Comments on: Working with UTF-8 in PDF::Writer and Ruby on Rails http://www.peterkrantz.com/2007/utf8-in-pdf-writer/ A blog about technology, visualization, music and unmanned vehicle experiments Sat, 06 Sep 2008 01:41:48 +0000 http://wordpress.org/?v=2.6.1 By: Rails PDF Plugin unter Rails2.1 | Code Schubser http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2308 Rails PDF Plugin unter Rails2.1 | Code Schubser Tue, 19 Aug 2008 16:05:44 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2308 [...] übrigens im PDF Dokument mit UTF-8 arbeiten will, sollte mal bei Peter Krantz [...] [...] übrigens im PDF Dokument mit UTF-8 arbeiten will, sollte mal bei Peter Krantz [...]

]]> By: gpm http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2306 gpm Thu, 14 Aug 2008 00:41:21 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2306 The PDF::Writer newsgroup brings this up briefly (http://groups.google.com/group/comp.lang.ruby/browse_thread/thread/ff2b849a9fc39a2b). It sounds like you need to convert to UTF-16 and specify a UxFEFF BOM before text string (or, as i understand it, the string must start with the bytes 254 then 255). I tried adding this to the solution here (http://pastie.org/252651) but it looks like the pdf is still rendering latin characters with the 254 and 255 bytes being rendered and prepended to everything. Please post if you get this working! Another alternative is to move over to prawn, which appears to support UTF-8 out of the box. The PDF::Writer newsgroup brings this up briefly (http://groups.google.com/group/comp.lang.ruby/browse_thread/thread/ff2b849a9fc39a2b). It sounds like you need to convert to UTF-16 and specify a UxFEFF BOM before text string (or, as i understand it, the string must start with the bytes 254 then 255).

I tried adding this to the solution here (http://pastie.org/252651) but it looks like the pdf is still rendering latin characters with the 254 and 255 bytes being rendered and prepended to everything.

Please post if you get this working! Another alternative is to move over to prawn, which appears to support UTF-8 out of the box.

]]> By: Andy http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2300 Andy Thu, 31 Jul 2008 20:19:26 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2300 Are there any ideas or techniques floating around here about how to do this if one is <i><b>not</b></i> only using Latin characters with diacritics? I have japanese utf-8 that has to go into PDF::Writer. Maybe I need to learn how to manipulate strings at a lower level or understand unicode more thoroughly, but here's what I've got so far - Currently in PDF::Writer: Iconv.conv('utf-16BE', 'utf-8', 'ローラー') ==> 0í0ü0é0ü While irb: Iconv.conv('utf-16BE', 'utf-8', 'ローラー') ==> "0\3550\3740\3510\374" Are there any ideas or techniques floating around here about how to do this if one is not only using Latin characters with diacritics? I have japanese utf-8 that has to go into PDF::Writer. Maybe I need to learn how to manipulate strings at a lower level or understand unicode more thoroughly, but here’s what I’ve got so far -

Currently in PDF::Writer:
Iconv.conv(’utf-16BE’, ‘utf-8′, ‘ローラー’) ==> 0í0ü0é0ü

While irb:
Iconv.conv(’utf-16BE’, ‘utf-8′, ‘ローラー’) ==> “0\3550\3740\3510\374″

]]> By: Boris http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2280 Boris Tue, 24 Jun 2008 13:51:32 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2280 I also ran into the recursion error "stack level too deep" and resolved it like this: alias_method :old_add_text, :add_text unless method_defined?(:old_add_text) This prevents the method to be overridden more often than once. Cheers, boris I also ran into the recursion error “stack level too deep” and resolved it like this:

alias_method :old_add_text, :add_text unless method_defined?(:old_add_text)

This prevents the method to be overridden more often than once.

Cheers,
boris

]]> By: BuGo http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2263 BuGo Thu, 15 May 2008 15:34:28 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2263 Am i the only one who gets : stack level too deep ? Am i the only one who gets :

stack level too deep

?

]]> By: Erik L. Underbjerg http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2247 Erik L. Underbjerg Wed, 23 Apr 2008 14:35:09 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2247 Brilliant! This was just what I needed :-) Brilliant! This was just what I needed :-)

]]> By: kikito and fjuan http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2236 kikito and fjuan Fri, 28 Mar 2008 15:50:00 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2236 We love you We love you

]]> By: Kai http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2166 Kai Fri, 21 Dec 2007 13:48:17 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2166 <code> def add_text(x, y, text, *args) old_add_text(x,y,ICONV_CONVERTER.iconv(text), *args) end </code> does not work with Time objects. The trick is to call iconv with text.to_s instead of just text. Then you are able to print dates (for example to put Time.now in the page footer). 

def add_text(x, y, text, *args)
  old_add_text(x,y,ICONV_CONVERTER.iconv(text), *args)
end

does not work with Time objects. The trick is to call iconv with text.to_s instead of just text. Then you are able to print dates (for example to put Time.now in the page footer).

]]> By: Pete http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2137 Pete Thu, 06 Dec 2007 18:05:40 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2137 I want to add that its better to use the method "add_text" instead of "text" because SimpleTable uses "add_text" and "text" uses "add_text" internally. I want to add that its better to use the method “add_text” instead of “text” because SimpleTable uses “add_text” and “text” uses “add_text” internally.

]]> By: Pete http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2136 Pete Thu, 06 Dec 2007 16:09:29 +0000 http://www.peterkrantz.com/2007/utf8-in-pdf-writer/#comment-2136 I have a small question: Why is the converter a constant? I have a small question: Why is the converter a constant?

]]>