gulesider 0.2.2 → 0.2.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- data/lib/screen_scraper.rb +16 -12
- metadata +3 -3
data/lib/screen_scraper.rb
CHANGED
@@ -30,14 +30,14 @@ class ScreenScraper
|
|
30
30
|
profile_info = vcard.search "div.profile-info"
|
31
31
|
|
32
32
|
profile_info.search("p").each do |p|
|
33
|
-
name = vcard.search("span.given-name").text + " " + vcard.search("span.family-name").text
|
34
|
-
phone = p.search("span.tel > a").text.strip
|
33
|
+
name = wash(vcard.search("span.given-name").text) + " " + wash(vcard.search("span.family-name").text)
|
34
|
+
phone = wash p.search("span.tel > a").text.strip
|
35
35
|
address = p.search("span.adr")
|
36
|
-
street_name = address.search("span.street-address").text
|
37
|
-
postal_code = address.search("span.postal-code").text
|
38
|
-
city = address.search("span.locality").text
|
39
|
-
lat = address.search("span.latitude").text
|
40
|
-
long = address.search("span.longitude").text
|
36
|
+
street_name = wash address.search("span.street-address").text
|
37
|
+
postal_code = wash address.search("span.postal-code").text
|
38
|
+
city = wash address.search("span.locality").text
|
39
|
+
lat = wash address.search("span.latitude").text
|
40
|
+
long = wash address.search("span.longitude").text
|
41
41
|
|
42
42
|
if phone != ""
|
43
43
|
search_result.push SearchResult.new({
|
@@ -56,6 +56,10 @@ class ScreenScraper
|
|
56
56
|
return search_result
|
57
57
|
end
|
58
58
|
|
59
|
+
def wash str
|
60
|
+
str.gsub(/\r|\n|\t/, "")
|
61
|
+
end
|
62
|
+
|
59
63
|
def scraped_company_data
|
60
64
|
search_result = []
|
61
65
|
list = @result.search "#result-list"
|
@@ -64,11 +68,11 @@ class ScreenScraper
|
|
64
68
|
hits.each do |hit|
|
65
69
|
search_result.push SearchResult.new({
|
66
70
|
:company => true,
|
67
|
-
:name => hit.search("div.header").search("a").text,
|
68
|
-
:phone => hit.search("li.tel:first-child").text,
|
69
|
-
:street_name => hit.search("span.street-address").text,
|
70
|
-
:postal_code => hit.search("span.postal-code").text,
|
71
|
-
:city => hit.search("span.locality").text
|
71
|
+
:name => wash(hit.search("div.header").search("a").text),
|
72
|
+
:phone => wash(hit.search("li.tel:first-child").text),
|
73
|
+
:street_name => wash(hit.search("span.street-address").text),
|
74
|
+
:postal_code => wash(hit.search("span.postal-code").text),
|
75
|
+
:city => wash(hit.search("span.locality").text)
|
72
76
|
})
|
73
77
|
end
|
74
78
|
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: gulesider
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.2.
|
4
|
+
version: 0.2.3
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -13,7 +13,7 @@ date: 2012-01-02 00:00:00.000000000Z
|
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: mechanize
|
16
|
-
requirement: &
|
16
|
+
requirement: &70211529548380 !ruby/object:Gem::Requirement
|
17
17
|
none: false
|
18
18
|
requirements:
|
19
19
|
- - ! '>='
|
@@ -21,7 +21,7 @@ dependencies:
|
|
21
21
|
version: '2.1'
|
22
22
|
type: :runtime
|
23
23
|
prerelease: false
|
24
|
-
version_requirements: *
|
24
|
+
version_requirements: *70211529548380
|
25
25
|
description:
|
26
26
|
email: mail@goeran.no
|
27
27
|
executables:
|