2010/11/26 Tom Lane :
> Pavel Stehule writes:
>> 2010/11/26 Itagaki Takahiro :
>>> Why did you change doctree and ctxt to global variables?
>>> I'm not sure why /* xmlFreeDoc(doctree); */ is commented out
>>> at the end of pgxml_xpath(), but is it enough to enable the code?
>
>> I am thinking, so
Alvaro Herrera writes:
> This looks great. As this fixes a problem that was reported to us two
> days ago, I'm interested in backpatching it. Are you going to do it?
Yeah, I'm on it. It's a bit tedious because the back branches are
different ...
regards, tom lane
--
Excerpts from Tom Lane's message of vie nov 26 16:14:08 -0300 2010:
> Those static variables are really ugly, and what's more this patch only
> stops some of the leakage. Per experimentation, the result object from
> pgxml_xpath has to be freed too, once it's been safely converted to
> whatever t
Pavel Stehule writes:
> 2010/11/26 Itagaki Takahiro :
>> Why did you change doctree and ctxt to global variables?
>> I'm not sure why /* xmlFreeDoc(doctree); */ is commented out
>> at the end of pgxml_xpath(), but is it enough to enable the code?
> I am thinking, so you must not to call xmlFreeDo
2010/11/26 Itagaki Takahiro :
> On Fri, Nov 26, 2010 at 17:59, Pavel Stehule wrote:
>> our customer showed a very significant memory leak in xml2.
>> It never release a memory allocated for context and doctree.
>
> Why did you change doctree and ctxt to global variables?
> I'm not sure why /* xmlF
On Fri, Nov 26, 2010 at 17:59, Pavel Stehule wrote:
> our customer showed a very significant memory leak in xml2.
> It never release a memory allocated for context and doctree.
Why did you change doctree and ctxt to global variables?
I'm not sure why /* xmlFreeDoc(doctree); */ is commented out
at
Hello
our customer showed a very significant memory leak in xml2.
try to
select xpath_number('' || generate_series || '','/data')
from generate_series(1,50);
attention! take all memory very fast
It never release a memory allocated for context and doctree.
Regards
Pavel Stehule
*** ./xpat
