0

I am using PDFKit to render a partial view as a PDF file. However, I'm running into some issues when trying to use render_to_string (preferred method), rather than an external web request.

The pdf file when rendered using a url:

html = "#{root_url}/contracts/#{@contract.id}"
pdf = PDFKit.new(html, page_size: 'Letter').to_pdf

Proper Text

The pdf file when rendered using render_to_string:

html = render_to_string :partial => "agreement"
pdf = PDFKit.new(html , page_size: 'Letter').to_pdf
*from the console*
html => "\n\n<style>\n  #contract h2{text-align: center; margin-bottom: 10px;}\n  #contract em{font-weight: bold; text-decoration: underline; font-style: normal;}\n  #contract p.tab{margin-left: 25px;}\n  #contract ol{list-style:lower-alpha;}\n  #contract ol li{margin-left: 25px; font-size: 1em; line-height: 1.615em; color: #555; margin-bottom: 5px;}\n  #contract b{font-weight: bold;}\n  #contract p p{margin-left: 10px;}\n</style>\n<div id=\"contract\">\n <p>This agreement (“<b>Agreement</b>”) is entered into...

Malformatted Text

What am I doing wrong?

Sean S
  • 485
  • 1
  • 5
  • 14
  • You are likely trying to show unicode symbols (quotes) within the page having `ASCII` encoding. The problem has nothing to do with `pdf`, add `` in your html’s `head`. – Aleksei Matiushkin Feb 16 '13 at 09:09
  • I tried, and that doesn't seem to do the trick. The same html is provided to the PDFKit in both cases, yet only one has the poor encoding. – Sean S Feb 16 '13 at 09:20

1 Answers1

3

You have curved quotes in the source partial. The console output shows this:

...<p>This agreement (“<b>Agreement</b>”)...

Curved quotes are UTF-8 characters, however PDFKit is parsing them as ASCII by default. See this question.

EDIT: Passing in a directly-rendered string requires the string to use UTF-8 encoding. Ruby 1.9 does this by default. In Ruby 1.8, try the following:

html = render_to_string :partial => "agreement", :encoding => "UTF-8"
Community
  • 1
  • 1
Substantial
  • 6,684
  • 2
  • 31
  • 40
  • I've tried adding the encoding above as shown in your edit, as well as passing in "html.toutf8()" directly into the PDFKit. Neither appear to work :-/ – Sean S Feb 16 '13 at 16:52
  • I added to the html header as suggested in the link you provided, and it did work. Thanks! – Sean S Feb 16 '13 at 16:55
  • Adding the meta tag into the from the comment above, worked very nicely for me as well :) – Bjorn Forsberg Feb 23 '15 at 12:10