From e1456b27916167c58f5163175d1d585e5ae37919 Mon Sep 17 00:00:00 2001
From: Ori Hoch <ori@uumpa.com>
Date: Wed, 10 Feb 2016 16:19:55 +0200
Subject: [PATCH] added scraper to get votes data from html (due to bug in
 dataservice)

---
 python/knesset_data/html_scrapers/__init__.py |  1 +
 .../html_scrapers/tests/__init__.py           |  1 +
 .../html_scrapers/tests/test_html_vote.py     |  9 +++++
 python/knesset_data/html_scrapers/votes.py    | 39 +++++++++++++++++++
 4 files changed, 50 insertions(+)
 create mode 100644 python/knesset_data/html_scrapers/__init__.py
 create mode 100644 python/knesset_data/html_scrapers/tests/__init__.py
 create mode 100644 python/knesset_data/html_scrapers/tests/test_html_vote.py
 create mode 100644 python/knesset_data/html_scrapers/votes.py

diff --git a/python/knesset_data/html_scrapers/__init__.py b/python/knesset_data/html_scrapers/__init__.py
new file mode 100644
index 0000000..68831bd
--- /dev/null
+++ b/python/knesset_data/html_scrapers/__init__.py
@@ -0,0 +1 @@
+__author__ = 'orihoch'
diff --git a/python/knesset_data/html_scrapers/tests/__init__.py b/python/knesset_data/html_scrapers/tests/__init__.py
new file mode 100644
index 0000000..68831bd
--- /dev/null
+++ b/python/knesset_data/html_scrapers/tests/__init__.py
@@ -0,0 +1 @@
+__author__ = 'orihoch'
diff --git a/python/knesset_data/html_scrapers/tests/test_html_vote.py b/python/knesset_data/html_scrapers/tests/test_html_vote.py
new file mode 100644
index 0000000..00437e9
--- /dev/null
+++ b/python/knesset_data/html_scrapers/tests/test_html_vote.py
@@ -0,0 +1,9 @@
+import unittest
+from knesset_data.html_scrapers.votes import HtmlVote
+
+
+class TestHtmlVote(unittest.TestCase):
+
+    def test(self):
+        html_vote = HtmlVote.get_from_vote_id(24084)
+        self.assertListEqual(html_vote.member_votes, [('863', 'voted for'), ('879', 'voted for'), ('953', 'voted for'), ('914', 'voted for'), ('918', 'voted for'), ('950', 'voted for'), ('922', 'voted for'), ('924', 'voted for'), ('932', 'voted for'), ('944', 'voted for'), ('230', 'voted for'), ('862', 'voted for'), ('865', 'voted for'), ('881', 'voted for'), ('818', 'voted for'), ('901', 'voted for'), ('216', 'voted for'), ('876', 'voted for'), ('915', 'voted for'), ('854', 'voted for'), ('941', 'voted for'), ('871', 'voted for')])
diff --git a/python/knesset_data/html_scrapers/votes.py b/python/knesset_data/html_scrapers/votes.py
new file mode 100644
index 0000000..5a9ecb6
--- /dev/null
+++ b/python/knesset_data/html_scrapers/votes.py
@@ -0,0 +1,39 @@
+import urllib2
+import re
+
+
+class HtmlVote(object):
+
+    def __init__(self, page):
+        self.page = page
+
+    @property
+    def member_votes(self):
+        """
+         Returns a tuple of (member_id, vote_result_code) describing the vote found in page, where:
+         member_id = link to dataservice member id
+         vote_result_code = "voted for"|"voted against"|"abstain"|"did not vote" - matching the codes in dataservice.votes.VoteMember.vote_result_code
+        """
+        results = []
+        pattern = re.compile("""Vote_Bord""")
+        match = pattern.split(self.page)
+        for i in match:
+            vote_result_code = ""
+            if(re.match("""_R1""", i)):
+                vote_result_code = "voted for"
+            if(re.match("""_R2""", i)):
+                vote_result_code = "voted against"
+            if(re.match("""_R3""", i)):
+                vote_result_code = "abstain"
+            if(re.match("""_R4""", i)):
+                vote_result_code = "did not vote"
+            if(vote_result_code != ""):
+                member_id = re.search("""mk_individual_id_t=(\d+)""",i).group(1)
+                results.append((member_id, vote_result_code))
+        return results
+
+    @classmethod
+    def get_from_vote_id(cls, vote_id):
+        url = 'http://www.knesset.gov.il/vote/heb/Vote_Res_Map.asp?vote_id_t=%s'%vote_id
+        page = urllib2.urlopen(url).read()
+        return cls(page)
\ No newline at end of file