"C-SPAN US Senate Schedule Scraper" seems to be url-encoding the 'source_url' field.

[onyxfish]

This doesn't seem to affect every entry.

e.g.

{
   "_id": "2007-04-04T00:00:00Z -  C-SPAN US Senate Schedule Scraper",
   "_rev": "1-32ca7948af083f2df06d7cd275bf38f7",
   "datetime": "2007-04-04T00:00:00Z",
   "end_datetime": null,
   "title": "The House adjourned pursuant to H. Con. Res. 103. The next meeting is scheduled for 2:00 p.m. on April 16, 2007.",
   "description": null,
   "branch": "Legislative",
   "entity": "Senate",
   "source_url": "http%3A%2F%2Fwww3.capwiz.com%2Fc-span%2Fdbq%2Fofficials%2Fschedule.dbq%3Fcommittee%3Dus_house%26amp%3Bcommand%3Dcommittee_schedules%26amp%3Bchambername%3DSenate%26amp%3Bchamber%3DS%26amp%3Bperiod%3D",
   "source_text": "APRIL     04, 2007
\u000a\u0009The House adjourned pursuant to H. Con. Res. 103. The next meeting is scheduled for 2:00 p.m. on April 16, 2007.
",
   "access_datetime": "2009-09-27T17:16:09Z",
   "parser_name": "C-SPAN US Senate Schedule Scraper",
   "parser_version": "0.1"
}

This does kick out as an error on the CouchDB validation report, but is labelled as "not an absolute url".