1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
|
# coding=utf-8
# YAML-tagger:
# Type: kommune
# Status: unfinished
# Name: Narvik kommune
# Format: HTML
# Datatype:
# Vendor: Visma
# Run: daily
# Missingfields: casedocseq
# Publish duration: perhaps unlimited
import datetime
import resource
import scraperwiki
import urllib2
import urlparse
import lxml.html
import dateutil.parser
agency = "Narvik kommune"
# Point scraperwiki GUI to the start page
starturl = "https://www.narvik.kommune.no/innsyn/postliste/"
scraperwiki.scrape(starturl)
postlistelib=scraperwiki.swimport('postliste-python-lib')
def cpu_spent():
usage = resource.getrusage(resource.RUSAGE_SELF)
return getattr(usage, 'ru_utime') + getattr(usage, 'ru_stime')
def cpu_available():
available = resource.getrlimit(resource.RLIMIT_CPU)[0]
# If no limit is set, assume 20 CPU seconds as the limit to avoid
# running for more than a few minutes every time.
if 0 > available:
available = 20
return available
def saver(unique_keys, data):
# return
#print "Not saving data"
scraperwiki.sqlite.save(unique_keys, data)
def expand_year(year):
year = int(year)
if year > 50:
year = year + 1900
else:
year = year + 2000
return year
fieldmap = {
'Journaldato:' : 'recorddate',
'Brevdato:' : 'docdate',
'Hjemmel:' : 'exemption',
'Dokumentansvarlig:' : 'saksansvarlig',
'Journalpostkategori:' : 'journalcategory',
}
def fetch_postjournal_entry(parser, datastore, entryurl):
print entryurl
html = postlistelib.fetch_url_harder(entryurl)
root = lxml.html.fromstring(html)
# print html
data = {
'agency' : parser.agency,
'scrapedurl' : entryurl,
'scrapestamputc' : datetime.datetime.now(),
'casedocseq' : 'unknown',
}
entryqueue = {}
for div in root.cssselect("table.i-bgw"):
trs = div.cssselect("tr")
for tr in trs:
field = tr.cssselect("th")[0].text_content().strip()
value = tr.cssselect("td")[0].text_content().strip()
# print "F: %s V: %s" % (field, value)
entryqueue[field] = value
if field in fieldmap:
data[fieldmap[field]] = value
# print entryqueue
if 'DokumentID:' not in entryqueue:
# No such ID, move along
return False
# F: DokumentID: V: 14/26261
# F: ArkivsakID: V: 14/1861
# F: Journaldato: V: 05.12.2014
# F: Brevdato: V: 04.12.2014
# F: Tittel på saken: V: Kommuneplanens arealdel - rullering
# F: Tittel på dokumentet: V: Jernbaneverkets uttalelse -Forslag til planprogram for Kommuneplanens arealdel 2014 - 2025 og varsel om oppstart -Narvik kommune sendt fra Jernbaneverket
# F: Dokumentansvarlig: V: Pedersen, Ingrid Sværd
(journalid, docdesc) = entryqueue['DokumentID:'].strip().split('-', 1)
data['docdesc'] = docdesc.strip()
journalid = journalid.strip()
(journalyear, journalseqnr) = journalid.split('/')
journalyear = expand_year(journalyear)
journalseqnr = int(journalseqnr)
journalid = "%d/%d" % (journalyear, journalseqnr)
(caseid, casedesc) = entryqueue['ArkivsakID:'].strip().split('-', 1)
data['casedesc'] = casedesc.strip()
caseid = caseid.strip()
(caseyear, caseseqnr) = caseid.split('/')
caseyear = expand_year(caseyear)
caseseqnr = int(caseseqnr)
caseid = "%d/%d" % (caseyear, caseseqnr)
doctypemap = {
u'Innkommende dokument' : 'I',
u'Utgående dokument' : 'U',
}
# doctype
doctypestr = root.cssselect("div.sec > h1")[0].text_content()
doctype = doctypemap[doctypestr]
for field in ['docdate', 'recorddate']:
data[field] = dateutil.parser.parse(data[field],
dayfirst=True).date()
# recipient/sender
mottakere = root.cssselect("div.dokmottakere")
if mottakere:
fratil = lxml.html.fromstring(lxml.html.tostring(mottakere[0]).replace('<br>', ", ")).text_content()
if parser.is_sender_doctype(doctype):
fratilfield = 'sender'
elif parser.is_recipient_doctype(doctype):
fratilfield = 'recipient'
if fratil is not None:
data[fratilfield] = fratil
data['caseid'] = caseid
