[Attempto] APE bug?

Kaarel Kaljurand kaljurand at gmail.com
Mon Jun 20 10:23:30 CEST 2011


Hi,

thanks for the interesting bug reports. I'll discuss the two sentences
in separate emails.

On Mon, Jun 20, 2011 at 04:19, Kenneth Jones <kennethjone at gmail.com> wrote:
> The following sentence elicits the error message "" from the APE Webclient
> at http://attempto.ifi.uzh.ch/site/tools/:
> The a:textual n:serialization of an a:XML element is the concatenation of
> "<" and the name of the element and the a:textual n:serializations of the
> attributes of the element and ">" and the a:textual n:serialization of the
> sequence of children of the element and "</" and the name of the element and
> ">".

Here there is a syntax error (missing determiner in front of
'children'). The parser correctly finds out about it but it takes it
28 seconds on my i3. I've tested with ape-6.6-101116.zip and
clex-6.5-090528.zip.

The Attempto server is easily 10 times slower than my laptop, so it
times out. I'm not sure why the "time limit exceeded" message is not
communicated to the user.

I would call this long runtime a bug. The work-around for the time
being could be a combination of:

* don't use prefixes (i.e. use a user lexicon, this also fixes the
problem of 'n:serialization' and 'n:serializations' having different
lemmas)
* avoid embedded of-constructions
* avoid the-references (in this sentence, most of them do not refer to
anything and are thus interpreted as either 'a' or 'at least 2')

--
kaarel


More information about the attempto mailing list