W3C home > Mailing lists > Public > www-svg@w3.org > July 2008

Question about getURL() (SVGT 1.2)

From: Julien Reichel <Julien.Reichel@spinetix.com>
Date: Tue, 22 Jul 2008 11:47:19 +0200
Message-ID: <364D31930BDB974AB40E22C5C2E584ED10D0FC@asterix.SPINETIX.local>
To: <www-svg@w3.org>

Hi al,

The getURL is described is section A.8.18 & A.8.20 of SVG Tiny 1.2.

1) The charset decoding rules are simplistic:
	"if the media type of the response body was in the text/*
hierarchy and specified a charset parameter, then the text must be
converted into the host programming language's native form" 
	But that will not work in most cases since the server typically
does not know the correct encoding and sends a default from its
configuration file, usually utf-8. It seems that rules from the W3C
XMLHttpRequest spec
(http://www.w3.org/TR/XMLHttpRequest/#text-response-entity-body) are
more appropriate to accommodate plain text (using the byte order marks),
XML and HTML strings. The problem is that they are not compatible with
the wording in SVGT in the case of xml and html documents. Do you have
any suggestion concerning this case ?

2) Converting first to a DOMString then parsing the XML means that the
XML may have to be modified in case it had a <?xml?> mark with charset.
Is this the intent ? 

3) If the content is binary, how is it supposed to be converted into a
DOMString ? A mapping that could make sense is that each byte of the
binary string be mapped to the character with the corresponding number
(from 0 to 255) in UNICODE. This would allow decoding with the string
access functions. 

Best regards



Julien Reichel, PhD
Technology Architect, SpinetiX S.A.
PSE-C, CH-1015, Lausanne, Switzerland

Tel:     +41 (0) 21 693 89 81
Mail:   julien.reichel@spinetix.com
Get Information : http://www.spinetix.com

Received on Tuesday, 22 July 2008 09:48:03 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:54:19 UTC