The Economist

bigrev

I have tried a number of different pages in the The Economist site in the last couple of weeks and I did all the troubleshooting sugestions. One of the examples is this story:

http://www.economist.com/world/unitedstates/displayStory.cfm?story_id=12918190&source=features_box_main

Can someone fix this? thanks :)

adamsmith

yes, I get a translator problem message, too. (haven't run through troubleshooting, though). In my case Zt 1.09 FF 3.05 Ubuntu 8.10

fbennett

It's being denied permission to access the DOM in a scraper function invoked by processDocuments. This may be something controlled at the site end. If I understand correctly, the way around it is to use doGet instead of processDocuments; it grabs a text copy of the HTML (instead of working from a pointer to the DOM passed through from doWeb). I'm not sure what you do from that point -- whether there is a way to instantiate a fresh DOM from the HTML in the scraper, or whether you're constrained to work from the text itself. Can anyone with technical chops provide a hint on what to do?

dstillman

An updated translator is now available that fixes single-page saves. We're still looking into what's causing the Permission Denied error, which should only be an issue with cross-domain requests.

(To answer your question, Frank, if the response text is valid XML, you can strip the DOCTYPE and use E4X to get node-based access. See the NCBI PubMed translator for an example. Otherwise you're stuck with the text.)