Improve parsing speed for xmltodict

Asked Sep 24 '14 at 14:58

Active Sep 24 '14 at 14:58

Viewed 820 times

I have a compressed FIXML ZIP file. I am trying to use xmltodict to parse its uncompressed file (~130MB uncompressed data) as follows, but the parsing takes about 3 mins:

with zipfile.ZipFile(ff, 'r') as fh:
     infile = fh.read(fh.namelist()[0])
o = xmltodict.parse(infile)

I have tried to use the latest version of xmltodict as well (0.9.0) since the release notes for a prior version outlines a peformance improvement, but this still didn't help...no speed improvement at all.

Any ideas on how this xmltodict parsing could be accomplished much faster?

Thanks.

asked Sep 24 '14 at 14:58

prd999

I'm not sure we can help here. it may be machine-dependent – Caffeinated Sep 24 '14 at 15:03
I have tried this on 2 flavors of linux - ubuntu 10 (8 cores) and RHEL6 (24 cores) and same issue. – prd999 Sep 24 '14 at 19:04
or any way to trace the bottleneck when the parsing is going on to rule out any system resource issue? – prd999 Sep 24 '14 at 19:05
sounds like you're getting closers – Caffeinated Sep 24 '14 at 23:07
Sorry, not sure what you mean by closers. Could you please elaborate? Thanks. – prd999 Sep 25 '14 at 23:47

Improve parsing speed for xmltodict

0 Answers0