statement 1.9.7 → 1.9.8
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.
- checksums.yaml +4 -4
- data/lib/statement/scraper.rb +5 -4
- data/lib/statement/version.rb +1 -1
- metadata +1 -1
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 87d85f7b50d23a90cbdaa02909c6361ec6408afe
|
4
|
+
data.tar.gz: 2d30a1e96a897bc211d62c9ed115f8f4eb9495e6
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 73695668ceca19e07c15001a397a6d67c163faa97fe9023ecae61a3a3200ca507d7b9e9afe5a219c6dc4297d933ef3839502ee823f4aa6efd34cf13e758ff222
|
7
|
+
data.tar.gz: ee18e827c2fac4a6b11e13ca2f7c8b6eff5bed5d4b253710906ee8a834f3f4f5fe13fd03bf6a208530279963767e14a6f0de468440d7ede105004c81a0fbb259
|
data/lib/statement/scraper.rb
CHANGED
@@ -284,10 +284,11 @@ module Statement
|
|
284
284
|
results
|
285
285
|
end
|
286
286
|
|
287
|
-
def self.cold_fusion(year=Date.today.year, month=nil)
|
287
|
+
def self.cold_fusion(year=Date.today.year, month=nil, skip_domains=[])
|
288
288
|
results = []
|
289
289
|
year = Date.today.year if not year
|
290
|
-
domains = ['www.ronjohnson.senate.gov','www.risch.senate.gov', 'www.lee.senate.gov', 'www.barrasso.senate.gov']
|
290
|
+
domains = ['www.ronjohnson.senate.gov','www.risch.senate.gov', 'www.lee.senate.gov', 'www.barrasso.senate.gov', 'www.heitkamp.senate.gov']
|
291
|
+
domains = domains - skip_domains if skip_domains
|
291
292
|
domains.each do |domain|
|
292
293
|
if domain == 'www.risch.senate.gov'
|
293
294
|
if not month
|
@@ -310,7 +311,7 @@ module Statement
|
|
310
311
|
end
|
311
312
|
doc = Statement::Scraper.open_html(url)
|
312
313
|
return if doc.nil?
|
313
|
-
if domain == 'www.lee.senate.gov' or domain == 'www.barrasso.senate.gov'
|
314
|
+
if domain == 'www.lee.senate.gov' or domain == 'www.barrasso.senate.gov' or domain == "www.heitkamp.senate.gov"
|
314
315
|
rows = doc.xpath("//tr")[1..-1]
|
315
316
|
else
|
316
317
|
rows = doc.xpath("//tr")[2..-1]
|
@@ -319,7 +320,7 @@ module Statement
|
|
319
320
|
date_text, title = row.children.map{|c| c.text.strip}.reject{|c| c.empty?}
|
320
321
|
next if date_text == 'Date' or date_text.size > 10
|
321
322
|
date = Date.parse(date_text)
|
322
|
-
|
323
|
+
scraped << { :source => url, :url => row.children[3].children.first['href'], :title => title, :date => date, :domain => domain }
|
323
324
|
end
|
324
325
|
end
|
325
326
|
results.flatten
|
data/lib/statement/version.rb
CHANGED