def parse(instr, maxrecs, progress):
numorgs = numopps = 0
instr = re.sub(r'<(/?db):', r'<\1_', instr)
opps = re.findall(r'<VolunteerOpportunity>.+?</VolunteerOpportunity>',
instr, re.DOTALL)
volopps = ""
for i, oppstr in enumerate(opps):
#if progress and i > 0 and i % 250 == 0:
# print str(datetime.now())+": ", i, " opportunities processed."
if (maxrecs > 0 and i > maxrecs):
break
xmlh.print_rps_progress("opps", progress, i, maxrecs)
item = xmlh.simple_parser(oppstr, known_elnames, progress=False)
orgid = register_org(item)
# logoURL -- sigh, this is for the opportunity not the org
volopps += '<VolunteerOpportunity>'
volopps += xmlh.output_val('volunteerOpportunityID', str(i))
volopps += xmlh.output_val('sponsoringOrganizationID', str(orgid))
volopps += xmlh.output_node('volunteerHubOrganizationID', item,
"LocalID")
volopps += xmlh.output_node('title', item, "Title")
volopps += xmlh.output_node('abstract', item, "Description")
volopps += xmlh.output_node('description', item, "Description")
volopps += xmlh.output_node('detailURL', item, "DetailURL")
volopps += xmlh.output_val('volunteersNeeded', "-8888")
try:
oppdates = item.getElementsByTagName("OpportunityDate")
except:
oppdates = []
if len(oppdates) > 1:
print datetime.now(), \
"parse_servenet.py: only 1 OpportunityDate supported."
#return None
oppdate = oppdates[0]
elif len(oppdates) == 0:
oppdate = None
else:
oppdate = oppdates[0]
volopps += '<dateTimeDurations><dateTimeDuration>'
if oppdate:
volopps += xmlh.output_val('openEnded', 'No')
volopps += xmlh.output_val(
'duration',
'P%s%s' % (xmlh.get_tag_val(oppdate, "DurationQuantity"),
xmlh.get_tag_val(oppdate, "DurationUnit")))
volopps += xmlh.output_val('commitmentHoursPerWeek', '0')
volopps += xmlh.output_node('startDate', oppdate, "StartDate")
volopps += xmlh.output_node('endDate', oppdate, "EndDate")
else:
volopps += xmlh.output_val('openEnded', 'Yes')
volopps += xmlh.output_val('commitmentHoursPerWeek', '0')
volopps += '</dateTimeDuration></dateTimeDurations>'
volopps += '<locations>'
try:
opplocs = item.getElementsByTagName("Location")
except:
opplocs = []
for opploc in opplocs:
volopps += '<location>'
virtual_tag = opploc.getElementsByTagName("Virtual")
if virtual_tag and xmlh.get_tag_val(
opploc, "Virtual").lower() == "yes":
volopps += xmlh.output_val('virtual', 'Yes')
else:
volopps += xmlh.output_node('region', opploc,
"StateOrProvince")
volopps += xmlh.output_node('country', opploc, "Country")
volopps += xmlh.output_node('postalCode', opploc,
"ZipOrPostalCode")
volopps += '</location>'
volopps += '</locations>'
volopps += '<categoryTags/>'
volopps += '</VolunteerOpportunity>'
numopps += 1
# convert to footprint format
outstr = '<?xml version="1.0" ?>'
outstr += '<FootprintFeed schemaVersion="0.1">'
outstr += '<FeedInfo>'
outstr += xmlh.output_val('providerID', providerID)
outstr += xmlh.output_val('providerName', providerName)
outstr += xmlh.output_val('feedID', feedID)
outstr += xmlh.output_val('createdDateTime', xmlh.current_ts())
outstr += xmlh.output_val('providerURL', providerURL)
outstr += xmlh.output_val('description', feedDescription)
# TODO: capture ts -- use now?!
outstr += '</FeedInfo>'
# hardcoded: Organization
outstr += '<Organizations>'
for key in ORGS:
outstr += ORGS[key]
numorgs += 1
outstr += '</Organizations>'
outstr += '<VolunteerOpportunities>'
outstr += volopps
outstr += '</VolunteerOpportunities>'
outstr += '</FootprintFeed>'
#outstr = re.sub(r'><([^/])', r'>\n<\1', outstr)
return outstr, numorgs, numopps
def parse(instr, maxrecs, progress):
"""return FPXML given craigslist data"""
if CL_LATLONGS == None:
load_craigslist_latlongs()
xmlh.print_progress("loading craigslist crawler output...")
crawl_craigslist.parse_cache_file(instr, listings_only=True)
xmlh.print_progress("loaded " + str(len(crawl_craigslist.pages)) + " craigslist pages.")
