Re: [Feedreader-development] Problem

[Toomas T. <to...@i-...>]

Marcus Hettlage wrote:

> The current developement version of FeedReader uses only the OpenXML 
> XDOM-Parser.
> This parser is stricter than the ICXML-parser that was used before.
> This results in the situation, that some XML-files can't be parsed, 
> i.e. loaded and used.
> This applies for example to the german-translation file where "umlaut 
> U" and other
> german specific characters are used that are not properly encoded 
> according to the
> procssing instruction at the beginning of the file. The file says that 
> it's UTF-8
> where in reality it's "Windows 12xx".

This is not problem at all. Users download files from www.feedreader.com 
and downloading is being managed by php script  where at the moment is 
print "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n..... It's quite easy 
for me to change this. Only problem is that i do not know right 
encodings. I can put diferent encoding for every translation file, if 
somebody gives me right encodings.

> This applies also to the headlines.xml and other files.
> What do you think is the best and easiest way to fix it?

What do you think is the right encoding for headlines,*.xml files? One 
possibility is to check for parsing error and if parsing error occures, 
then manually do string replace on  xml stream? From utf-8 to something 
new so that xml parser will not complain?

Can you tell me, what kind of encoding is utf-8? What are right and what 
are not right characters for this encoding. And if somebody has japanese 
characters and german characters at the same time, what kind of encoding 
is this?

> But, please remember this code is used only once for older installations,
> where these files already exist, and only for the first time the new .EXE
> is started.
>  
>     Marcus

Greetings
Toomas