Recommendations about xmltramp

[Jonas Hei]

I'd like to get experts' opinion about
xmltramp (http://www.aaronsw.com/2002/xmltramp/).

I am still in early stages of exploration - I need to
determine to process XML in my next project.
Currently I like ElementTree (@ effbot.org) and xmltramp.
But I am bit concerned about Mark Pilgrim's comments about
xmltree(http://diveintomark.org/archives/bli...20031218035253)

Can anyone elaborate/explain about scenarios in which xmltramp can corrupt
data?
Or was it all just a joke...

[Istvan Albert]

Jonas Hei wrote:

Can anyone elaborate/explain about scenarios in which xmltramp can
corrupt data?

xmltramp is a XML data reading framework. It cannot corrupt data.
It may parse it incorrectly though.

Istvan.

[Jonas Hei]

On Fri, 17 Sep 2004 13:39:36 -0400, Istvan Albert <ia*****@mailblocks.com>
wrote:

xmltramp is a XML data reading framework. It cannot corrupt data.
It may parse it incorrectly though.

Istvan.

It appears to be a little bit more than that.
Check out 'Mutations and re-serialization' section at
http://www.xml.com/lpt/a/2003/12/17/py-xml.html

[Nelson Minar]

"Jonas Hei" <ma******@hotmail.com> writes:

I'd like to get experts' opinion about
xmltramp (http://www.aaronsw.com/2002/xmltramp/).

I don't know about expert, but I've used xmltramp for quick hacks on a
bunch of projects and I love it. It's the simplest API I've ever used
for accessing XML. It builds trees in memory so it's not going to work
for 100 megabyte documents. And it's fast and loose enough that I
might be nervous about using it for very meticulous work, although I
have no specific evidence of it failing. Overall it's really nice.

Here's a sample:

rssFeed = urllib.urlopen("http://.../index.rss")
rss = xmltramp.parse(rssFeed.read())

print rss.channel.title
for i in rss.channel:
if i._name == 'item':
print i.title

More here:
http://www.nelson.monkey.org/~nelson.../xmltramp.html

[djw]

Jonas Hei wrote:

I'd like to get experts' opinion about
xmltramp (http://www.aaronsw.com/2002/xmltramp/).

I am still in early stages of exploration - I need to
determine to process XML in my next project.
Currently I like ElementTree (@ effbot.org) and xmltramp.
But I am bit concerned about Mark Pilgrim's comments about
xmltree(http://diveintomark.org/archives/bli...20031218035253)

Can anyone elaborate/explain about scenarios in which xmltramp can
corrupt data?
Or was it all just a joke...

The PyRXP parser (http://www.reportlab.org/pyrxp.html) lets you access
XML in a way similar to xmltree using the TagWrapper class in
xmlutils.xml. I've used it and it has worked very well (its also very fast).

-Don

[Istvan Albert]

Jonas Hei wrote:

It appears to be a little bit more than that.
Check out 'Mutations and re-serialization' section at
http://www.xml.com/lpt/a/2003/12/17/py-xml.html

Oh, I see, this feature was not evident from the
xmltramp page.

But reading the information above it appears that
it is a very ad-hoc and hackish solution that might
have been the reason for the negative comment that
the original poster was inquiring about.

Istvan.