# convert to footprint format
outstr = '<?xml version="1.0" ?>'
outstr += '<FootprintFeed schemaVersion="0.1">'
outstr += "<FeedInfo>"
outstr += xmlh.output_val("providerID", "105")
outstr += xmlh.output_val("providerName", "craigslist")
outstr += xmlh.output_val("feedID", "craigslist")
outstr += xmlh.output_val("createdDateTime", xmlh.current_ts())
outstr += xmlh.output_val("providerURL", "http://www.craigslist.org/")
outstr += "</FeedInfo>"
numorgs = numopps = 0
# no "organization" in craigslist postings
outstr += "<Organizations>"
outstr += "<Organization>"
outstr += "<organizationID>0</organizationID>"
outstr += "<nationalEIN></nationalEIN>"
outstr += "<name></name>"
outstr += "<missionStatement></missionStatement>"
outstr += "<description></description>"
outstr += "<location>"
outstr += xmlh.output_val("city", "")
outstr += xmlh.output_val("region", "")
outstr += xmlh.output_val("postalCode", "")
outstr += "</location>"
outstr += "<organizationURL></organizationURL>"
outstr += "<donateURL></donateURL>"
outstr += "<logoURL></logoURL>"
outstr += "<detailURL></detailURL>"
outstr += "</Organization>"
numorgs += 1
outstr += "</Organizations>"
skipped_listings = {}
skipped_listings["body"] = skipped_listings["title"] = skipped_listings["not-ok"] = 0
outstr += "<VolunteerOpportunities>"
for i, url in enumerate(crawl_craigslist.pages):
page = crawl_craigslist.pages[url]
ok = extract(page, "it's OK to distribute this " + "charitable volunteerism opportunity")
if ok == "":
skipped_listings["not-ok"] += 1
continue
title = extract(page, "<title>(.+?)</title>")
if title == "":
skipped_listings["title"] += 1
continue
body = extract(page, '<div id="userbody">(.+?)<')
if len(body) < 25:
skipped_listings["body"] += 1
continue
item_id = extract(url, "/vol/(.+?)[.]html$")
locstr = extract(page, "Location: (.+?)<")
datestr = extract(page, "Date: (.+?)<")
ts = dateutil.parser.parse(datestr)
datetimestr = ts.strftime("%Y-%m-%dT%H:%M:%S")
datestr = ts.strftime("%Y-%m-%d")
if maxrecs > 0 and i > maxrecs:
break
xmlh.print_rps_progress("opps", progress, i, maxrecs)
if progress and i > 0 and i % 250 == 0:
msg = "skipped " + str(skipped_listings["title"] + skipped_listings["body"])
msg += " listings (" + str(skipped_listings["title"]) + " for no-title and "
msg += str(skipped_listings["body"]) + " for short body and "
msg += str(skipped_listings["not-ok"]) + " for no-redistrib)"
xmlh.print_progress(msg)
# print "---"
# print "title:",title
# print "loc:",locstr
# print "date:",datestr
# print "body:",body[0:100]
# craigslist is full of weird escapes-- strip them
body = re.sub(r"&[a-z]+;", "", body)
title = re.sub(r"&[a-z]+;", "", title)
locstr = re.sub(r"&[a-z]+;", "", locstr)
outstr += "<VolunteerOpportunity>"
outstr += "<volunteerOpportunityID>%s</volunteerOpportunityID>" % (item_id)
outstr += "<sponsoringOrganizationIDs><sponsoringOrganizationID>0</sponsoringOrganizationID></sponsoringOrganizationIDs>"
outstr += "<volunteerHubOrganizationIDs><volunteerHubOrganizationID>0</volunteerHubOrganizationID></volunteerHubOrganizationIDs>"
outstr += "<title>%s</title>" % (title)
outstr += "<detailURL>%s</detailURL>" % (url)
# avoid CDATA in body...
esc_body = xml.sax.saxutils.escape(body)
esc_body100 = xml.sax.saxutils.escape(body[0:100])
outstr += "<description>%s</description>" % (esc_body)
outstr += "<abstract>%s</abstract>" % (esc_body100 + "...")
