__rulesetname__ = "preen" __version__ = "0.7" __author__ = "Ted Percival " # Purpose: # 1. Align TIGER-generated tags with OpenStreetMap and geo-local conventions. # 2. Derive tags from TIGER- & GNIS-sourced names. # # In other words, preen the data. # # How to use: # Get change_tags.py from # http://trac.openstreetmap.org/browser/applications/utils/change_tags/ # and use the --module option to specify this module (preen). # # This ruleset was formerly known as gridstreetnames, but with the addition # of non-streetname-related functionality it was renamed to preen. import re def preen(tags, type): changed = False # Street name munging if strip_st_from_grid_streets(tags, type): changed = True if collapse_whitespace(tags,type): changed = True if expand_directions(tags, type): changed = True if streetname_suffixes(tags, type): changed = True # Tag detection if add_highway_refs(tags, type): changed = True if tag_places_of_worship(tags, type): changed = True return changed def created_by(): return "change_tags.py: %s %s" % (__rulesetname__, __version__) def strip_st_from_grid_streets(tags, type): """For grid streets with names such as "South 400 West St", remove the "St" or "Street" suffix because it is not used.""" if type == "way" and "highway" in tags and "name" in tags: new_name = re.sub( r'\b(N(orth)?|S(outh)?|E(ast)?|W(est)?) +(\w+) +(N(orth)?|S(outh)?|E(ast)?|W(est)?) +St(reet)?$', r'\1 \6 \7', tags["name"]) if new_name != tags["name"]: tags["name"] = new_name tags["created_by"] = created_by() return True def expand_directions(tags, type): """For ways with names such as "S 400 West", expand the "S" to "South" and so on.""" map = { r'\bN\b': r'North', r'\bS\b': r'South', r'\bE\b': r'East', r'\bW\b': r'West', } remap = {} for key, value in map.items(): pat = re.compile(key) remap[pat] = value changed = False if type == "way" and "name" in tags: for key, value in remap.items(): new_name = key.sub(value, tags["name"]) if new_name != tags["name"]: tags["name"] = new_name tags["created_by"] = created_by() changed = True return changed def collapse_whitespace(tags, type): """Collapse sequences of two or more spaces into a single space in the given object's "name" tag""" if "name" in tags: new_name = re.sub(r' +', r' ', tags["name"]) if new_name != tags["name"]: tags["name"] = new_name tags["created_by"] = created_by() return True return False def streetname_suffixes(tags, type): """Expand street name suffixes, for example St -> Street and Ave -> Avenue. """ # List of keys based on # `grep name_type area.osm | sort -u` map = { "Aly": "Alley", "Ave": "Avenue", "Blvd": "Boulevard", "Brg": "Bridge", "Cir": "Circle", "Ct": "Court", "Cv": "Cove", "Dr": "Drive", "Expy": "Expressway", "Hwy": "Highway", "Ln": "Lane", #"Loop": "Loop", "Mal": "Mall", "Pky": "Parkway", "Pl": "Place", "Rd": "Road", #"Run": "Run", "St": "Street", "Ter": "Terrace", "Tpke": "Turnpike", "Trl": "Trail", #"Way": "Way", "Xing": "Crossing", } remap = {} for key, value in map.items(): # Abbreviations must be the last word in the value pat = re.compile(" " + key + "$") remap[pat] = " " + value changed = False if type == "way" and "name" in tags: for key, value in remap.items(): new_name = key.sub(value, tags["name"]) if new_name != tags["name"]: tags["name"] = new_name tags["created_by"] = created_by() changed = True break # There will not be subsequent matches return changed def add_highway_refs(tags, type): """Detect highway reference numbers from TIGER-imports. These tend to be in a supplementary name tag (eg "name_1" or "name 1") containing a string like "State Route nn" or "United States Highway nn" Iff there is no "ref" tag, it is added based on the detected reference(s). There may be multiple references. """ # Skip elements we don't care about # In particular, if the way already has a "ref" then we don't touch it. if type != "way" or "highway" not in tags or "ref" in tags: return False changed = False new_refs = [] for key, val in tags.iteritems(): if key.startswith("name"): # Include keys like "name_1" or "name 1" match = re.match("State Route (\w+)", val) if match: new_refs.append("SR-%s" % match.group(1)) match = re.match("United States Highway (\w+)", val) if match: new_refs.append("US-%s" % match.group(1)) if len(new_refs) != 0: changed = True # TODO: sort I-* before US-* before SR-* tags["ref"] = ";".join(new_refs) if changed: tags["created_by"] = created_by() return changed def tag_places_of_worship(tags, type): """The GNIS import contains places of worship. Religion and denomination tags can be derived from the name tag in some cases.""" # Originally based on the GNIS data: # grep '|Church|' UT_Features_20090203.txt | cut -d'|' -f2 | sort -u maps = [ # lowercase_search_term, religion, denomination ["apostolic", "christian", "apostolic"], ["baptist", "christian", "baptist"], ["catholic", "christian", "catholic"], ["christ scientist", "christian", "christ_scientist"], ["christian science", "christian", "christ_scientist"], ["cogic", "christian", "pentecostal"], ["episcopal", "christian", "anglican"], ["foursquare", "christian", "foursquare"], ["greek orthodox", "christian", "greek_orthodox"], ["jehovah", "christian", "jehovahs_witness"], ["latter day saints", "christian", "mormon"], ["lutheran", "christian", "lutheran"], ["methodist", "christian", "methodist"], ["pentecostal", "christian", "pentecostal"], ["presbyterian", "christian", "presbyterian"], ["seventh day adventist","christian", "seventh_day_adventist"], ["wesleyan", "christian", "methodist"], # Non-denominational matches ["buddhist", "buddhist", None], ["christian", "christian", None], ["islamic", "muslim", None], ["scientology", "scientologist",None], # Hopefully "church" doesn't have too many false positives. # Since we are only matching amentity=place_of_worship, it should # be reasonably accurate. Must be *after* "scientolgy" ["church", "christian", None], ["synagogue", "jewish", None], ] changed = False if "amenity" in tags and tags["amenity"] == "place_of_worship": for [search_term, religion, denomination] in maps: if "name" in tags and tags["name"].lower().find(search_term) != -1: if "religion" not in tags: tags["religion"] = religion changed = True if denomination is not None and \ "denomination" not in tags and tags["religion"] == religion: tags["denomination"] = denomination changed = True break # Don't bother searching for other matches if changed: tags["created_by"] = created_by() return changed # vim: ts=4 sw=4 et