On Wed, Oct 26, 2011 at 07:18, Bryce Nesbitt <bry...@obviously.com> wrote: > KeepRight, at http://keepright.ipax.at/ > now validates "website" tags on ways and nodes. > The script loads each page and performs a fuzzy match... > a restaurant for example needs to match on name, phone number or address > (any one will do). > > The script also checks for hijacked domains, though this detection could use > additional match strings. >
Great stuff, the best of armchair mapping. You do surveys of websites.. Is it possible to download these keepright lists in batch form? I would like to do a survey with paper of some places and then it would be cool to have your error messages available. Bugs: But you don't match phone numbers with spaces in them, at least over here people use these chars [0-9 +()[]-] to write a phone number. No OCR, and no flash parsing. (ha! joking) Could you add support for website2=*? Who reads what you write in the comment field in KeepRight? > The "http error" check could use some help also: many http errors are > transient, which leads to unfortunate flagging of non-problems. Patches welcome? -- /emj _______________________________________________ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
