OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index]
Re: [xml-dev] The text illusion
• From: David Carlisle <d.p.carlisle@gmail.com>
• To: "Costello, Roger L." <costello@mitre.org>
• Date: Fri, 15 Jul 2016 13:05:57 +0100

On 15 July 2016 at 12:52, Costello, Roger L. wrote:

Hi Folks,

What is the altitude?

<altitude>0</altitude>

The altitude is zero (0), right?

No, the altitude is 48. I will now prove it.

The content (value) of <altitude> is a single character and that character is represented by hex 30. You can see this if you open the XML in a hex editor:

Note: hex 30 equals decimal 48.

So, if we interpret the value of <altitude> as a bunch of bits representing a number, then the answer to the question is this: The altitude is 48.

Q.E.D.

My “proof” is flawed. My proof assumed the bits within <altitude> represents a number. However, bits can represent anything. XML chooses to represent all bits as an encoding of characters. In the ASCII character encoding scheme, the bits 0011 0000 (hex 30) represent the character ‘0’. So the altitude is ‘0’.

Now, you might want to convert the character ‘0’ to a number (cast the value of <altitude> to a number). Such a converter will need to know (a) the bits are an encoding of a character, (b) the bits are an encoding of an ASCII character (and not, say, an EBCDIC character), (c) the character is a character version of a digit (e.g., ‘0’ is a character version of the digit 0), and (d) how to convert the character digit to a numeric digit.

Lesson Learned: When you look at XML you might think you are seeing numbers but you are not. It’s an illusion. You are merely seeing characters.

/Roger

It would be wrong to just look at a fragment of the file content in that way and guess the encoding is ascii or otherwise. The interpretation of the xml content should always be as a sequence of unicode code points. The mapping between bits in the file encoding unicode is specified in the xml declaration at the top of the file (or other places such as http headers as specified in the xml rec)

so whether the file is ebcdic encoded or ascii or has &#x30; the interpretation should first be to U+0030  and then to interpret that as the number zero if appropriate.

David

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index]

News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 1993-2007 XML.org. This site is hosted by OASIS