Notice: Over the next few months, we're reorganizing the App Engine documentation site to make it easier to find content and better align with the rest of Google Cloud products. The same content will be available, but the navigation will now match the rest of the Cloud products. If you have feedback or questions as you navigate the site, click Send Feedback.

Python 2 is no longer supported by the community. We recommend that you migrate Python 2 apps to Python 3.
Stay organized with collections Save and categorize content based on your preferences.

Source code for google.appengine.ext.admin

#!/usr/bin/env python
# Copyright 2007 Google Inc.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# See the License for the specific language governing permissions and
# limitations under the License.

"""Simple datastore view and interactive console, for use in dev_appserver."""

import cgi
import collections
import csv
import cStringIO
import datetime
import decimal
import logging
import math
import mimetypes
import os
import os.path
import pickle
import pprint
import random
import sys
import time
import traceback
import types
import urllib
import urlparse

  from google.appengine.cron import groctimespecification
  from google.appengine.api import croninfo
except ImportError:
  HAVE_CRON = False
  HAVE_CRON = True

from google.appengine.api import apiproxy_stub_map
from google.appengine.api import datastore
from google.appengine.api import datastore_errors
from google.appengine.api import datastore_types
from google.appengine.api import memcache
from google.appengine.api import search
from google.appengine.api import taskqueue
from google.appengine.api import users
from google.appengine.api.taskqueue import taskqueue_service_pb
from google.appengine.api.taskqueue import taskqueue_stub
from google.appengine.datastore import datastore_stats_generator
from google.appengine.ext import db
from google.appengine.ext import webapp
from google.appengine.ext.db import metadata
from google.appengine.ext.webapp import _template
from google.appengine.ext.webapp import util
from google.appengine.runtime import apiproxy_errors

_DEBUG = True

QUEUE_MODE = taskqueue_service_pb.TaskQueueMode

_UsecToSec = taskqueue_stub._UsecToSec
_FormatEta = taskqueue_stub._FormatEta
_EtaDelta = taskqueue_stub._EtaDelta

    'If your app uses memcache to cache entities (e.g. uses NDB), you may see '
    'stale results unless you flush memcache.')


class _AhAdminXsrfToken_(db.Model):
  """Model class used to persist the XSRF token."""


  xsrf_token = db.StringProperty()

[docs]def get_xsrf_token(): """Return the XSRF token. This is read from the datastore. If no token is present in the datastore, we create a random token and insert it. """ entity = _AhAdminXsrfToken_.get_by_key_name(_AhAdminXsrfToken_.XSRF_KEY_NAME) if not entity: randints = ['%08x' % (random.randrange(-2**31, 2**31-1) & (2**32-1)) for i in range(6)] xsrf_token = '_'.join(randints) entity = _AhAdminXsrfToken_(key_name=_AhAdminXsrfToken_.XSRF_KEY_NAME, xsrf_token=xsrf_token) entity.put() return entity.xsrf_token
[docs]def xsrf_required(method): """Decorator to protect post() handlers against XSRF attacks.""" def xsrf_required_decorator(self): expected_token = get_xsrf_token() actual_token = self.request.get('xsrf_token') if actual_token != expected_token: self.response.set_status(403, 'Invalid XSRF token') self.response.out.write('<h1>Invalid XSRF token</h1>\n' + '<p>Please reload the form page</n>\n' + ' '*512) else: method(self) return xsrf_required_decorator
[docs]def ustr(value): """Like str(), but UTF-8-encodes Unicode instead of failing.""" try: return str(value) except UnicodeError: return unicode(value).encode('UTF-8')
[docs]def urepr(value): """Like repr(), but UTF-8-encodes Unicode inside a list.""" if isinstance(value, list): return '[' + ', '.join(map(urepr, value)) + ']' if isinstance(value, unicode): return ('u"' + value.encode('utf-8').replace('\"', '\\"').replace('\\', '\\\\') + '"') return repr(value)
[docs]def TruncateValue(value): """Truncates potentially very long string to a fixed maximum length.""" value = ustr(value) if len(value) > 32: return value[:32] + '...' return value
[docs]class Document(object): """Simple representation of document.""" def __init__(self, doc_id): self.doc_id = doc_id self.fields = {}
[docs]class ImageHandler(webapp.RequestHandler): """Serves a static image. This exists because we don't want to burden the user with specifying a static file handler for the image resources used by the admin tool. """ PATH = '/images/.*'
[docs] def get(self): image_name = os.path.basename(self.request.path) content_type, encoding = mimetypes.guess_type(image_name) if not content_type or not content_type.startswith('image/'): logging.debug('image_name=%r, content_type=%r, encoding=%r', image_name, content_type, encoding) self.error(404) return directory = os.path.dirname(__file__) path = os.path.join(directory, 'templates', 'images', image_name) try: image_stream = open(path, 'rb') except IOError, e: logging.error('Cannot open image %s: %s', image_name, e) self.error(404) return try: image_data = finally: image_stream.close() self.response.headers['Content-Type'] = content_type self.response.out.write(image_data)
[docs]class BaseRequestHandler(webapp.RequestHandler): """Supplies a common template generation function. When you call generate(), we augment the template variables supplied with the current user in the 'user' variable and the current webapp request in the 'request' variable. """