outstr += "<lastUpdated>%s</lastUpdated>" % (datetimestr)
# TODO: expires
# TODO: synthesize location from metro...
outstr += "<locations><location>"
outstr += "<name>%s</name>" % (xml.sax.saxutils.escape(locstr))
# what about the few that do geocode?
lat, lng = "", ""
try:
domain, unused = url.split("vol/")
lat, lng = CL_LATLONGS[domain].split(",")
except:
# ignore for now
# print url
# continue
pass
outstr += "<latitude>%s</latitude>" % (lat)
outstr += "<longitude>%s</longitude>" % (lng)
outstr += "</location></locations>"
# outstr += '<locations><location>'
# outstr += '<city>%s</city>' % (
# outstr += '<region>%s</region>' % (
# outstr += '</location></locations>'
outstr += "<dateTimeDurations><dateTimeDuration>"
outstr += "<openEnded>No</openEnded>"
outstr += "<startDate>%s</startDate>" % (datestr)
# TODO: endDate = startDate + N=14 days?
# TODO: timezone???
# outstr += '<endDate>%s</endDate>' % (
outstr += "</dateTimeDuration></dateTimeDurations>"
# TODO: categories???
# outstr += '<categoryTags>'
outstr += "</VolunteerOpportunity>"
numopps += 1
outstr += "</VolunteerOpportunities>"
outstr += "</FootprintFeed>"
# outstr = re.sub(r'><([^/])', r'>\n<\1', outstr)
return outstr, numorgs, numopps
def parse(instr, maxrecs, progress):
numorgs = numopps = 0
instr = re.sub(r'<(/?db):', r'<\1_', instr)
opps = re.findall(r'<VolunteerOpportunity>.+?</VolunteerOpportunity>',
instr, re.DOTALL)
volopps = ""
for i, oppstr in enumerate(opps):
#if progress and i > 0 and i % 250 == 0:
# print str(datetime.now())+": ", i, " opportunities processed."
if (maxrecs > 0 and i > maxrecs):
break
xmlh.print_rps_progress("opps", progress, i, maxrecs)
item = xmlh.simple_parser(oppstr, known_elnames, progress=False)
orgid = register_org(item)
# logoURL -- sigh, this is for the opportunity not the org
volopps += '<VolunteerOpportunity>'
volopps += xmlh.output_val('volunteerOpportunityID', str(i))
volopps += xmlh.output_val('sponsoringOrganizationID', str(orgid))
volopps += xmlh.output_node('volunteerHubOrganizationID', item, "LocalID")
volopps += xmlh.output_node('title', item, "Title")
volopps += xmlh.output_node('abstract', item, "Description")
volopps += xmlh.output_node('description', item, "Description")
volopps += xmlh.output_node('detailURL', item, "DetailURL")
volopps += xmlh.output_val('volunteersNeeded', "-8888")
try:
oppdates = item.getElementsByTagName("OpportunityDate")
except:
oppdates = []
if len(oppdates) > 1:
print datetime.now(), \
"parse_servenet.py: only 1 OpportunityDate supported."
#return None
oppdate = oppdates[0]
elif len(oppdates) == 0:
oppdate = None
else:
oppdate = oppdates[0]
volopps += '<dateTimeDurations><dateTimeDuration>'
if oppdate:
volopps += xmlh.output_val('openEnded', 'No')
volopps += xmlh.output_val('duration', 'P%s%s' %
(xmlh.get_tag_val(oppdate, "DurationQuantity"),
xmlh.get_tag_val(oppdate, "DurationUnit")))
volopps += xmlh.output_val('commitmentHoursPerWeek', '0')
volopps += xmlh.output_node('startDate', oppdate, "StartDate")
volopps += xmlh.output_node('endDate', oppdate, "EndDate")
else:
volopps += xmlh.output_val('openEnded', 'Yes')
volopps += xmlh.output_val('commitmentHoursPerWeek', '0')
volopps += '</dateTimeDuration></dateTimeDurations>'
volopps += '<locations>'
try:
opplocs = item.getElementsByTagName("Location")
except:
opplocs = []
for opploc in opplocs:
volopps += '<location>'
virtual_tag = opploc.getElementsByTagName("Virtual")
if virtual_tag and xmlh.get_tag_val(opploc, "Virtual").lower() == "yes":
volopps += xmlh.output_val('virtual', 'Yes')
else:
volopps += xmlh.output_node('region', opploc, "StateOrProvince")
volopps += xmlh.output_node('country', opploc, "Country")
volopps += xmlh.output_node('postalCode', opploc, "ZipOrPostalCode")
volopps += '</location>'
volopps += '</locations>'
volopps += '<categoryTags/>'
volopps += '</VolunteerOpportunity>'
numopps += 1
# convert to footprint format
outstr = '<?xml version="1.0" ?>'
outstr += '<FootprintFeed schemaVersion="0.1">'
outstr += '<FeedInfo>'
outstr += xmlh.output_val('providerID', providerID)
outstr += xmlh.output_val('providerName', providerName)
outstr += xmlh.output_val('feedID', feedID)
outstr += xmlh.output_val('createdDateTime', xmlh.current_ts())
outstr += xmlh.output_val('providerURL', providerURL)
outstr += xmlh.output_val('description', feedDescription)