data['doctype'] = doctype
data['caseyear'] = caseyear
data['caseseqnr'] = caseseqnr
# data['casedocseq'] = int(casedocseq)
data['journalyear'] = journalyear
data['journalseqnr'] = journalseqnr
data['journalid'] = journalid
# data['saksbehandler'] = saksbehandler
# data['saksansvarlig'] = saksansvarlig.strip()
# data['saksansvarligenhet'] = saksansvarligenhet.strip()
# data['arkivsaksref'] = arkivsaksref
# data['laapenr'] = laapenr
# print data
parser.verify_entry(data)
datastore.append(data)
return False
def journalid2webid(journalid):
(year, seqnr) = str(journalid).split('/')
return "%d%06d" % (int(year), int(seqnr))
def fetch_day(parser, date):
datestr = date.strftime("%Y-%m-%d")
datastore = []
startrow = 0
rowinc = 10
totalcount = 0
while True:
url = 'https://www.narvik.kommune.no/innsyn.aspx?response=journalpost_postliste&MId1=1543&scripturi=/innsyn.aspx&skin=infolink&fradato=%sT00:00:00&startrow=%d' % (datestr, startrow)
html = postlistelib.fetch_url_harder(url)
root = lxml.html.fromstring(html)
urls = root.cssselect("a.content-link")
urllist = []
for ahref in urls:
href = ahref.attrib['href']
if -1 != href.find("response=journalpost_detaljer"):
url = urlparse.urljoin(url, href)
urllist.append(url)
# print urllist
if 0 == len(urllist):
break
for entryurl in urllist:
fetch_postjournal_entry(parser=parser, datastore=datastore,
entryurl=entryurl)
totalcount = totalcount + 1
startrow = startrow + rowinc
scraperwiki.sqlite.save(unique_keys=['journalid'], data=datastore)
return totalcount
def scrape_some_days(parser):
aday = datetime.timedelta(1) # one day delta
newest = None
try:
newest = dateutil.parser.parse(scraperwiki.sqlite.select("max(recorddate) as max from swdata")[0]["max"], dayfirst=False).date()
oldest = dateutil.parser.parse(scraperwiki.sqlite.select("min(recorddate) as min from swdata")[0]["min"], dayfirst=False).date()
except scraperwiki.sqlite.SqliteError:
# Table not created yet, ignore the error
pass
if not newest:
# Bootstrap a month ago
newest = datetime.datetime.today() - aday * 30
oldest = newest
#print oldest, newest
skiplimit = 10
totalcount = 0
# Look forward one week to at least get past the weekends, rescan
# the last day in case new records showed up in the mean time.
# Next, scan backwards, one day before the oldest entry in the
# database.
for n in range(0, skiplimit, 1):
day = newest + aday * n
# print day
totalcount = totalcount + fetch_day(parser, day)
if cpu_spent() > (cpu_available() - 3):
print "Running short on CPU time, exiting"
return 0
for n in range(-1, -skiplimit, -1):
day = oldest + aday * n
# print day
totalcount = totalcount + fetch_day(parser, day)
if cpu_spent() > (cpu_available() - 3):
print "Running short on CPU time, exiting"
return 0
print "Fetched %d journal entries" % totalcount
# Need to rescan after a while to make sure we get the entries
# that take a while to show up when moving forward. Idea: Revisit
# all days where the record date is less than 30 days after the
# scraper date, allowing records to change for 30 days until we
# stop rescraping them. But wait 15 days before scraping again,
# to avoid scraping the same day over and over.
totalcount = 0
for drec in scraperwiki.sqlite.select("DISTINCT(recorddate) as d FROM swdata WHERE JULIANDAY(scrapestamputc) - JULIANDAY(recorddate) < 30 AND JULIANDAY('now') - JULIANDAY(scrapestamputc) > 15"):
day = dateutil.parser.parse(drec['d'], dayfirst=False).date()
print day
totalcount = totalcount + fetch_day(parser, day)
if cpu_spent() > (cpu_available() - 3):
print "Running short on CPU time, exiting"
return 0
print "Rescanned %d journal entries" % totalcount
def main():
print "Fetching public journal!"
parser = postlistelib.JournalParser(agency=agency)
datastore = []
scrape_some_days(parser)
main()
|