[docs] def generate(self, template_name, template_values={}): base_path = self.base_path() values = { 'application_name': self.request.environ['APPLICATION_ID'], 'sdk_version': self.request.environ.get('SDK_VERSION', 'Unknown'), 'user': users.get_current_user(), 'request': self.request, 'home_path': base_path + DefaultPageHandler.PATH, 'datastore_path': base_path + DatastoreQueryHandler.PATH, 'datastore_indexes': base_path + DatastoreGetIndexesHandler.PATH, 'datastore_edit_path': base_path + DatastoreEditHandler.PATH, 'datastore_batch_edit_path': base_path + DatastoreBatchEditHandler.PATH, 'datastore_stats_path': base_path + DatastoreStatsHandler.PATH, 'interactive_path': base_path + InteractivePageHandler.PATH, 'interactive_execute_path': base_path + InteractiveExecuteHandler.PATH, 'memcache_path': base_path + MemcachePageHandler.PATH, 'queues_path': base_path + QueuesPageHandler.PATH, 'search_path': base_path + SearchIndexesListHandler.PATH, 'search_index_path': base_path + SearchIndexHandler.PATH, 'search_document_path': base_path + SearchDocumentHandler.PATH, 'search_batch_delete_path': base_path + SearchBatchDeleteHandler.PATH, 'tasks_path': base_path + TasksPageHandler.PATH, 'xmpp_path': base_path + XMPPPageHandler.PATH, 'inboundmail_path': base_path + InboundMailPageHandler.PATH, 'backends_path': base_path + BackendsPageHandler.PATH, 'xsrf_token': get_xsrf_token(), } if HAVE_CRON: values['cron_path'] = base_path + CronPageHandler.PATH if 'X-AppEngine-Datastore-Admin-Enabled' in self.request.headers: values['datastore_admin_path'] = base_path + DatastoreAdminHandler.PATH values['interactive_console'] = self.interactive_console_enabled() values.update(template_values) directory = os.path.dirname(__file__) path = os.path.join(directory, os.path.join('templates', template_name)) self.response.out.write(_template.render(path, values, debug=_DEBUG))
[docs] def base_path(self): """Returns the base path of this admin app, which is chosen by the user. The user specifies which paths map to this application in their app.cfg. You can get that base path with this method. Combine with the constant paths specified by the classes to construct URLs. """ path = self.__class__.PATH return self.request.path[:-len(path)]
[docs] def filter_url(self, args): """Filters the current URL to only have the given list of arguments. For example, if your URL is /search?q=foo&num=100&start=10, then self.filter_url(['start', 'num']) => /search?num=100&start=10 self.filter_url(['q']) => /search?q=10 self.filter_url(['random']) => /search? """ queries = [] for arg in args: value = self.request.get(arg) if value: queries.append(arg + '=' + urllib.quote_plus( ustr(self.request.get(arg)))) return self.request.path + '?' + '&'.join(queries)
[docs] def in_production(self): """Detects if app is running in production. Returns a boolean. """ server_software = os.getenv('SERVER_SOFTWARE') if server_software is None: return False return not server_software.startswith('Development')
[docs] def interactive_console_enabled(self): return 'True' == self.request.headers.get( 'X-AppEngine-Interactive-Console-Enabled', 'True')
[docs]class DefaultPageHandler(BaseRequestHandler): """Redirects to the Datastore application by default.""" PATH = '/'
[docs] def get(self): if self.request.path.endswith('/'): base = self.request.path[:-1] else: base = self.request.path self.redirect(base + DatastoreQueryHandler.PATH)
[docs]class DatastoreAdminHandler(BaseRequestHandler): """Loads the Datastore Admin handler in an iframe.""" PATH = '/datastore_admin'
[docs] def get(self): self.generate('datastore_admin_frame.html')
[docs]class InteractivePageHandler(BaseRequestHandler): """Shows our interactive console HTML.""" PATH = '/interactive'
[docs] def get(self): if users.is_current_user_admin(): self.generate('interactive.html') else: logging.warning( 'Non admin user from IP %s attempted to use interactive console', self.request.remote_addr) self.error(404)
[docs]class InteractiveExecuteHandler(BaseRequestHandler): """Executes the Python code submitted in a POST within this context. For obvious reasons, this should only be available to administrators of the applications. """ PATH = InteractivePageHandler.PATH + '/execute'
[docs] @xsrf_required def post(self): if users.is_current_user_admin(): if self.interactive_console_enabled(): save_stdout = sys.stdout results_io = cStringIO.StringIO() try: sys.stdout = results_io code = self.request.get('code') code = code.replace('\r\n', '\n') try: compiled_code = compile(code, '<string>', 'exec') exec(compiled_code, globals()) except Exception, e: traceback.print_exc(file=results_io) finally: sys.stdout = save_stdout results = results_io.getvalue() else: results = """The interactive console has been disabled for security because the dev_appserver is listening on a non-default address. If you would like to re-enable the console, invoke dev_appserver with the --enable_console argument. See for more information.""" self.generate('interactive-output.html', {'output': results}) else: logging.warning( 'Non admin user from IP %s attempted to use interactive console', self.request.remote_addr) self.error(404)
[docs]class CronPageHandler(BaseRequestHandler): """Shows information about configured cron jobs in this application.""" PATH = '/cron'
[docs] def get(self, now=None): """Shows template displaying the configured cron jobs.""" if not now: now = datetime.datetime.utcnow() values = {'request': self.request} cron_info = _ParseCronYaml() values['cronjobs'] = [] values['now'] = str(now) if cron_info and cron_info.cron: for entry in cron_info.cron: job = {} values['cronjobs'].append(job) if entry.description: job['description'] = entry.description else: job['description'] = '(no description)' if entry.timezone: job['timezone'] = entry.timezone job['url'] = entry.url job['schedule'] = entry.schedule schedule = groctimespecification.GrocTimeSpecification(entry.schedule) matches = schedule.GetMatches(now, 3) job['times'] = [] for match in matches: job['times'].append({'runtime': match.strftime('%Y-%m-%d %H:%M:%SZ'), 'difference': str(match - now)}) self.generate('cron.html', values)
[docs]class XMPPPageHandler(BaseRequestHandler): """Tests XMPP requests.""" PATH = '/xmpp'
