[lxml-dev] build & performance issues with 2.0beta2

[Stefan Behnel]

Hi Holger,

jholg at gmx.de wrote:
> I tried to do a little bit of performance comp. This is the stuff that
> seems to be >20% slower for me since 2.0alpha:
> 
> 0 lb54320 at adevp02 .../lxml-2.0beta2 $ /data/pydev/hjoukl/python/pysource/tools/lxml_benchcmp.py /data/tmp/pytaf/benchmarks/etree_2.0alpha.log /data/tmp/pytaf/benchmarks/etree_2.0beta2.log --tolerance 20 --loglevel MUCHSLOWER
>                    lxe: index_slice_neg (--TR T1 ):      0.02000000 <<<      0.14900000 msec/pass (+6.450000) !!!
>                    lxe: index_slice_neg (--TR T4 ):      0.00790000 <<<      0.10700000 msec/pass (+12.544304) !!!
>                   lxe: replace_children (--TC T2 ):      0.27700000 <<<      0.39790000 msec/pass (+0.436462) !!!
>                   lxe: replace_children (--TC T1 ):      0.03290000 <<<      0.04200000 msec/pass (+0.276596) !!!
>                        lxe: index_slice (--TR T3 ):      0.01100000 <<<      0.01410000 msec/pass (+0.281818) !!!
>                   lxe: replace_children (--TC T4 ):      0.03290000 <<<      0.04080000 msec/pass (+0.240122) !!!
> 0 lb54320 at adevp02 .../lxml-2.0beta2 $

Hmm, interesting. I'll look over that when I find the time. This is not
release critical.

Thanks,
Stefan