On Thu, Oct 18, 2012 at 06:19:31PM +0200, Graham Leggett wrote:
> On 18 Oct 2012, at 6:07 PM, Daniel Veillard wrote:
>
> > See xmlByteConsumed() but it's more complex for us than for expat
> > as we convert the initial byte stream to UTF-8 if it was in a different
> > encoding. See the xmlByteCo
Thanks Liam
We are building a platform to which codes containing xpaths are submitted by
external users. Manual optimization of xpaths are infeasible. Do you know about
any tools that can automate it?
-
Z
On Oct 18, 2012, at 7:15 PM, Liam R E Quin wrote:
> On Thu, 2012-10-18 at 18:00 -0700,
On Thu, 2012-10-18 at 18:00 -0700, Zhigang Chen wrote:
> Hi
>
> We sometimes run into the situation where a pretty expensive xpath
> (e.g. .//table//td[@class]) is run on a big document (~ 9M) and it
> takes very very long. In fact we never see it finish.
[resending from the right account, sorry]
Hi
We sometimes run into the situation where a pretty expensive xpath (e.g.
.//table//td[@class]) is run on a big document (~ 9M) and it takes very very
long. In fact we never see it finish.
I did some digging and found that it is spending very long time on
xmlXPathNodeSetMergeAndClear. So I w
On 18 Oct 2012, at 6:07 PM, Daniel Veillard wrote:
> See xmlByteConsumed() but it's more complex for us than for expat
> as we convert the initial byte stream to UTF-8 if it was in a different
> encoding. See the xmlByteConsumed() code.
The docs say "This function provides the current index of
On Thu, Oct 18, 2012 at 04:35:25PM +0200, Graham Leggett wrote:
> Hi all,
>
> I am currently tasked with replacing the expat parser within an application
> with the more lenient html parser found in libxml2.
>
> I am using the parser to work out the location within the document of certain
> ele
Hi all,
I am currently tasked with replacing the expat parser within an application
with the more lenient html parser found in libxml2.
I am using the parser to work out the location within the document of certain
elements (tags), and once I have found the element I am looking for, I need to
k