[docs] def get(self): """Shows template displaying the XMPP.""" xmpp_configured = True values = { 'xmpp_configured': xmpp_configured, 'request': self.request } self.generate('xmpp.html', values)
[docs]class InboundMailPageHandler(BaseRequestHandler): """Tests Mail requests.""" PATH = '/inboundmail'
[docs] def get(self): """Shows template displaying the Inbound Mail form.""" inboundmail_configured = True values = { 'inboundmail_configured': inboundmail_configured, 'request': self.request } self.generate('inboundmail.html', values)
[docs]class TaskQueueHelper(object): """Taskqueue rpc wrapper.""" def _make_sync_call(self, rpc_name, request): """Make a synchronous taskqueue api call. Args: rpc_name: The name of the rpc to call. request: The protocol buffer to be used as the request. Returns: The rpc response. This is an instance of the correct response protocol buffer for the request 'rpc_name'. """ response = getattr(taskqueue_service_pb, 'TaskQueue%sResponse' % rpc_name)() apiproxy_stub_map.MakeSyncCall('taskqueue', rpc_name, request, response) return response
[docs] def get_queues(self, now): """Get a list of queue in the application. Args: now: The current time. A datetime.datetime object with a utc timezone. Returns: A list of queue dicts corresponding to the tasks for this application. """ request = taskqueue_service_pb.TaskQueueFetchQueuesRequest() request.set_max_rows(1000) response = self._make_sync_call('FetchQueues', request) queue_stats_request = taskqueue_service_pb.TaskQueueFetchQueueStatsRequest() queue_stats_request.set_max_num_tasks(0) queues = [] for queue_proto in response.queue_list(): queue = {'name': queue_proto.queue_name(), 'mode': queue_proto.mode(), 'rate': queue_proto.user_specified_rate(), 'bucket_size': queue_proto.bucket_capacity()} queues.append(queue) queue_stats_request.queue_name_list().append(queue_proto.queue_name()) queue_stats_response = self._make_sync_call( 'FetchQueueStats', queue_stats_request) for queue, queue_stats in zip(queues, queue_stats_response.queuestats_list()): queue['tasks_in_queue'] = queue_stats.num_tasks() if queue_stats.oldest_eta_usec() != -1: queue['oldest_task'] = _FormatEta(queue_stats.oldest_eta_usec()) queue['eta_delta'] = _EtaDelta(queue_stats.oldest_eta_usec(), now) return queues
[docs] def get_number_tasks_in_queue(self, queue_name): """Returns the number of tasks in the named queue. Args: queue_name: The name of the queue. Returns: The number of tasks in the queue. """ queue_stats_request = taskqueue_service_pb.TaskQueueFetchQueueStatsRequest() queue_stats_request.set_max_num_tasks(0) queue_stats_request.add_queue_name(queue_name) queue_stats_response = self._make_sync_call( 'FetchQueueStats', queue_stats_request) assert queue_stats_response.queuestats_size() == 1 return queue_stats_response.queuestats(0).num_tasks()
[docs] def get_tasks(self, now, queue_name, start_eta_usec, start_task_name, num_tasks): """Fetch the specified tasks from taskqueue. Note: This only searchs by eta. Args: now: The current time. This is used to calculate the EtaFromNow. Must be a datetime.datetime in the utc timezone. queue_name: The queue to search for tasks. start_eta_usec: The earliest eta to return. start_task_name: For tasks with the same eta_usec, this is used as a tie breaker. num_tasks: The maximum number of tasks to return. Returns: A list of task dicts (as returned by taskqueue_stub.QueryTasksResponseToDict). """ request = taskqueue_service_pb.TaskQueueQueryTasksRequest() request.set_queue_name(queue_name) request.set_start_task_name(start_task_name) request.set_start_eta_usec(start_eta_usec) request.set_max_rows(num_tasks) response = self._make_sync_call('QueryTasks', request) tasks = [] for task in response.task_list(): tasks.append(taskqueue_stub.QueryTasksResponseToDict( queue_name, task, now)) return tasks
[docs] def delete_task(self, queue_name, task_name): """Delete the named task. Args: queue_name: The name of the queue. task_name: The name of the task. """ request = taskqueue_service_pb.TaskQueueDeleteRequest() request.set_queue_name(queue_name) request.task_name_list().append(task_name) self._make_sync_call('Delete', request)
[docs] def purge_queue(self, queue_name): """Purge the named queue. Args: queue_name: the name of the queue. """ request = taskqueue_service_pb.TaskQueuePurgeQueueRequest() request.set_queue_name(queue_name) self._make_sync_call('PurgeQueue', request)
[docs]class QueueBatch(object): """Collection of push queues or pull queues.""" def __init__(self, title, run_manually, rate_limited, contents): self.title = title self.run_manually = run_manually self.rate_limited = rate_limited self.contents = contents def __eq__(self, other): if type(self) is not type(other): return NotImplemented return (self.title == other.title and self.run_manually == other.run_manually and self.rate_limited == other.rate_limited and self.contents == other.contents) def __iter__(self): return self.contents.__iter__()
[docs]class QueuesPageHandler(BaseRequestHandler): """Shows information about configured (and default) task queues.""" PATH = '/queues' def __init__(self, *args, **kwargs): super(QueuesPageHandler, self).__init__(*args, **kwargs) self.helper = TaskQueueHelper()
[docs] def get(self): """Shows template displaying the configured task queues.""" def is_push_queue(queue): return queue['mode'] == QUEUE_MODE.PUSH def is_pull_queue(queue): return queue['mode'] == QUEUE_MODE.PULL now = datetime.datetime.utcnow() values = {} try: queues = self.helper.get_queues(now) push_queues = QueueBatch('Push Queues', True, True, filter(is_push_queue, queues)) pull_queues = QueueBatch('Pull Queues', False, False, filter(is_pull_queue, queues)) values['queueBatches'] = [push_queues, pull_queues] except apiproxy_errors.ApplicationError: logging.exception('Could not fetch list of queues.') self.generate('queues.html', values)
