From 6a599fef2febaf4322a2950ea7418b4b8df36d3c Mon Sep 17 00:00:00 2001
From: Tim Showers <showerst@gmail.com>
Date: Thu, 15 Dec 2016 10:25:16 -0500
Subject: [PATCH] Updated SC bills to scrape prefiles

---
 openstates/sc/bills.py | 31 ++++++++++++++++++++++++++++++-
 1 file changed, 30 insertions(+), 1 deletion(-)

diff --git a/openstates/sc/bills.py b/openstates/sc/bills.py
index 9cb3083538..eef2671500 100644
--- a/openstates/sc/bills.py
+++ b/openstates/sc/bills.py
@@ -63,9 +63,11 @@ class SCBillScraper(BillScraper):
     urls = {
         'lower' : {
           'daily-bill-index': "http://www.scstatehouse.gov/hintro/hintros.php",
+          'prefile-index': "http://www.scstatehouse.gov/sessphp/prefil17.php",
         },
         'upper' : {
           'daily-bill-index': "http://www.scstatehouse.gov/sintro/sintros.php",
+          'prefile-index': "http://www.scstatehouse.gov/sessphp/prefil17.php",
         }
     }
 
@@ -252,7 +254,7 @@ def scrape_details(self, bill_detail_url, session, chamber, bill_id):
     def scrape(self, chamber, session):
         # start with subjects
         session_code = self.metadata['session_details'][session]['_code']
-        self.scrape_subjects(session_code)
+        #self.scrape_subjects(session_code)
 
         # get bill index
         index_url = self.urls[chamber]['daily-bill-index']
@@ -278,3 +280,30 @@ def scrape(self, chamber, session):
                 if bill_id.startswith(chamber_letter):
                     self.scrape_details(bill_a.get('href'), session, chamber,
                                         bill_id)
+
+        prefile_url = self.urls[chamber]['prefile-index']
+        page = self.get(prefile_url).text
+        doc = lxml.html.fromstring(page)
+        doc.make_links_absolute(prefile_url)
+
+        # visit each day and extract bill ids
+        days = ''
+        if chamber == 'lower':
+            days = doc.xpath('//dd[contains(text(),"House")]/a/@href')
+        else:
+            days = doc.xpath('//dd[contains(text(),"Senate")]/a/@href')
+            
+        for day_url in days:
+            try:
+                data = self.get(day_url).text
+            except scrapelib.HTTPError:
+                continue
+
+            doc = lxml.html.fromstring(data)
+            doc.make_links_absolute(day_url)
+
+            for bill_a in doc.xpath('//p/a[1]'):
+                bill_id = bill_a.text.replace('.', '')
+                if bill_id.startswith(chamber_letter):
+                    self.scrape_details(bill_a.get('href'), session, chamber,
+                                        bill_id)
\ No newline at end of file