Re: XML Translation Problem

[Scott Klement <rpg400-l@xxxxxxxxxxxxxxxx>]

Hiya Michael,

XML doesn't require you to use an entity reference for Unicode 
characters. Provided you are using a Unicode encoding, you can just put 
the character in the document, unescaped.

The problem most people run into (whether escaped or not) is that EBCDIC 
simply doesn't contain one of the characters in their XML.  (This 
includes your "greater than or equal" character... it doesn't exist in 
any flavor of EBCDIC that I'm familiar with. (And it's not alone... 
there are perhaps 300,000 characters that exist in Unicode but aren't 
available in ASCII or EBCDIC.)

So, you can have RPG parse it into a UCS-2 (another Unicode) field, and 
it should work nicely.  But, as soon as you try to convert it to EBCDIC 
(for example, to display it on a 5250 screen, or save it to an EBCDIC 
field in a database, etc) you'll have a challenge to overcome.  This is 
where you might do a find/replace (as Henrik suggests) to convert the 
character to something else.

If you store it wholly in Unicode fields (UCS-2, UTF-8, UTF-16, etc) 
then you'll be okay. You could even send it back out in an HTML, XML, 
JSON, etc document, and be fine. However, experience has shown that on 
IBM i, folks usually need to convert it to EBCDIC at some point, and 
that's where things go awry.

-SK


On 9/5/2012 12:30 PM, Michael Naughton wrote:
>
> I'm generating the XML in a Lotus Domino agent on a Windows server
> (very much like a Visual Basic environment). Basically, right now I'm
> sending everything with an ASCII number greater than 127 as a &# ...
> string.
>