[docs] @xsrf_required def post(self): """Handle modifying actions and/or redirect to GET page.""" queue_name = self.request.get('queue') if self.request.get('action:purgequeue'): self.helper.purge_queue(queue_name) self.redirect(self.request.path_url)
[docs]class TasksPageHandler(BaseRequestHandler): """Shows information about a queue's tasks.""" PATH = '/tasks' PAGE_SIZE = 20 MAX_TASKS_TO_FETCH = 1000 MIN_TASKS_TO_FETCH = 200 def __init__(self, *args, **kwargs): super(TasksPageHandler, self).__init__(*args, **kwargs) self.helper = TaskQueueHelper() self.prev_page = None self.next_page = None self.this_page = None
[docs] def parse_arguments(self): """Parse the arguments passed into the request and store them on self.""" self.queue_name = self.request.get('queue') self.start_name = self.request.get('start_name', '') self.start_eta = int(self.request.get('start_eta', '0')) self.per_page = int(self.request.get('per_page', self.PAGE_SIZE)) self.page_no = int(self.request.get('page_no', '1')) assert self.per_page > 0
[docs] def redirect_to_tasks(self, keep_offset=True): """Perform a redirect to the tasks page. Args: keep_offset: If true, will keep the 'start_eta', 'start_name' and 'page_no' fields. """ params = {'queue': self.queue_name, 'per_page': self.per_page} if keep_offset: params['start_name'] = self.start_name params['start_eta'] = self.start_eta params['page_no'] = self.page_no self.redirect('%s?%s' % (self.request.path, urllib.urlencode(params)))
def _generate_page_params(self, page_dict): """Generate the params for a page link.""" params = [ ('queue', self.queue_name), ('start_eta', page_dict['start_eta']), ('start_name', page_dict['start_name']), ('per_page', self.per_page), ('page_no', page_dict['number']), ] return urllib.urlencode(params)
[docs] def generate_page_dicts(self, start_tasks, end_tasks): """Generate the page dicts from a list of tasks. Args: tasks: A list of task dicts, sorted by eta. Returns: A list of page dicts containing the following keys: 'start_name', 'start_eta', 'number', 'has_gap'. """ page_map = {} for i, task in enumerate(start_tasks[::self.per_page]): page_no = i + 1 page_map[page_no] = { 'start_name': task['name'], 'start_eta': task['eta_usec'], 'number': page_no} if page_map and (page_no < self.page_no - 1): page_map[page_no]['has_gap'] = True for i, task in enumerate(end_tasks[::self.per_page]): page_no = self.page_no + i page_map[page_no] = { 'start_name': task['name'], 'start_eta': task['eta_usec'], 'number': self.page_no + i} page_map[1] = {'start_name': '', 'start_eta': 0, 'number': 1} pages = sorted(sorted(page_map.values()), key=lambda page: page['number']) for page in pages: page['url'] = self._generate_page_params(page) self.this_page = page_map[self.page_no] if self.page_no - 1 in page_map: self.prev_page = page_map[self.page_no - 1] if self.page_no + 1 in page_map: self.next_page = page_map[self.page_no + 1] return pages
[docs] def get(self): """Shows template displaying the queue's tasks.""" self.parse_arguments() now = datetime.datetime.utcnow() tasks_to_fetch = min(self.MAX_TASKS_TO_FETCH, max(self.MIN_TASKS_TO_FETCH, self.per_page * 10)) try: tasks = self.helper.get_tasks(now, self.queue_name, self.start_eta, self.start_name, tasks_to_fetch) except apiproxy_errors.ApplicationError: logging.exception('Could not fetch list of tasks.') tasks = [] if self.start_eta or self.start_name: if not tasks: self.redirect_to_tasks(keep_offset=False) return first_tasks = self.helper.get_tasks(now, self.queue_name, 0, '', tasks_to_fetch) else: first_tasks = [] pages = self.generate_page_dicts(first_tasks, tasks) if len(tasks) == tasks_to_fetch: pages[-1]['has_gap'] = True tasks = tasks[:self.per_page] def is_this_push_queue(queue): return (queue['name'] == self.queue_name and queue['mode'] == QUEUE_MODE.PUSH) values = { 'queue': self.queue_name, 'per_page': self.per_page, 'tasks': tasks, 'prev_page': self.prev_page, 'next_page': self.next_page, 'this_page': self.this_page, 'pages': pages, 'page_no': self.page_no, } if any(filter(is_this_push_queue, self.helper.get_queues(now))): values['is_push_queue'] = 'true' self.generate('tasks.html', values)
[docs] @xsrf_required def post(self): self.parse_arguments() self.task_name = self.request.get('task') if self.request.get('action:deletetask'): self.helper.delete_task(self.queue_name, self.task_name) self.redirect_to_tasks(keep_offset=True)
[docs]class BackendsPageHandler(BaseRequestHandler): """Shows information about an app's backends.""" PATH = '/backends' def __init__(self, *args, **kwargs): super(BackendsPageHandler, self).__init__(*args, **kwargs) self.stub = apiproxy_stub_map.apiproxy.GetStub('system')
[docs] def get(self): """Shows template displaying the app's backends or a single backend.""" backend_name = self.request.get('backendName') if backend_name: return self.render_backend_page(backend_name) else: return self.render_backends_page()
[docs] def render_backends_page(self): """Shows template displaying all the app's backends.""" from google.appengine.api import backends if hasattr(self.stub, 'get_backend_info'): backend_info = self.stub.get_backend_info() or [] else: backend_info = [] backend_list = [] for backend in backend_info: backend_list.append({ 'name':, 'instances': backend.instances, 'instanceclass': backend.get_class() or 'B2', 'address': backends.get_hostname(, 'state': 'running', 'options': backend.options, }) values = { 'request': self.request, 'backends': backend_list, 'backend_path': self.base_path() + self.PATH, } self.generate('backends.html', values)
[docs] def get_backend_entry(self, backend_name): """Get the BackendEntry for a single backend.""" if not hasattr(self.stub, 'get_backend_info'): return None backend_entries = self.stub.get_backend_info() or [] for backend in backend_entries: if == backend_name: return backend return None
