Hello,
For some reason, with one of my feeds the magic parser stops when it sees '&'-character. Only products before '&' get imported. As I remove '&' characters from the feed, more products get imported and when I remove all '&'-characters, all products get imported. This only happens with one feed and another one works fine with '&'-characters.
What could be the reason for this? Has anyone else had this problem? It's very confusing.
This happens always when & is found in the feed. Also with & ä etc. (But only this one feed, not with the other one)
I don't think the problem is in Magic Parser, because also Firefox says "XML Parsing Error: undefined entity". I still don't get what's the difference between these two files, except that one is ISO-8859-1 and the working one is utf-8.
Firefox points that there is problem with & on line:
<product_description>Some text with ö</product_description>Hi,
Yes - the problem there is that the feed has been generated with HTML entities without using CDATA tags (see my earlier post). The correct formatting for the above XML should be:
<product_description><![CDATA[Some text with ö]]></product_description>It also sounds like there is a character set mis-match between the declared character set of the feed and the data within it. If the "problem" feed is declared as "ISO-8859-1" at the top but contains UTF-8 data the feed will not parse... This could happen if, for example, the system generating the feed is not making sure that the data it is pulling from a database in order to construct the feed is of the same format as that declared in the feed itself.
Cheers,
David.
I got this worked out, there was something wrong with the feed and the merchant repaired it
Hi,
This sounds like the & is not being encoded correctly within the XML. & has a special meaning within XML documents, so where you need it to appear on its own, you must use the entity:
&The only situation where this is not the case, which may apply in the other feed where you are able to add & characters without any problem, is within CDATA tags, which instruct parsers to treat all characters "as is" and not to interpret anything as being part of the XML markup, for example:
<description><![CDATA[It is ok to use & here!]]></description>Hope thie helps;
Cheers,
David.