| import logging | import logging |
| import operator | import operator |
| import ckan.lib.base as base | import ckan.lib.base as base |
| import ckan.model as model | import ckan.model as model |
| from ckanext.ga_report.ga_model import GA_Url, GA_Publisher | from ckanext.ga_report.ga_model import GA_Url, GA_Publisher |
| from ckanext.ga_report.controller import _get_publishers | from ckanext.ga_report.controller import _get_publishers |
| _log = logging.getLogger(__name__) | _log = logging.getLogger(__name__) |
| def popular_datasets(count=10): | def popular_datasets(count=10): |
| import random | import random |
| publisher = None | publisher = None |
| publishers = _get_publishers(30) | publishers = _get_publishers(30) |
| total = len(publishers) | total = len(publishers) |
| while not publisher or not datasets: | while not publisher or not datasets: |
| rand = random.randrange(0, total) | rand = random.randrange(0, total) |
| publisher = publishers[rand][0] | publisher = publishers[rand][0] |
| if not publisher.state == 'active': | if not publisher.state == 'active': |
| publisher = None | publisher = None |
| continue | continue |
| datasets = _datasets_for_publisher(publisher, 10)[:count] | datasets = _datasets_for_publisher(publisher, 10)[:count] |
| ctx = { | ctx = { |
| 'datasets': datasets, | 'datasets': datasets, |
| 'publisher': publisher | 'publisher': publisher |
| } | } |
| return base.render_snippet('ga_report/ga_popular_datasets.html', **ctx) | return base.render_snippet('ga_report/ga_popular_datasets.html', **ctx) |
| def single_popular_dataset(top=20): | def single_popular_dataset(top=20): |
| '''Returns a random dataset from the most popular ones. | |
| :param top: the number of top datasets to select from | |
| ''' | |
| import random | import random |
| datasets = {} | top_datasets = model.Session.query(GA_Url).\ |
| rand = random.randrange(0, top) | filter(GA_Url.url.like('/dataset/%')).\ |
| entry = model.Session.query(GA_Url).\ | order_by('ga_url.pageviews::int desc') |
| filter(GA_Url.url.like('/dataset/%')).\ | num_top_datasets = top_datasets.count() |
| order_by('ga_url.pageviews::int desc')[rand] | |
| if num_top_datasets: | |
| dataset = None | dataset = None |
| while not dataset: | while not dataset: |
| dataset = model.Package.get(entry.url[len('/dataset/'):]) | rand = random.randrange(0, min(top, num_top_datasets)) |
| if dataset and not dataset.state == 'active': | ga_url = top_datasets[rand] |
| dataset = None | dataset = model.Package.get(ga_url.url[len('/dataset/'):]) |
| else: | if dataset and not dataset.state == 'active': |
| publisher = model.Group.get(entry.department_id) | dataset = None |
| else: | |
| ctx = { | dataset = model.Session.query(model.Package)\ |
| .filter_by(state='active').first() | |
| publisher = dataset.get_groups('publisher')[0] | |
| return { | |
| 'dataset': dataset, | 'dataset': dataset, |
| 'publisher': publisher | 'publisher': publisher |
| } | } |
| return base.render_snippet('ga_report/ga_popular_single.html', **ctx) | |
| def single_popular_dataset_html(top=20): | |
| context = single_popular_dataset(top) | |
| return base.render_snippet('ga_report/ga_popular_single.html', **context) | |
| def most_popular_datasets(publisher, count=20): | def most_popular_datasets(publisher, count=20): |
| if not publisher: | if not publisher: |
| _log.error("No valid publisher passed to 'most_popular_datasets'") | _log.error("No valid publisher passed to 'most_popular_datasets'") |
| return "" | return "" |
| results = _datasets_for_publisher(publisher, count) | results = _datasets_for_publisher(publisher, count) |
| ctx = { | ctx = { |
| 'dataset_count': len(datasets), | 'dataset_count': len(results), |
| 'datasets': results, | 'datasets': results, |
| 'publisher': publisher | 'publisher': publisher |
| } | } |
| return base.render_snippet('ga_report/publisher/popular.html', **ctx) | return base.render_snippet('ga_report/publisher/popular.html', **ctx) |
| def _datasets_for_publisher(publisher, count): | def _datasets_for_publisher(publisher, count): |
| datasets = {} | datasets = {} |
| entries = model.Session.query(GA_Url).\ | entries = model.Session.query(GA_Url).\ |
| filter(GA_Url.department_id==publisher.name).\ | filter(GA_Url.department_id==publisher.name).\ |
| filter(GA_Url.url.like('/dataset/%')).\ | filter(GA_Url.url.like('/dataset/%')).\ |
| order_by('ga_url.pageviews::int desc').all() | order_by('ga_url.pageviews::int desc').all() |
| for entry in entries: | for entry in entries: |
| if len(datasets) < count: | if len(datasets) < count: |
| p = model.Package.get(entry.url[len('/dataset/'):]) | p = model.Package.get(entry.url[len('/dataset/'):]) |
| if not p in datasets: | if not p in datasets: |
| datasets[p] = {'views':0, 'visits': 0} | datasets[p] = {'views':0, 'visits': 0} |
| datasets[p]['views'] = datasets[p]['views'] + int(entry.pageviews) | datasets[p]['views'] = datasets[p]['views'] + int(entry.pageviews) |
| datasets[p]['visits'] = datasets[p]['visits'] + int(entry.visitors) | datasets[p]['visits'] = datasets[p]['visits'] + int(entry.visitors) |
| results = [] | results = [] |
| for k, v in datasets.iteritems(): | for k, v in datasets.iteritems(): |
| results.append((k,v['views'],v['visits'])) | results.append((k,v['views'],v['visits'])) |
| return sorted(results, key=operator.itemgetter(1), reverse=True) | return sorted(results, key=operator.itemgetter(1), reverse=True) |