[docs] def render_backend_page(self, backend_name): """Shows template displaying a single backend.""" from google.appengine.api import backends backend = self.get_backend_entry(backend_name) instances = [] if backend: for i in range(backend.instances): instances.append({ 'id': i, 'address': backends.get_hostname(backend_name, i), 'state': 'running', }) values = { 'request': self.request, 'backend_name': backend_name, 'backend_path': self.base_path() + self.PATH, 'instances': instances, } self.generate('backend.html', values)
[docs] @xsrf_required def post(self): if self.request.get('action:startbackend'): self.stub.start_backend(self.request.get('backend')) if self.request.get('action:stopbackend'): self.stub.stop_backend(self.request.get('backend')) self.redirect(self.request.path_url) return
[docs]class MemcachePageHandler(BaseRequestHandler): """Shows stats about memcache and query form to get values.""" PATH = '/memcache' TYPES = ((str, str, 'String'), (unicode, unicode, 'Unicode String'), (bool, lambda value: MemcachePageHandler._ToBool(value), 'Boolean'), (int, int, 'Integer'), (long, long, 'Long Integer'), (float, float, 'Float')) DEFAULT_TYPESTR_FOR_NEW = 'String' @staticmethod def _ToBool(string_value): """Convert string to boolean value. Args: string_value: A string. Returns: Boolean. True if string_value is "true", False if string_value is "false". This is case-insensitive. Raises: ValueError: string_value not "true" or "false". """ string_value_low = string_value.lower() if string_value_low not in ('false', 'true'): raise ValueError('invalid literal for boolean: %s' % string_value) return string_value_low == 'true' def _GetValueAndType(self, key): """Fetch value from memcache and detect its type. Args: key: String Returns: (value, type), value is a Python object or None if the key was not set in the cache, type is a string describing the type of the value. """ try: value = memcache.get(key) except (pickle.UnpicklingError, AttributeError, EOFError, ImportError, IndexError), e: msg = 'Failed to retrieve value from cache: %s' % e return msg, 'error' if value is None: return None, self.DEFAULT_TYPESTR_FOR_NEW for typeobj, _, typestr in self.TYPES: if isinstance(value, typeobj): break else: typestr = 'pickled' value = pprint.pformat(value, indent=2) return value, typestr def _SetValue(self, key, type_, value): """Convert a string value and store the result in memcache. Args: key: String type_: String, describing what type the value should have in the cache. value: String, will be converted according to type_. Returns: Result of memcache.set(key, converted_value). True if value was set. Raises: ValueError: Value can't be converted according to type_. """ for _, converter, typestr in self.TYPES: if typestr == type_: value = converter(value) break else: raise ValueError('Type %s not supported.' % type_) return memcache.set(key, value)
[docs] def get(self): """Show template and prepare stats and/or key+value to display/edit.""" values = {'request': self.request, 'message': self.request.get('message')} edit = self.request.get('edit') key = self.request.get('key') if edit: key = edit values['show_stats'] = False values['show_value'] = False values['show_valueform'] = True values['types'] = [typestr for _, _, typestr in self.TYPES] elif key: values['show_stats'] = True values['show_value'] = True values['show_valueform'] = False else: values['show_stats'] = True values['show_valueform'] = False values['show_value'] = False if key: values['key'] = key values['value'], values['type'] = self._GetValueAndType(key) values['key_exists'] = values['value'] is not None if values['type'] in ('pickled', 'error'): values['writable'] = False else: values['writable'] = True if values['show_stats']: memcache_stats = memcache.get_stats() if not memcache_stats: memcache_stats = {'hits': 0, 'misses': 0, 'byte_hits': 0, 'items': 0, 'bytes': 0, 'oldest_item_age': 0} values['stats'] = memcache_stats try: hitratio = memcache_stats['hits'] * 100 / (memcache_stats['hits'] + memcache_stats['misses']) except ZeroDivisionError: hitratio = 0 values['hitratio'] = hitratio delta_t = datetime.timedelta(seconds=memcache_stats['oldest_item_age']) values['oldest_item_age'] = - delta_t self.generate('memcache.html', values)
def _urlencode(self, query): """Encode a dictionary into a URL query string. In contrast to urllib this encodes unicode characters as UTF8. Args: query: Dictionary of key/value pairs. Returns: String. """ return '&'.join('%s=%s' % (urllib.quote_plus(k.encode('utf8')), urllib.quote_plus(v.encode('utf8'))) for k, v in query.iteritems())
[docs] @xsrf_required def post(self): """Handle modifying actions and/or redirect to GET page.""" next_param = {} if self.request.get('action:flush'): if memcache.flush_all(): next_param['message'] = 'Cache flushed, all keys dropped.' else: next_param['message'] = 'Flushing the cache failed. Please try again.' elif self.request.get('action:display'): next_param['key'] = self.request.get('key') elif self.request.get('action:edit'): next_param['edit'] = self.request.get('key') elif self.request.get('action:delete'): key = self.request.get('key') result = memcache.delete(key) if result == memcache.DELETE_NETWORK_FAILURE: next_param['message'] = ('ERROR: Network failure, key "%s" not deleted.' % key) elif result == memcache.DELETE_ITEM_MISSING: next_param['message'] = 'Key "%s" not in cache.' % key elif result == memcache.DELETE_SUCCESSFUL: next_param['message'] = 'Key "%s" deleted.' % key else: next_param['message'] = ('Unknown return value. Key "%s" might still ' 'exist.' % key) elif self.request.get('action:save'): key = self.request.get('key') value = self.request.get('value') type_ = self.request.get('type') next_param['key'] = key try: if self._SetValue(key, type_, value): next_param['message'] = 'Key "%s" saved.' % key else: next_param['message'] = 'ERROR: Failed to save key "%s".' % key except ValueError, e: next_param['message'] = 'ERROR: Unable to encode value: %s' % e elif self.request.get('action:cancel'): next_param['key'] = self.request.get('key') else: next_param['message'] = 'Unknown action.' next = self.request.path_url if next_param: next = '%s?%s' % (next, self._urlencode(next_param)) self.redirect(next)