# TODO: capture ts -- use now?!
outstr += '</FeedInfo>'
# hardcoded: Organization
outstr += '<Organizations>'
for key in ORGS:
outstr += ORGS[key]
numorgs += 1
outstr += '</Organizations>'
outstr += '<VolunteerOpportunities>'
outstr += volopps
outstr += '</VolunteerOpportunities>'
outstr += '</FootprintFeed>'
#outstr = re.sub(r'><([^/])', r'>\n<\1', outstr)
return outstr, numorgs, numopps
def parse(instr, maxrecs, progress):
"""return FPXML given craigslist data"""
if CL_LATLONGS == None:
load_craigslist_latlongs()
xmlh.print_progress("loading craigslist crawler output...")
crawl_craigslist.parse_cache_file(instr, listings_only=True)
xmlh.print_progress("loaded "+str(len(crawl_craigslist.pages))+" craigslist pages.")
# convert to footprint format
outstr = '<?xml version="1.0" ?>'
outstr += '<FootprintFeed schemaVersion="0.1">'
outstr += '<FeedInfo>'
outstr += xmlh.output_val('providerID', "105")
outstr += xmlh.output_val('providerName', "craigslist")
outstr += xmlh.output_val('feedID', "craigslist")
outstr += xmlh.output_val('createdDateTime', xmlh.current_ts())
outstr += xmlh.output_val('providerURL', "http://www.craigslist.org/")
outstr += '</FeedInfo>'
numorgs = numopps = 0
# no "organization" in craigslist postings
outstr += '<Organizations>'
outstr += '<Organization>'
outstr += '<organizationID>0</organizationID>'
outstr += '<nationalEIN></nationalEIN>'
outstr += '<name></name>'
outstr += '<missionStatement></missionStatement>'
outstr += '<description></description>'
outstr += '<location>'
outstr += xmlh.output_val("city", "")
outstr += xmlh.output_val("region", "")
outstr += xmlh.output_val("postalCode", "")
outstr += '</location>'
outstr += '<organizationURL></organizationURL>'
outstr += '<donateURL></donateURL>'
outstr += '<logoURL></logoURL>'
outstr += '<detailURL></detailURL>'
outstr += '</Organization>'
numorgs += 1
outstr += '</Organizations>'
skipped_listings = {}
skipped_listings["body"] = skipped_listings["title"] = \
skipped_listings["not-ok"] = 0
outstr += '<VolunteerOpportunities>'
for i, url in enumerate(crawl_craigslist.pages):
page = crawl_craigslist.pages[url]
ok = extract(page, "it's OK to distribute this "+
"charitable volunteerism opportunity")
if ok == "":
skipped_listings["not-ok"] += 1
continue
title = extract(page, "<title>(.+?)</title>")
if title == "":
skipped_listings["title"] += 1
continue
body = extract(page, '<div id="userbody">(.+?)<')
if len(body) < 25:
skipped_listings["body"] += 1
continue
item_id = extract(url, "/vol/(.+?)[.]html$")
locstr = extract(page, "Location: (.+?)<")
datestr = extract(page, "Date: (.+?)<")
ts = dateutil.parser.parse(datestr)
datetimestr = ts.strftime("%Y-%m-%dT%H:%M:%S")
datestr = ts.strftime("%Y-%m-%d")
if (maxrecs>0 and i>maxrecs):
break
xmlh.print_rps_progress("opps", progress, i, maxrecs)
if progress and i > 0 and i % 250 == 0:
msg = "skipped " + str(skipped_listings["title"]+skipped_listings["body"])
msg += " listings ("+str(skipped_listings["title"]) + " for no-title and "
msg += str(skipped_listings["body"]) + " for short body and "
msg += str(skipped_listings["not-ok"]) + " for no-redistrib)"
xmlh.print_progress(msg)
#print "---"
#print "title:",title
#print "loc:",locstr
#print "date:",datestr
#print "body:",body[0:100]
# craigslist is full of weird escapes-- strip them
body = re.sub(r'&[a-z]+;', '', body)
title = re.sub(r'&[a-z]+;', '', title)
locstr = re.sub(r'&[a-z]+;', '', locstr)
outstr += '<VolunteerOpportunity>'
outstr += '<volunteerOpportunityID>%s</volunteerOpportunityID>' % (item_id)
outstr += '<sponsoringOrganizationIDs><sponsoringOrganizationID>0</sponsoringOrganizationID></sponsoringOrganizationIDs>'
outstr += '<volunteerHubOrganizationIDs><volunteerHubOrganizationID>0</volunteerHubOrganizationID></volunteerHubOrganizationIDs>'
outstr += '<title>%s</title>' % (title)
outstr += '<detailURL>%s</detailURL>' % (url)
