click_distributions.py 文件源码

def pickle_category_counts_distribution():
    results =  {}
    db = MySQLDatabase(DATABASE_HOST, DATABASE_USER, DATABASE_PASSWORD, DATABASE_NAME)
    db_worker_view = db.get_work_view()
    cursor = db_worker_view._cursor
    for category in ['lead', 'infobox', 'body', 'left-body', 'navbox']:
        try:
            cursor.execute('select counts from link_features where counts is not null and visual_region=%s;', (category,))
            result = cursor.fetchall()
            results[category] = result
        except MySQLdb.Error, e:
            print e

    try:
        cursor.execute('select counts from clickstream_derived_internal_links;')
        result = cursor.fetchall()
        results['counts'] = result
    except MySQLdb.Error, e:
        print e

    write_pickle(HOME+'output/category_counts_distribution.obj', results)