On Jan 13, 2006, at 3:07 PM, Steve Dunham wrote:
> I posted a new scraper for sfgate.com restaurant searches to
> simile's semantic bank. The dc:comment contains the urls for the
> generic search page, and the page with the top 100 restaurants. (I
> also posted the "top 100" restaurants themselves to the bank.)
>
> It'd be nice to have a consensus on ontology for restaurants. (And
> I wouldn't mind working out a recipe ontology with people.)
>
> I'm scraping some floating point numeric values as strings, is
> there a way to make these numeric in a piggy bank scraper?
>
>
> With this scraper I'm hitting a bug with my local piggy bank. It's
> only happening on one machine, so I don't know if the bank is in a
> weird state, or if it's because I'm using a nightly build. (It
> doesn't occur on another machine with an older piggy bank.)
>
> If I scrape a result set with more that 63 items in it, piggy bank
> starts behaving oddly. e.g. if I load:
>
> http://www.cse.msu.edu/~dunham/bank/works.n3
>
> everything is fine, but if I load:
>
> http://www.cse.msu.edu/~dunham/bank/doesnt.n3
quick 2 sec look indicates the problem might be the '&' character in
line 700 of the file. this may be the problem?
-e
>
> I get a box with "$object.getLabel()" at the top of the results
> list, and can't save or export the data.
>
> I'd appreciate if someone could let me know if they can reproduce
> it or not. (I seem to be able to reproduce this with a nightly
> build, I don't think it occurs with 2.1.1.)
>
> Thanks,
> Steve
>
Received on Fri Jan 13 2006 - 20:52:51 EST