Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 29 additions & 0 deletions pension_crawler/sites/spiders.py
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,7 @@ def __init__(self, crawler, data, *args, **kwargs):
super(SitesSpider, self).__init__(*args, **kwargs)
self.crawler = crawler
self.data = data
self.years = [str(year) for year in range(1960,2019,1)]

# class methods

Expand Down Expand Up @@ -57,6 +58,34 @@ def _process_item(self, url, node):
loader.add_xpath('text', 'text()')
loader.add_value('file_urls', [href])
loader.add_value('timestamp', datetime.now().isoformat())

'''
Code added by Joseph J. Bautista
- Context: Some links were missing in the output.csv file while it was clear that the year
was present in the link. This is the proposed solution for that.
'''
href = href.split('/')[-1].split('.')[0]
i = 0
year_stack = []
for year in self.years:
if year in link:
i = i + 1
year_stack.append(year)
elif "FY{}".format(year[-2:]) in link:
i = i + 1
year_stack.append(year)
elif "FY-{}".format(year[-2:]) in link:
i = i + 1
year_stack.append(year)
elif "fy{}".format(year[-2:]) in link:
i = i + 1
elif "fy-{}".format(year[-2:]) in link:
i = i + 1
if == 1:
loader.add_value('year', year)
elif i > 1:
loader.add_value('year', year_stack[-1])

return loader.load_item()

# class method overrides
Expand Down