[docs]class DatastoreGetIndexesHandler(BaseRequestHandler): """Our main request handler that displays indexes""" PATH = '/datastore_indexes'
[docs] def get(self): indexes = collections.defaultdict(list) for index, state in datastore.GetIndexes(): properties = [] for property_name, sort_direction in index.Properties(): properties.append({ 'name': property_name, 'sort_symbol': ('&#x25b2;', '&#x25bc;')[sort_direction - 1], 'sort_direction': ('ASCENDING', 'DESCENDING')[sort_direction - 1] }) kind = str(index.Kind()) kind_indexes = indexes[kind] kind_indexes.append({ 'id': str(index.Id()), 'status': ('BUILDING', 'SERVING', 'DELETING', 'ERROR')[state], 'has_ancestor': bool(index.HasAncestor()), 'properties': properties }) self.generate('datastore_indexes.html', {'request': self.request, 'indexes': sorted(indexes.items())})
[docs]class DatastoreRequestHandler(BaseRequestHandler): """The base request handler for our datastore admin pages. We provide utility functions for querying the datastore and inferring the types of entity properties. """
[docs] def start(self): """Returns the santized "start" argument from the URL.""" return self.request.get_range('start', min_value=0, default=0)
[docs] def num(self): """Returns the sanitized "num" argument from the URL.""" return self.request.get_range('num', min_value=1, max_value=100, default=10)
[docs] def execute_query(self, start=0, num=0, no_order=False): """Parses the URL arguments and executes the query. Args: start: How many entities from the beginning of the result list should be skipped from the query. num: How many entities should be returned, if 0 (default) then a reasonable default will be chosen. Returns: A tuple (list of entities, total entity count). If inappropriate URL arguments are given, we return an empty set of results and 0 for the entity count. """ kind = self.request.get('kind') namespace = self.request.get('namespace') if not namespace: namespace = None if not kind: return ([], 0) query = datastore.Query(kind, _namespace=namespace) order = self.request.get('order') order_type = self.request.get('order_type') if order and order_type: order_type = DataType.get_by_name(order_type).python_type() if order.startswith('-'): direction = datastore.Query.DESCENDING order = order[1:] else: direction = datastore.Query.ASCENDING try: query.Order((order, order_type, direction)) except datastore_errors.BadArgumentError: pass if not start: start = self.start() if not num: num = self.num() total = query.Count() entities = query.Get(start + num)[start:] return (entities, total)
[docs] def get_key_values(self, entities): """Returns the union of key names used by the given list of entities. We return the union as a dictionary mapping the key names to a sample value from one of the entities for the key name. """ key_dict = {} for entity in entities: for key, value in entity.iteritems(): if key_dict.has_key(key): key_dict[key].append(value) else: key_dict[key] = [value] return key_dict
[docs] def redirect_with_message(self, message): """Redirect to the 'next' url with message added as the msg parameter.""" quoted_message = urllib.quote_plus(message) redirect_url = self.request.get('next') if '?' in redirect_url: redirect_url += '&msg=%s' % quoted_message else: redirect_url += '?msg=%s' % quoted_message self.redirect(redirect_url)
[docs]class DatastoreQueryHandler(DatastoreRequestHandler): """Our main request handler that executes queries and lists entities. We use execute_query() in our base request handler to parse URL arguments and execute the datastore query. """ PATH = '/datastore' _ONE_MILLION = decimal.Decimal(1000000) _DOLLARS_PER_WRITE = 1/_ONE_MILLION _PENNIES_PER_WRITE = _DOLLARS_PER_WRITE/100 def _writes_to_pennies(self, writes): return self._PENNIES_PER_WRITE * writes def _calculate_writes_for_built_in_indices(self, entity): writes = 0 for prop_name in entity.keys(): if not prop_name in entity.unindexed_properties(): prop_vals = entity[prop_name] if isinstance(prop_vals, (list)): num_prop_vals = len(prop_vals) else: num_prop_vals = 1 writes += 2 * num_prop_vals return writes def _calculate_writes_for_composite_index(self, entity, index): composite_index_value_count = 1 for prop_name, _ in index.Properties(): if not prop_name in entity.keys() or ( prop_name in entity.unindexed_properties()): return 0 prop_vals = entity[prop_name] if isinstance(prop_vals, (list)): composite_index_value_count = ( composite_index_value_count * len(prop_vals)) ancestor_count = 1 if index.HasAncestor(): key = entity.key().parent() while key != None: ancestor_count = ancestor_count + 1 key = key.parent() return composite_index_value_count * ancestor_count def _get_write_ops(self, entity): writes = 2 + self._calculate_writes_for_built_in_indices(entity) for index, _ in datastore.GetIndexes(): if index.Kind() != entity.kind(): continue writes = writes + self._calculate_writes_for_composite_index( entity, index) return writes def _get_creation_cost_analysis(self, entity): write_ops = self._get_write_ops(entity) return (write_ops, self._writes_to_pennies(write_ops))
[docs] def get_kinds(self, namespace): """Get sorted list of kind names the datastore knows about. This should only be called in the development environment as metadata queries are expensive and no caching is done. Args: namespace: The namespace to fetch the schema for e.g. ''. It is an error to pass in None. Returns: A sorted list of kinds e.g. ['Book', 'Guest', Post'] (encoded in utf-8). """ assert namespace is not None q = metadata.Kind.all(namespace=namespace) return [x.kind_name.encode('utf-8') for x in]
