U++ forum: Welcome to the forum

Re: Small bug in DeXml [message #20298 is a reply to message #20284]

Sat, 07 March 2009 14:38

phirox
Messages: 49
Registered: December 2007

Member

Thank you both for your reactions.

Rereading this part, I have to agree that blank lines are preserved thus showing that a \n character should not be ignored by the parser and considered part of the data. As every end-of-line combination is normalized to LF for inter-compatibility; encoding CR, which is currently done, is correct. But as far as I can tell this only goes for tag elements.

Quote:

One more issue, perhaps we should use your DeXml variant for attributes....

This is where I found the problem. In my application I use the Xmlize function to store a String object. At first I did this as a tag, but later I switched to arguments because of whitespace. But the parser for arguments does not pick up \n breaks. As I believe it shouldn't, because here breaks are ignored but whitespace is not.

edit: as reply to Tomas's last post, I agree that tags should be left alone for now. And the function should only adapt for arguments as I explained above.

[Updated on: Sat, 07 March 2009 14:50]

Report message to a moderator