Re: [xml] Adding parsed 8859-1 content to a UTF-8 document ...

On Sat, Jul 19, 2008 at 03:29:32PM -0700, denverrox denver wrote:
Greetings list!

I'm using libxml2 (specifically the HTMLparser/tree modules, and the xpath library) to perform 
transformation operations on HTML input files, and have run into a character encoding issue:

Specifically, I have two HTML documents, one in 8859-1 encoding, and the other in UTF-8.

First I parse both documents into DOM trees.

Then, I'm performing an XPath on the 8859-1 document, cloning the resultset nodes using "xmlCopyNodeList," 
then using "xmlAddNextSibling" to add the 8859-1 document content into a document that was originally UTF-8 

This results in the 8859-1 content not being correctly serialized if I output the UTF-8 document.  Special 
characters are garbled, etc.

  I guess that assertion need a more precise description. Any character
in 8859-1 will en encoded with 1 or 2 bytes in UTF-8 without problem.

Based on the libxml2 encodings webpage ( 
), it seems that libxml2 converts all character encodings to UTF-8 internally. Therefore unless I'm 
misunderstanding something, the 8859-1 document should be in UTF-8 after parsing.  

  yes it is in UTF-8 internally

Is there any reason why this serialization problem should occur, if both the 8859-1 document and UTF-8 
document are converted to native UTF-8 by libxml2?  Shouldn't it "just work"? My impression is that you can 
freely copy cloned nodesets between documents, as they're all internally in UTF-8.  Careful review of the 
libXML2 encodings page seems to agree with this assertion, so I'm quite stumped.

  I don't think there should be any problem


Red Hat Virtualization group
Daniel Veillard      | virtualization library
veillard redhat com  | libxml GNOME XML XSLT toolkit | Rpmfind RPM search engine

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]