NH: Created committee scraper.

openstates/nh/__init__.py

@@ -1,6 +1,7 @@
 import lxml.html
 from .bills import NHBillScraper
 from .legislators import NHLegislatorScraper
+from .committees import NHCommitteeScraper
 from . import utils
 
 metadata = {
@@ -18,7 +19,7 @@
          'start_year': 2011, 'end_year': 2012},
         {'name': '2013-2014', 'sessions': ['2013', '2014'],
          'start_year': 2013, 'end_year': 2014},
-        {'name': '2015-2016', 'sessions': ['2015','2016'],
+        {'name': '2015-2016', 'sessions': ['2015', '2016'],
          'start_year': 2015, 'end_year': 2016}
     ],
     'session_details': {

openstates/nh/committees.py

@@ -0,0 +1,91 @@
+import re
+
+from billy.scrape.committees import CommitteeScraper, Committee
+
+from .utils import db_cursor
+
+
+class NHCommitteeScraper(CommitteeScraper):
+    jurisdiction = 'nh'
+
+    _chamber_map = {
+        'upper': 'S',
+        'lower': 'H',
+    }
+
+    def _parse_committees(self, chamber):
+        """Queries and returns a dict for legislative committees."""
+        chamber_code = NHCommitteeScraper._chamber_map[chamber]
+
+        query = "SELECT "\
+            "Committees.CommitteeCode AS committee_code, "\
+            "Committees.committeename AS committee_name "\
+            "FROM Committees "\
+            "WHERE Committees.CommitteeCode LIKE '{}%' "\
+            "AND DATALENGTH(Committees.CommitteeEmailAddress) > 0"\
+            .format(chamber_code)
+
+        self.db_cursor.execute(query)
+
+        committees = {}
+        for row in self.db_cursor.fetchall():
+            # There is technically a potential for duplicated committee
+            # records to cause issues here. Haven't decided how to
+            # handle this yet, so we're throwing an exception instead.
+            if row['committee_code'] not in committees.keys():
+                committee = Committee(chamber, row['committee_name'])
+                committee_url = 'http://www.gencourt.state.nh.us/house/'\
+                    'committees/committeedetails.aspx?code={}'\
+                    .format(row['committee_code'])
+                committee.add_source(committee_url)
+                committees[row['committee_code']] = committee
+            else:
+                raise ValueError('{} committee is already present.'.format(
+                    row['committee_name']))
+
+        return committees
+
+    def _parse_committee_members(self, committees):
+        """Queries and parses the members for a given dict of committees."""
+        # Adopting the following method of parsing committee members to
+        # reduce querying. Grabs a list of members for all committees
+        # given instead of querying for each committee individually.
+        query = "SELECT "\
+            "CommitteeMembers.CommitteeCode AS committee_code, "\
+            "CommitteeMembers.comments AS comment, "\
+            "Legislators.LastName AS last_name, "\
+            "Legislators.FirstName AS first_name, "\
+            "Legislators.MiddleName AS middle_name "\
+            "FROM CommitteeMembers "\
+            "LEFT OUTER JOIN Legislators "\
+            "ON CommitteeMembers.EmployeeNumber = Legislators.Employeeno "\
+            "WHERE CommitteeMembers.CommitteeCode IN ('{}') "\
+            "AND CommitteeMembers.ActiveMember = 1 "\
+            "ORDER BY CommitteeCode ASC, SequenceNumber ASC"\
+            .format('\', \''.join(committees))
+
+        self.db_cursor.execute(query)
+
+        for row in self.db_cursor.fetchall():
+            committee = committees[row['committee_code']]
+            member_name = '{} {} {}'.format(row['first_name'],
+                row['middle_name'], row['last_name'])
+            member_name = re.sub(r'[\s]{2,}', ' ', member_name)
+
+            if row['comment'] in ('Chairman', 'V Chairman'):
+                member_role = 'chair'
+            else:
+                member_role = 'member'
+
+            committee.add_member(member_name, member_role)
+
+        return committees
+
+    def scrape(self, chamber, term):
+        self.db_cursor = db_cursor()
+
+        committees = self._parse_committee_members(
+            self._parse_committees(chamber))
+
+        for committee_code, committee in committees.iteritems():
+            self.save_committee(committee)