/
scraper.py
164 lines (131 loc) · 4.71 KB
/
scraper.py
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
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
This scraper scrapes the data from the Ministry of Finance of The
Slovak Republic. It processes the PDF list of the real-estate
property of the state.
"""
# INITIAL DATA
site_url = 'https://www.finance.gov.sk/' # main page
start_page = 'Default.aspx?CatID=4733' # subpage url
header_row = 0 # use row number N as headers
column_count_row = 1 # use row number N as column indexes (optional)
ignore_rows = 2 # ignore first N rows per page (must set explicitly, even if header is set, eg. N = 1 with headers)
column_count = 38
diff_vert = 3
diff_horiz = 6
cellmap = {
66: 'ID',
79: 'ID2',
226: 'Zariadenie',
378: 'Typ',
406: 'Druh',
475: 'Druh2',
631: 'Inventárne číslo',
684: 'Rok nadobudnutia a kraj',
740: 'Názov okresu',
794: 'Názov obce',
867: 'Názov KÚ',
925: 'Ulica',
994: 'Číslo VL',
1021: 'Spoluvl. podiel',
1069: 'Výmera v m^2',
1179: 'Parcelné číslo',
1209: 'Kolaudácia a správca objektu',
1323: 'Užívateľ objektu',
1423: 'Obstarávacia cena v EUR',
1459: 'Zostatková cena v EUR',
}
import scraperwiki
import urllib2
import lxml
import lxml.html
import sys
import re
import collections
def process_columns(row):
"""
Column post-processing (accepts a row of results)
The values are inconsistent across columns - values bleed to previous/next cells, this function
attemps to create a list of consistent and usable values.
"""
# specify a standard list of colums for every row in the final resultset
item = collections.OrderedDict()
cols = 'id organizacia zariadenie typ druh_1 druh_2'.split(' ')
for col in cols:
item[col] = None
# id, organizacia
if re.match('^\d+$', row['ID']):
item['id'] = int(row['ID'])
item['organizacia'] = row.get('ID2', None)
else:
results = re.findall('^(\d+) (.*)$', row['ID'])
if results:
item['id'] = int(results[0][0])
item['organizacia'] = results[0][1]
else:
return None
# zariadenie
item['zariadenie'] = row.get('Zariadenie', None)
# typ
item['typ'] = row.get('Typ', None)
# druh
item['druh_1'] = row.get('Druh', None)
item['druh_2'] = row.get('Druh2', None)
return item
html = scraperwiki.scrape(site_url + start_page)
# get all pdf links
root = lxml.html.fromstring(html)
pdf_urls = root.cssselect("li.pdf > a")
for pdf_url in pdf_urls:
pdf_url_text = site_url + pdf_url.get('href')
print pdf_url_text
pdf_text = scraperwiki.scrape(pdf_url_text)
data = scraperwiki.pdftoxml(pdf_text)
tree = lxml.etree.fromstring(data)
#tree = lxml.etree.parse('data.xml')
missed_rows_global = 0
for p, page in enumerate(tree.xpath('page')):
print "processing page" + page.get('number')
rows = {}
xmlcells = page.xpath('text')
lastrow = 0
missed_rows = 0
for xmlcell in xmlcells:
top = int(xmlcell.get('top'))
for dev in range(diff_vert+1):
if top+dev in rows:
rows[top+dev].append(xmlcell)
break
elif top-dev in rows:
rows[top-dev].append(xmlcell)
break
else:
pass
else:
rows[top] = []
rows[top].append(xmlcell)
pagedata = []
for key in sorted(rows.keys()):
itemvalues = {}
for column in rows[key]:
left = int(column.get('left'))
for dev in range(diff_horiz+1):
if left+dev in cellmap:
itemvalues[cellmap[left+dev]] = column.xpath('string()')
elif left-dev in cellmap:
itemvalues[cellmap[left-dev]] = column.xpath('string()')
# we only want records with an ID
id = itemvalues.get('ID', '')
if re.match('^\d+ \D.*', id) or (re.match('^\d+$', id) and 'ID2' in itemvalues):
itemvalues_processed = process_columns(itemvalues)
if itemvalues_processed is not None:
pagedata.append(itemvalues_processed)
else:
missed_rows += 1
else:
missed_rows += 1
scraperwiki.sqlite.save(unique_keys=['id'],data=pagedata)
print "Missed rows: %s" % missed_rows
missed_rows_global += missed_rows
print "Processed %s pages. Missed %s rows." % (p+1, missed_rows_global)