XML library reliable?

[Mark Brownell]

On Monday, July 21, 2003, at 10:35  PM, Joël Guillod wrote:

> 1) Consider that you:
>
> set the htmltext of field 1 to that -- : ...
>
> <p>©2003 by Joël Guillod</p>
> <p><font color="#0000FF"><u>joel.guillod at net2000.ch</u></font></p>
> <p></p>
>
>
> 2) execute this test script:
>
>     on test
>       put "<fld1>" & (the HTMLtext of fld 1) & "</fld1>" into xmldata
>       put revCreateXMLTree(xmldata,false,true,false) into fldTreeID

[snip]

> 3) Why dont I get the same than the original data: i.e. why xmldata is  
> not
> xmlDataRestored? The procedure looses the diacritics (the euml for ë  
> and
> nothing for the copyright character). I actually get:
>
> <fld1><p>2003 by Jol Guillod
> joel.guillod at net2000.ch
> </p></fld1>

Warning, untested thinking,

I haven't tested for this but off hand I would say by your results that  
the XML parser is handling the text according to the rules of SGML  
correctly. Line-returns, in combination with the HTML "<p> </p>," are  
not being restored because they are not allowed in XML. Like empty  
space that is reserved for the creation of attributes in SGML, you  
might need to place all your text data on a single line. This might  
then result in a returned chunk of HTML text that will render correctly.

Example:
<p>©2003 by Joël Guillod</p><p><font  
color="#0000FF"><u>joel.guillod at net2000.ch</u></font></p><p></p>

You might further test XML by placing all your HTML text in one line &  
into an attribute in combination with including the escape-sequences  
for the quote characters.

Example:
<htmlChunk text="<p>©2003 by Joël Guillod</p><p><font color=  
"e;#0000FF"e;><u>joel.guillod at net2000.ch</u></font></p><p></ 
p>"></htmlChunk>

Mark Brownell


-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 1879 bytes
Desc: not available
URL: <http://lists.runrev.com/pipermail/use-livecode/attachments/20030722/815ca780/attachment.bin>