[docs] def get(self): """Formats the results from execute_query() for datastore.html. The only complex part of that process is calculating the pager variables to generate the Gooooogle pager at the bottom of the page. """ result_set, total = self.execute_query() key_values = self.get_key_values(result_set) keys = key_values.keys() keys.sort() headers = [] for key in keys[:DEFAULT_MAX_DATASTORE_VIEWER_COLUMNS]: sample_value = key_values[key][0] headers.append({ 'name': ustr(key), 'type': DataType.get(sample_value).name(), }) entities = [] edit_path = self.base_path() + DatastoreEditHandler.PATH for entity in result_set: write_ops = self._get_write_ops(entity) attributes = [] for key in keys[:DEFAULT_MAX_DATASTORE_VIEWER_COLUMNS]: if entity.has_key(key): raw_value = entity[key] data_type = DataType.get(raw_value) value = data_type.format(raw_value) short_value = data_type.short_format(raw_value) additional_html = data_type.additional_short_value_html(raw_value) else: value = '' short_value = '' additional_html = '' attributes.append({ 'name': ustr(key), 'value': ustr(value), 'short_value': short_value, 'additional_html': ustr(additional_html), }) entities.append({ 'key': ustr(entity.key()), 'key_name': ustr(entity.key().name()), 'key_id': entity.key().id(), 'write_ops' : write_ops, 'shortened_key': str(entity.key())[:8] + '...', 'attributes': attributes, 'edit_uri': edit_path + '?key=' + str(entity.key()) + '&kind=' + urllib.quote(ustr(self.request.get('kind'))) + '&next=' + urllib.quote(ustr(self.filter_url( ['kind', 'order', 'order_type', 'namespace', 'num']))), }) start = self.start() num = self.num() max_pager_links = 8 current_page = start / num num_pages = int(math.ceil(total * 1.0 / num)) page_start = max(int(math.floor(current_page - max_pager_links / 2)), 0) page_end = min(page_start + max_pager_links, num_pages) pages = [] for page in range(page_start + 1, page_end + 1): pages.append({ 'number': page, 'start': (page - 1) * num, }) current_page += 1 in_production = self.in_production() if in_production: kinds = None else: kinds = self.get_kinds(self.request.get('namespace')) values = { 'request': self.request, 'in_production': in_production, 'kinds': kinds, 'kind': ustr(self.request.get('kind')), 'order': self.request.get('order'), 'headers': headers, 'entities': entities, 'message': self.request.get('msg'), 'pages': pages, 'current_page': current_page, 'namespace': self.request.get('namespace'), 'show_namespace': self.request.get('namespace', None) is not None, 'num': num, 'next_start': -1, 'prev_start': -1, 'start': start, 'total': total, 'start_base_url': self.filter_url(['kind', 'order', 'order_type', 'namespace', 'num']), 'order_base_url': self.filter_url(['kind', 'namespace', 'num']), 'property_overflow': len(keys) > DEFAULT_MAX_DATASTORE_VIEWER_COLUMNS, } if current_page > 1: values['prev_start'] = int((current_page - 2) * num) if current_page < num_pages: values['next_start'] = int(current_page * num) self.generate('datastore.html', values)
[docs]class DatastoreBatchEditHandler(DatastoreRequestHandler): """Request handler for a batch operation on entities. Supports deleting multiple entities by key, then redirecting to another url. """ PATH = DatastoreQueryHandler.PATH + '/batchedit'
[docs] @xsrf_required def post(self): """Handle POST.""" if self.request.get('flush_memcache'): if memcache.flush_all(): message = 'Cache flushed, all keys dropped.' else: message = 'Flushing the cache failed. Please try again.' self.redirect_with_message(message) return kind = self.request.get('kind') keys = [] index = 0 num_keys = int(self.request.get('numkeys')) for i in xrange(1, num_keys+1): key = self.request.get('key%d' % i) if key: keys.append(key) if self.request.get('action') == 'Delete': num_deleted = 0 for key in keys: datastore.Delete(datastore.Key(key)) num_deleted = num_deleted + 1 message = '%d entit%s deleted. %s' % ( num_deleted, ('ies', 'y')[num_deleted == 1], _DATASTORE_CACHING_WARNING) self.redirect_with_message(message) return self.error(404)
[docs]class DatastoreEditHandler(DatastoreRequestHandler): """Request handler for the entity create/edit form. We determine how to generate a form to edit an entity by doing a query on the entity kind and looking at the set of keys and their types in the result set. We use the DataType subclasses for those introspected types to generate the form and parse the form results. """ PATH = DatastoreQueryHandler.PATH + '/edit'
[docs] def get(self): entity_key = self.request.get('key') if entity_key: key_instance = datastore.Key(entity_key) entity_key_name = if entity_key_name: entity_key_name = ustr(entity_key_name) entity_key_id = namespace = key_instance.namespace() parent_key = key_instance.parent() kind = key_instance.kind() entity = datastore.Get(key_instance) sample_entities = [entity] else: kind = self.request.get('kind') sample_entities = self.execute_query()[0] if len(sample_entities) < 1: next_uri = self.request.get('next') next_uri += '&msg=%s' % urllib.quote_plus( "The kind %s doesn't exist in the %s namespace" % ( kind, self.request.get('namespace', '<Empty>'))) kind_param = 'kind=%s' % kind if not kind_param in next_uri: if '?' in next_uri: next_uri += '&' + kind_param else: next_uri += '?' + kind_param self.redirect(next_uri) return if not entity_key: key_instance = None entity_key_name = None entity_key_id = None namespace = self.request.get('namespace') parent_key = None entity = None if parent_key: parent_kind = parent_key.kind() parent_key_string = PseudoBreadcrumbs(parent_key) else: parent_kind = None parent_key_string = None fields = [] key_values = self.get_key_values(sample_entities) for key, sample_values in sorted(key_values.iteritems()): if entity and entity.has_key(key): data_type = DataType.get(entity[key]) else: data_type = DataType.get(sample_values[0]) name = + "|" + key if entity and entity.has_key(key): value = entity[key] else: value = None field = data_type.input_field(name, value, sample_values) fields.append((ustr(key),, field)) self.generate('datastore_edit.html', { 'kind': ustr(kind), 'key': entity_key, 'key_name': entity_key_name, 'key_id': entity_key_id, 'fields': fields, 'focus': self.request.get('focus'), 'namespace': namespace, 'next': self.request.get('next'), 'parent_key': parent_key, 'parent_kind': parent_kind, 'parent_key_string': parent_key_string, })
