + end + + def contains_images? + body.match(/<img[^>]*>/mi) ? true : false + end + + def convert + # Get the attachment body outside of the chdir call as getting + # the body may require opening files too + text = attachment_body + + @converted ||= Dir.chdir(tmpdir) do + tempfile = create_tempfile(text) + + html = AlaveteliExternalCommand.run("pdftohtml", + "-nodrm", "-zoom", "1.0", "-stdout", "-enc", "UTF-8", + "-noframes", tempfile.path, :timeout => 30 + ) + + cleanup_tempfile(tempfile) + + html + end + end + + def create_tempfile(text) + tempfile = if RUBY_VERSION.to_f >= 1.9 + Tempfile.new('foiextract', '.', + :encoding => text.encoding) + else + Tempfile.new('foiextract', '.') + end + tempfile.print(text) + tempfile.flush + tempfile + end + + def cleanup_tempfile(tempfile) + tempfile.close + tempfile.delete + end + + def attachment_body + @attachment_body ||= attachment.body + end + + end + end +end diff --git a/lib/attachment_to_html/adapters/rtf.rb b/lib/attachment_to_html/adapters/rtf.rb new file mode 100644 index 000000000..859c0e541 --- /dev/null +++ b/lib/attachment_to_html/adapters/rtf.rb @@ -0,0 +1,107 @@ +module AttachmentToHTML + module Adapters + # Convert application/rtf documents in to HTML + class RTF + + attr_reader :attachment, :tmpdir + + # Public: Initialize a RTF converter + # + # attachment - the FoiAttachment to convert to HTML + # opts - a Hash of options (default: {}): + # :tmpdir - String name of directory to store the + # converted document + def initialize(attachment, opts = {}) + @attachment = attachment + @tmpdir = opts.fetch(:tmpdir, ::Rails.root.join('tmp')) + end + + # Public: The title to use in the <title> tag + # + # Returns a String + 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? + # + # Returns a Boolean + def success? + has_content? || contains_images? + end + + private + + def parse_body + match = convert.match(/<body[^>]*>(.*?)<\/body>/mi) + match ? match[1] : '' + end + + def has_content? + !body.gsub(/\s+/,"").gsub(/\<[^\>]*\>/, "").empty? + end + + def contains_images? + body.match(/<img[^>]*>/mi) ? true : false + end + + def convert + # Get the attachment body outside of the chdir call as getting + # the body may require opening files too + text = attachment_body + + @converted ||= Dir.chdir(tmpdir) do + tempfile = create_tempfile(text) + + html = AlaveteliExternalCommand.run("unrtf", "--html", + tempfile.path, :timeout => 120 + ) + + cleanup_tempfile(tempfile) + + sanitize_converted(html) + end + + end + + # Works around http://savannah.gnu.org/bugs/?42015 in unrtf ~> 0.21 + def sanitize_converted(html) + invalid = %Q(<!DOCTYPE html PUBLIC -//W3C//DTD HTML 4.01 Transitional//EN>) + valid = %Q(<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN>") + if html.include?(invalid) + html.sub!(invalid, valid) + end + html + end + + def create_tempfile(text) + tempfile = if RUBY_VERSION.to_f >= 1.9 + Tempfile.new('foiextract', '.', + :encoding => text.encoding) + else + Tempfile.new('foiextract', '.') + end + tempfile.print(text) + tempfile.flush + tempfile + end + + def cleanup_tempfile(tempfile) + tempfile.close + tempfile.delete + end + + def attachment_body + @attachment_body ||= attachment.body + end + + end + end +end diff --git a/lib/attachment_to_html/adapters/text.rb b/lib/attachment_to_html/adapters/text.rb new file mode 100644 index 000000000..b431ada5e --- /dev/null +++ b/lib/attachment_to_html/adapters/text.rb @@ -0,0 +1,63 @@ +require 'nokogiri' + +module AttachmentToHTML + module Adapters + # Convert text/plain documents in to HTML + class Text + + attr_reader :attachment + + # Public: Initialize a Text converter + # + # attachment - the FoiAttachment to convert to HTML + # opts - a Hash of options (default: {}): + # No options currently accepted + def initialize(attachment, opts = {}) + @attachment = attachment + end + + # Public: The title to use in the <title> tag + # + # Returns a String + 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? + # + # Returns a Boolean + def success? + has_content? || contains_images? + end + + private + + def convert + text = attachment.body.strip + text = CGI.escapeHTML(text) + text = MySociety::Format.make_clickable(text) + text = text.gsub(/\n/, '<br>') + end + + def parse_body + convert + end + + def has_content? + !body.gsub(/\s+/,"").gsub(/\<[^\>]*\>/, "").empty? + end + + def contains_images? + body.match(/<img[^>]*>/mi) ? true : false + end + + end + end +end |