# avoid CDATA in body...
esc_body = xml.sax.saxutils.escape(body)
esc_body100 = xml.sax.saxutils.escape(body[0:100])
outstr += '<description>%s</description>' % (esc_body)
outstr += '<abstract>%s</abstract>' % (esc_body100 + "...")
outstr += '<lastUpdated>%s</lastUpdated>' % (datetimestr)
# TODO: expires
# TODO: synthesize location from metro...
outstr += '<locations><location>'
outstr += '<name>%s</name>' % (xml.sax.saxutils.escape(locstr))
# what about the few that do geocode?
lat, lng = "", ""
try:
domain, unused = url.split("vol/")
lat, lng = CL_LATLONGS[domain].split(",")
except:
# ignore for now
#print url
#continue
pass
outstr += '<latitude>%s</latitude>' % (lat)
outstr += '<longitude>%s</longitude>' % (lng)
outstr += '</location></locations>'
#outstr += '<locations><location>'
#outstr += '<city>%s</city>' % (
#outstr += '<region>%s</region>' % (
#outstr += '</location></locations>'
outstr += '<dateTimeDurations><dateTimeDuration>'
outstr += '<openEnded>No</openEnded>'
outstr += '<startDate>%s</startDate>' % (datestr)
# TODO: endDate = startDate + N=14 days?
# TODO: timezone???
#outstr += '<endDate>%s</endDate>' % (
outstr += '</dateTimeDuration></dateTimeDurations>'
# TODO: categories???
#outstr += '<categoryTags>'
outstr += '</VolunteerOpportunity>'
numopps += 1
outstr += '</VolunteerOpportunities>'
outstr += '</FootprintFeed>'
#outstr = re.sub(r'><([^/])', r'>\n<\1', outstr)
return outstr, numorgs, numopps
def parse(instr, maxrecs, progress):
"""return FPXML given usaservice data"""
# TODO: progress
known_elnames = [
'channel',
'db:abstract',
'db:address',
'db:attendee_count',
'db:categories',
'db:city',
'db:country',
'db:county',
'db:dateTime',
'db:event',
'db:eventType',
'db:guest_total',
'db:host',
'db:latitude',
'db:length',
'db:longitude',
'db:rsvp',
'db:scheduledTime',
'db:state',
'db:street',
'db:title',
'db:venue_name',
'db:zipcode',
'description',
'docs',
'guid',
'item',
'language',
'link',
'pubDate',
'rss',
'title',
]
# convert to footprint format
s = '<?xml version="1.0" ?>'
s += '<FootprintFeed schemaVersion="0.1">'
s += '<FeedInfo>'
# TODO: assign provider IDs?
s += '<providerID>101</providerID>'
s += '<providerName>usaservice.org</providerName>'
s += '<feedID>1</feedID>'
s += '<createdDateTime>%s</createdDateTime>' % xmlh.current_ts()
s += '<providerURL>http://www.usaservice.org/</providerURL>'
s += '<description>Syndicated events</description>'
# TODO: capture ts -- use now?!
s += '</FeedInfo>'
numorgs = numopps = 0
# hardcoded: Organization
s += '<Organizations>'
s += '<Organization>'
s += '<organizationID>0</organizationID>'
s += '<nationalEIN></nationalEIN>'
s += '<name></name>'
s += '<missionStatement></missionStatement>'
s += '<description></description>'
s += '<location><city></city><region></region><postalCode></postalCode></location>'
s += '<organizationURL></organizationURL>'
s += '<donateURL></donateURL>'
s += '<logoURL></logoURL>'
s += '<detailURL></detailURL>'
s += '</Organization>'
numorgs += 1
s += '</Organizations>'
s += '<VolunteerOpportunities>'
instr = re.sub(r'<(/?db):', r'<\1_', instr)
for i, line in enumerate(instr.splitlines()):
if (maxrecs > 0 and i > maxrecs):
break
xmlh.print_rps_progress("opps", progress, i, maxrecs)
item = xmlh.simple_parser(line, known_elnames, progress=False)
