Re: [xml] libxml2 issue with corrupted data

[Daniel Veillard <veillard redhat com>]

On Thu, Aug 02, 2012 at 02:43:16PM +0000, Gidon Sidis wrote:
[...]
> when the parser reaches to the <Tag> section than there is a call to startElement however there is no call 
> for the endElement for at least 200 ms. What happens after this 200ms is that a new XML message arrives and 
> our application calls xmlFreeParserCtxt for the ctxt member and call xmlParseChunk again thus losing the 
> info the previous XML (not a good behavior on our side).
>
> However if there is 1000ms delay between the messages than the parser succeeds to parse the XML and there 
> is call to endElement.
>
> Also if the string value in <Tag> is not messed up than the parse calls startElement & endElement within 
> the same millisecond.
>
> Can you please tell me why it takes so long for the parser to handle corrupted data? The parser doesn't 
> fail to parse but rather takes his time... Is that an expected behavior?


  I have no idea, and can't tell without the actual input.
Provide data which reproduce the problem with xmllint --push or
xmllint --stream and then I will look at what is going on and try to
fix it. But without a way to reproduce the problem, sorry I can't
give an answer !

Daniel

-- 
Daniel Veillard      | libxml Gnome XML XSLT toolkit  http://xmlsoft.org/
daniel veillard com  | Rpmfind RPM search engine http://rpmfind.net/
http://veillard.com/ | virtualization library  http://libvirt.org/