From 1c7ba35aebf1305c5db14533d377a0d520f99909 Mon Sep 17 00:00:00 2001
From: Ed Summers <ehs@pobox.com>
Date: Sun, 20 Nov 2011 22:51:20 -0500
Subject: [PATCH] added readme

---
 .gitignore     |  2 ++
 README.md      |  5 +++++
 __init__.py    |  0
 jobs/models.py | 20 +++++++++++++++++++-
 jobs/tests.py  | 10 +++++-----
 5 files changed, 31 insertions(+), 6 deletions(-)
 create mode 100644 README.md
 create mode 100644 __init__.py

diff --git a/.gitignore b/.gitignore
index 449b550..055a648 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,4 @@
 mboxes/*
 *.pyc
+settings.py
+*.db
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..eea5117
--- /dev/null
+++ b/README.md
@@ -0,0 +1,5 @@
+jobs4lib
+--------
+
+A django app that collects job announcements from the code4lib discussion 
+list and puts them on the Web.
diff --git a/__init__.py b/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/jobs/models.py b/jobs/models.py
index 39a7800..9370056 100644
--- a/jobs/models.py
+++ b/jobs/models.py
@@ -9,7 +9,13 @@
 
 from django.db import models
 
+class EmailKeyword(models.Model):
+    created = models.DateTimeField(auto_now_add=True)
+    name = models.CharField(max_length=255)
+    emails = models.ManyToManyField('JobEmail', related_name='keywords')
+
 class JobEmail(models.Model):
+    created = models.DateTimeField(auto_now_add=True)
     from_name = models.CharField(max_length=255)
     from_address = models.CharField(max_length=255)
     from_domain = models.CharField(max_length=255)
@@ -38,7 +44,6 @@ def tags(self):
         words = nltk.word_tokenize(self.body)
         return nltk.pos_tag(words)
 
-
     @classmethod
     def new_from_msg(klass, msg):
         if not is_job(msg):
@@ -63,6 +68,19 @@ def new_from_msg(klass, msg):
             return None
 
         e.save()
+
+        # add keywords
+        for n in e.proper_nouns():
+            n = n.lower()
+            try:
+                kw = EmailKeyword.objects.get(name=n)
+                kw.emails.add(e)
+                kw.save()
+            except EmailKeyword.DoesNotExist:
+                kw = EmailKeyword.objects.create(name=n)
+                kw.emails.add(e)
+                kw.save()
+
         return e
 
 def normalize_name(name):
diff --git a/jobs/tests.py b/jobs/tests.py
index a39e0d4..bca2f88 100644
--- a/jobs/tests.py
+++ b/jobs/tests.py
@@ -5,6 +5,9 @@
 
 class JobsTests(unittest.TestCase):
 
+    def setUp(self):
+        JobEmail.objects.all().delete()
+
     def test_email(self):
         msg = email.message_from_file(open("test-data/job-email"))
         e = JobEmail.new_from_msg(msg)
@@ -14,8 +17,5 @@ def test_email(self):
         self.assertEqual(e.subject, '[CODE4LIB] Job Posting: Head of Web &  Emerging Technologies, University of Miami - revised')
         self.assertTrue('collaborates' in e.body)
         self.assertTrue(e.message_id, '<7933CD19EEFCC94392323A994F6F1EDF01DBB52AE8@MBX03.cgcent.miami.edu>')
-
-    def test_tagging(self):
-        msg = email.message_from_file(open("test-data/job-email"))
-        e = JobEmail.new_from_msg(msg)
-        print e.tags()
+        keywords = [kw.name for kw in e.keywords.all()]
+        self.assertTrue('drupal' in keywords)