various improvements to the crawler (better use of coroutines, test if an article should be updated). tags are now retrieved for the k-means clustering (previously achived with the content of articles)

4 files changed, 34 insertions, 2 deletions

diff --git a/src/web/models/__init__.py b/src/web/models/__init__.py
index 53b692e2..1fc0c3e0 100644
--- a/src/web/models/__init__.py
+++ b/src/web/models/__init__.py
@@ -32,8 +32,9 @@ from .user import User
 from .article import Article
 from .icon import Icon
 from .category import Category
+from .tag import Tag
 
-__all__ = ['Feed', 'Role', 'User', 'Article', 'Icon', 'Category']
+__all__ = ['Feed', 'Role', 'User', 'Article', 'Icon', 'Category', 'Tag']
 
 import os
 
diff --git a/src/web/models/article.py b/src/web/models/article.py
index 23708f6b..5261cb0d 100644
--- a/src/web/models/article.py
+++ b/src/web/models/article.py
@@ -29,6 +29,8 @@ __license__ = "GPLv3"
 from bootstrap import db
 from datetime import datetime
 from sqlalchemy import asc, desc, Index
+from sqlalchemy.ext.associationproxy import association_proxy
+
 from web.models.right_mixin import RightMixin
 
 
@@ -49,6 +51,13 @@ class Article(db.Model, RightMixin):
     feed_id = db.Column(db.Integer(), db.ForeignKey('feed.id'))
     category_id = db.Column(db.Integer(), db.ForeignKey('category.id'))
 
+    # relationships
+    tag_objs = db.relationship('Tag', back_populates='article',
+                            cascade='all,delete-orphan',
+                            lazy=False,
+                            foreign_keys='[Tag.article_id]')
+    tags = association_proxy('tag_objs', 'text')
+
     # index
     idx_article_uid = Index('user_id')
     idx_article_uid_cid = Index('user_id', 'category_id')
diff --git a/src/web/models/tag.py b/src/web/models/tag.py
new file mode 100644
index 00000000..8d7fe4d4
--- /dev/null
+++ b/src/web/models/tag.py
@@ -0,0 +1,22 @@
+#! /usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from sqlalchemy import Column, ForeignKey, Integer, String
+from sqlalchemy.orm import relationship
+
+from bootstrap import db
+
+
+class Tag(db.Model):
+    text = Column(String, primary_key=True, unique=False)
+
+    # foreign keys
+    article_id = Column(Integer, ForeignKey('article.id', ondelete='CASCADE'),
+                        primary_key=True)
+
+    # relationships
+    article = relationship('Article', back_populates='tag_objs',
+                           foreign_keys=[article_id])
+
+    def __init__(self, text):
+        self.text = text
diff --git a/src/web/models/user.py b/src/web/models/user.py
index 37bc78fd..460958e0 100644
--- a/src/web/models/user.py
+++ b/src/web/models/user.py
@@ -64,7 +64,7 @@ class User(db.Model, UserMixin, RightMixin):
     is_admin = db.Column(db.Boolean(), default=False)
     is_api = db.Column(db.Boolean(), default=False)
 
-    # relationship
+    # relationships
     categories = db.relationship('Category', backref='user',
                               cascade='all, delete-orphan',
                             foreign_keys=[Category.user_id])