[docs] @xsrf_required def post(self): kind = self.request.get('kind') entity_key = self.request.get('key') if entity_key: if self.request.get('action') == 'Delete': datastore.Delete(datastore.Key(entity_key)) self.redirect_with_message( 'Entity deleted. %s' % _DATASTORE_CACHING_WARNING) return entity = datastore.Get(datastore.Key(entity_key)) else: namespace = self.request.get('namespace') if not namespace: namespace = None entity = datastore.Entity(kind, _namespace=namespace) args = self.request.arguments() for arg in args: bar = arg.find('|') if bar > 0: data_type_name = arg[:bar] field_name = arg[bar + 1:] form_value = self.request.get(arg) data_type = DataType.get_by_name(data_type_name) if entity and entity.has_key(field_name): old_formatted_value = data_type.format(entity[field_name]) if old_formatted_value == ustr(form_value): continue if len(form_value) > 0: value = data_type.parse(form_value) entity[field_name] = value elif entity.has_key(field_name): del entity[field_name] datastore.Put(entity) if entity_key: self.redirect_with_message( 'Entity updated. %s' % _DATASTORE_CACHING_WARNING) else: self.redirect(self.request.get('next'))
[docs]class DatastoreStatsHandler(BaseRequestHandler): """Allows computation of datastore stats.""" PATH = '/datastore_stats'
[docs] def get(self): """Shows Datastore Stats generator button.""" values = { 'request': self.request, 'app_id': self.request.get('app_id', None), 'status': self.request.get('status', None), 'msg': self.request.get('msg', None)} self.generate('datastore_stats.html', values)
[docs] @xsrf_required def post(self): """Handle actions and redirect to GET page.""" app_id = self.request.get('app_id', None) if self.request.get('action:compute_stats'): status = 'OK' msg = self.generate_stats(_app=app_id) else: status = 'FAIL' msg = 'No processing requested' uri = self.request.path_url self.redirect('%s?%s' % (uri, urllib.urlencode( [('msg', msg), ('status', status)])))
[docs] def generate_stats(self, _app=None): """Generate datastore stats.""" processor = datastore_stats_generator.DatastoreStatsProcessor(_app) return processor.Run().Report()
[docs]class SearchIndexesListHandler(BaseRequestHandler): """FTS main page with list on indexes.""" PATH = '/search'
[docs] def get(self): """Displays list of FTS indexes.""" start = self.request.get_range('start', min_value=0, default=0) limit = self.request.get_range('num', min_value=1, max_value=100, default=10) namespace = self.request.get('namespace', default_value=None) resp = search.get_indexes(offset=start, limit=limit+1, namespace=namespace or '') has_more = len(resp.results) > limit indexes = resp.results[:limit] current_page = start / limit + 1 values = { 'request': self.request, 'namespace': namespace, 'has_namespace': namespace is not None, 'current_page': current_page, 'next_start': -1, 'prev_start': -1, 'num': limit, 'start': start, 'start_base_url': self.filter_url(['num', 'namespace']), 'next': urllib.quote(ustr(self.request.uri)), 'indexes': indexes} if current_page > 1: values['prev_start'] = int((current_page - 2) * limit) values['paging'] = True if has_more: values['next_start'] = int(current_page * limit) values['paging'] = True self.generate('search.html', values)
[docs]class SearchIndexHandler(BaseRequestHandler): """FTS index information.""" PATH = '/search_index' def _ProcessSearchResponse(self, response): """Format document list and produce corresponding hdf representation.""" documents = [] field_names = set() for result in response.results: doc = Document(result.doc_id) for field in result.fields: field_names.add( doc.fields[] = field documents.append(doc) field_names = sorted(field_names) docs = [] for doc in documents: doc_fields = [] for field_name in field_names: if field_name in doc.fields: value = TruncateValue(doc.fields[field_name].value) else: value = '' doc_fields.append(value) docs.append({ 'doc_id': doc.doc_id, 'fields': doc_fields }) return { 'documents': docs, 'field_names': field_names, }
[docs] def get(self): """Displays documents in a FTS index.""" start = self.request.get_range('start', min_value=0, default=0) query = self.request.get('query') namespace = self.request.get('namespace') limit = self.request.get_range('num', min_value=1, max_value=100, default=10) index_name = self.request.get('index') or 'index' index = search.Index(name=index_name, namespace=namespace) resp = query_string=query, options=search.QueryOptions(offset=start, limit=limit))) has_more = resp.number_found > start + limit current_page = start / limit + 1 values = { 'request': self.request, 'namespace': namespace, 'index': index_name, 'query': query, 'current_page': current_page, 'next_start': -1, 'prev_start': -1, 'start_base_url': self.filter_url([ 'query', 'index', 'num', 'namespace']), 'next': urllib.quote(ustr(self.request.uri)), 'values': self._ProcessSearchResponse(resp), 'prev': self.request.get( 'next', default_value=self.base_path() + SearchIndexesListHandler.PATH)} if current_page > 1: values['prev_start'] = int((current_page - 2) * limit) values['paging'] = True if has_more: values['next_start'] = int(current_page * limit) values['paging'] = True self.generate('search_index.html', values)
[docs]class SearchDocumentHandler(BaseRequestHandler): """FTS document information.""" PATH = '/search_document'
[docs] def get(self): """Displays FTS document.""" index_name = self.request.get('index') doc_id = self.request.get('id') namespace = self.request.get('namespace') doc = None index = search.Index(name=index_name, namespace=