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
|
# -*- coding: utf-8 -*-
# YAML-tagger:
# Type: kommune
# Status: finished
# Name: Ås kommune
# Format: HTML
# Datatype:
# Vendor:
# Run: daily
# Missingfields: journalseqnr, journalyear, journalid
# Publish duration: 3 months
import scraperwiki
import urllib2
import lxml.html
import re
import dateutil.parser
from collections import deque
import datetime
from dateutil.relativedelta import relativedelta
scraperwiki.scrape("http://www.as.kommune.no/offentlig-journal-og-innsyn-i-arkiv.352152.no.html")
postlistelib=scraperwiki.swimport('postliste-python-lib')
agency = u'Ås kommune'
baseurl = "http://www.as.kommune.no"
print "Fetching public journal for %s!" % agency
parser = postlistelib.JournalParser(agency=agency)
fieldmap = {
'Tilhører sak:' : 'casedesc',
'Dokumentdato:' : 'docdate',
'Tilgangskode:' : 'exemption',
'Dokumenttype:' : 'doctype',
'Ansvarlig enhet:' : 'saksansvarligenhet',
}
typemap = {
u'Inngående dokument (I)' : 'I',
u'Utgående dokument (U)' : 'U',
}
class NoDataEntries(LookupError):
pass
def expand_year(year):
year = int(year)
if year > 50:
year = year + 1900
else:
year = year + 2000
return year
def parse_day_html(parser, datastore, dayurl, html):
root = lxml.html.fromstring(html)
# count = 0
for tr in root.cssselect("table.postjournal > tr"):
data = {
'agency' : parser.agency,
'scrapedurl' : dayurl,
'scrapestamputc' : datetime.datetime.now()
}
# count = count + 1
# print "=========== %d =============" % count
# print tr.text_content()
doknrroot = tr.cssselect("td div.doknr")
if not doknrroot:
# No records found, just return
msg = "No entries found in %s" % dayurl
print msg
raise NoDataEntries(msg)
arkivsaksref = doknrroot[0].text_content().strip()
caseyear = 0
caseseqnr = 0
casedocseq = 0
caseid = 'unknown'
matchObj = re.match( r'(\d+)/(\d+)\s*-\s*(\d+)$', arkivsaksref, re.M|re.I)
if matchObj:
caseyear = int(matchObj.group(1))
data['caseseqnr'] = int(matchObj.group(2))
data['casedocseq'] = int(matchObj.group(3))
data['caseyear'] = expand_year(caseyear)
data['caseid'] = str(data['caseyear']) + "/" + str(data['caseseqnr'])
data['arkivsaksref'] = arkivsaksref
else:
print "error: really broken Arkivsaksnr: %s" % arkivsaksref
raise Exception("unable to parse url %s" % dayurl)
data['docdesc'] = tr.cssselect("div.tittel")[0].text_content().strip()
datofratil = tr.cssselect("div.fratil")[0]
for dtr in tr.cssselect("table.postjournaldetaljer > tr"):
entry = dtr.cssselect('td')
heading = entry[0].text_content().strip()
if heading in fieldmap:
data[fieldmap[heading]] = entry[1].text_content()
if data['doctype'] in typemap:
data['doctype'] = typemap[data['doctype']]
else:
raise Exception("unknown document type")
if 'docdate' in data:
data['docdate'] = dateutil.parser.parse(data['docdate'],
dayfirst=True).date()
if 'exemption' in data:
data['exemption'] = data['exemption'].replace('Unntatt offentlighet, ', '')
dato, fratil = datofratil.text_content().split('-', 1)
data['recorddate'] = dateutil.parser.parse(dato.replace('Dato: ', '').strip(), dayfirst=True).date()
fratil = fratil.strip().replace('Avsender:', '').strip()
fratil = fratil.strip().replace('Mottaker:', '').strip()
if parser.is_sender_doctype(data['doctype']):
fratilfield = 'sender'
elif parser.is_recipient_doctype(data['doctype']):
fratilfield = 'recipient'
data[fratilfield] = fratil
# print data
parser.verify_entry(data)
datastore.append(data)
def fetch_day(parser, day):
datastore = []
dayurl = 'http://www.as.kommune.no/offentlig-journal-og-innsyn-i-arkiv.352152.no.html?pjdate=%s&pjfind=&pjdoktype=&cat=352152' % day.strftime('%d.%m.%Y')
html = postlistelib.fetch_url_harder(dayurl).decode('utf-8')
# print html
try:
parse_day_html(parser, datastore, dayurl, html)
scraperwiki.sqlite.save(unique_keys=['arkivsaksref'], data=datastore)
except NoDataEntries, e:
return
except Exception, e:
print html
raise
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
skiplimit = 10
# Look forward one week to at least get past the weekends
for n in xrange(skiplimit):
fetch_day(parser, newest + aday * n)
for n in xrange(skiplimit):
fetch_day(parser, oldest - aday * n)
|
