parsing massive ODT doc in Writer?

[Jan-Marek Glogowski]

Hi Yury

Am 10.02.2016 um 00:33 schrieb Yury Tarasievich:
> A question to developers: what could even cause the drastic slowdown of
> parsing the ODT XML parts or whatever goes in the writer after the ODT
> zip archive is unpacked?
> 
> System is linux amd64, libraries compiled-in are internal.

I guess you won't find much without profiling.

As you're already building LO from source, I already wrote you how to do
the profiling. If you have problems with these instructions or the
analysis of the result, please ask on IRC. See my previous mail for
reference:

https://lists.freedesktop.org/archives/libreoffice/2016-January/072315.html

And as you already wrote, the "stock" build doesn't have these problems,
so IMHO there is not much which can be done by a different developer:

https://lists.freedesktop.org/archives/libreoffice/2016-January/072257.html

The compile options from

https://lists.freedesktop.org/archives/libreoffice/2016-January/072153.html

look generally ok

For the official build options see: distro-configs/README

Jan-Marek