# unmapped: db_rsvp (seems to be same as link, but with #rsvp at end of url?)
# unmapped: db_host (no equivalent?)
# unmapped: db_county (seems to be empty)
# unmapped: attendee_count
# unmapped: guest_total
# unmapped: db_title (dup of title, above)
s += '<VolunteerOpportunity>'
s += '<volunteerOpportunityID>%s</volunteerOpportunityID>' % (
xmlh.get_tag_val(item, "guid"))
# hardcoded: sponsoringOrganizationID
s += '<sponsoringOrganizationIDs><sponsoringOrganizationID>0</sponsoringOrganizationID></sponsoringOrganizationIDs>'
# hardcoded: volunteerHubOrganizationID
s += '<volunteerHubOrganizationIDs><volunteerHubOrganizationID>0</volunteerHubOrganizationID></volunteerHubOrganizationIDs>'
s += '<title>%s</title>' % (xmlh.get_tag_val(item, "title"))
s += '<abstract>%s</abstract>' % (xmlh.get_tag_val(item, "abstract"))
s += '<volunteersNeeded>-8888</volunteersNeeded>'
dbscheduledTimes = item.getElementsByTagName("db_scheduledTime")
if (dbscheduledTimes.length != 1):
print datetime.now(
), "parse_usaservice: only 1 db_scheduledTime supported."
return None
dbscheduledTime = dbscheduledTimes[0]
s += '<dateTimeDurations><dateTimeDuration>'
length = xmlh.get_tag_val(dbscheduledTime, "db_length")
if length == "" or length == "-1":
s += '<openEnded>Yes</openEnded>'
else:
s += '<openEnded>No</openEnded>'
date, time = xmlh.get_tag_val(dbscheduledTime,
"db_dateTime").split(" ")
s += '<startDate>%s</startDate>' % (date)
# TODO: timezone???
s += '<startTime>%s</startTime>' % (time)
s += '</dateTimeDuration></dateTimeDurations>'
dbaddresses = item.getElementsByTagName("db_address")
if (dbaddresses.length != 1):
print datetime.now(
), "parse_usaservice: only 1 db_address supported."
return None
dbaddress = dbaddresses[0]
s += '<locations><location>'
s += '<name>%s</name>' % (xmlh.get_tag_val(item, "db_venue_name"))
s += '<streetAddress1>%s</streetAddress1>' % (xmlh.get_tag_val(
dbaddress, "db_street"))
s += '<city>%s</city>' % (xmlh.get_tag_val(dbaddress, "db_city"))
s += '<region>%s</region>' % (xmlh.get_tag_val(dbaddress, "db_state"))
s += '<country>%s</country>' % (xmlh.get_tag_val(
dbaddress, "db_country"))
s += '<postalCode>%s</postalCode>' % (xmlh.get_tag_val(
dbaddress, "db_zipcode"))
s += '<latitude>%s</latitude>' % (xmlh.get_tag_val(
item, "db_latitude"))
s += '<longitude>%s</longitude>' % (xmlh.get_tag_val(
item, "db_longitude"))
s += '</location></locations>'
type = xmlh.get_tag_val(item, "db_eventType")
s += '<categoryTags><categoryTag>%s</categoryTag></categoryTags>' % (
type)
s += '<contactName>%s</contactName>' % xmlh.get_tag_val(
item, "db_host")
s += '<detailURL>%s</detailURL>' % (xmlh.get_tag_val(item, "link"))
s += '<description>%s</description>' % (xmlh.get_tag_val(
item, "description"))
pubdate = xmlh.get_tag_val(item, "pubDate")
if re.search("[0-9][0-9] [A-Z][a-z][a-z] [0-9][0-9][0-9][0-9]",
pubdate):
# TODO: parse() is ignoring timzone...
ts = dateutil.parser.parse(pubdate)
pubdate = ts.strftime("%Y-%m-%dT%H:%M:%S")
s += '<lastUpdated>%s</lastUpdated>' % (pubdate)
s += '</VolunteerOpportunity>'
numopps += 1
s += '</VolunteerOpportunities>'
s += '</FootprintFeed>'
#s = re.sub(r'><([^/])', r'>\n<\1', s)
return s, numorgs, numopps