Files
edx-platform/xmodule/modulestore/mongo/base.py

1902 lines
78 KiB
Python

"""
Modulestore backed by Mongodb.
Stores individual XModules as single documents with the following
structure:
{
'_id': <location.as_dict>,
'metadata': <dict containing all Scope.settings fields>
'definition': <dict containing all Scope.content fields>
'definition.children': <list of all child text_type(location)s>
}
"""
import copy
import logging
import re
import sys
from datetime import datetime
from importlib import import_module
from uuid import uuid4
import pymongo
from bson.son import SON
from fs.osfs import OSFS
from mongodb_proxy import autoretry_read
from opaque_keys.edx.keys import CourseKey, UsageKey
from opaque_keys.edx.locator import BlockUsageLocator, CourseLocator, LibraryLocator
from path import Path as path
from pytz import UTC
from xblock.core import XBlock
from xblock.exceptions import InvalidScopeError
from xblock.fields import Reference, ReferenceList, ReferenceValueDict, Scope, ScopeIds
from xblock.runtime import KvsFieldData
from xmodule.assetstore import AssetMetadata, CourseAssetsFromStorage
from xmodule.course_block import CourseSummary
from xmodule.error_block import ErrorBlock
from xmodule.errortracker import exc_info_to_str, null_error_tracker
from xmodule.exceptions import HeartbeatFailure
from xmodule.mako_block import MakoDescriptorSystem
from xmodule.modulestore import BulkOperationsMixin, ModuleStoreEnum, ModuleStoreWriteBase
from xmodule.modulestore.draft_and_published import DIRECT_ONLY_CATEGORIES, ModuleStoreDraftAndPublished
from xmodule.modulestore.edit_info import EditInfoRuntimeMixin
from xmodule.modulestore.exceptions import DuplicateCourseError, ItemNotFoundError, ReferentialIntegrityError
from xmodule.modulestore.inheritance import InheritanceKeyValueStore, InheritanceMixin, inherit_metadata
from xmodule.modulestore.store_utilities import DETACHED_XBLOCK_TYPES
from xmodule.modulestore.xml import CourseLocationManager
from xmodule.mongo_utils import connect_to_mongodb, create_collection_index
from xmodule.partitions.partitions_service import PartitionService
from xmodule.services import SettingsService
log = logging.getLogger(__name__)
# sort order that returns DRAFT items first
SORT_REVISION_FAVOR_DRAFT = ('_id.revision', pymongo.DESCENDING)
# sort order that returns PUBLISHED items first
SORT_REVISION_FAVOR_PUBLISHED = ('_id.revision', pymongo.ASCENDING)
BLOCK_TYPES_WITH_CHILDREN = list({
name for name, class_ in XBlock.load_classes() if getattr(class_, 'has_children', False)
})
# Allow us to call _from_deprecated_(son|string) throughout the file
# pylint: disable=protected-access
# at module level, cache one instance of OSFS per filesystem root.
_OSFS_INSTANCE = {}
class MongoRevisionKey:
"""
Key Revision constants to use for Location and Usage Keys in the Mongo modulestore
Note: These values are persisted in the database, so should not be changed without migrations
"""
draft = 'draft'
published = None
class InvalidWriteError(Exception):
"""
Raised to indicate that writing to a particular key
in the KeyValueStore is disabled
"""
pass # lint-amnesty, pylint: disable=unnecessary-pass
class MongoKeyValueStore(InheritanceKeyValueStore):
"""
A KeyValueStore that maps keyed data access to one of the 3 data areas
known to the MongoModuleStore (data, children, and metadata)
"""
def __init__(self, data, parent, children, metadata):
super().__init__()
if not isinstance(data, dict):
self._data = {'data': data}
else:
self._data = data
self._parent = parent
self._children = children
self._metadata = metadata
def get(self, key):
if key.scope == Scope.children:
return self._children
elif key.scope == Scope.parent:
return self._parent
elif key.scope == Scope.settings:
return self._metadata[key.field_name]
elif key.scope == Scope.content:
return self._data[key.field_name]
else:
raise InvalidScopeError(
key,
(Scope.children, Scope.parent, Scope.settings, Scope.content),
)
def set(self, key, value):
if key.scope == Scope.children:
self._children = value
elif key.scope == Scope.parent:
self._parent = value
elif key.scope == Scope.settings:
self._metadata[key.field_name] = value
elif key.scope == Scope.content:
self._data[key.field_name] = value
else:
raise InvalidScopeError(
key,
(Scope.children, Scope.settings, Scope.content),
)
def delete(self, key):
if key.scope == Scope.children:
self._children = []
elif key.scope == Scope.settings:
if key.field_name in self._metadata:
del self._metadata[key.field_name]
elif key.scope == Scope.content:
if key.field_name in self._data:
del self._data[key.field_name]
else:
raise InvalidScopeError(
key,
(Scope.children, Scope.settings, Scope.content),
)
def has(self, key):
if key.scope in (Scope.children, Scope.parent):
return True
elif key.scope == Scope.settings:
return key.field_name in self._metadata
elif key.scope == Scope.content:
return key.field_name in self._data
else:
return False
def __repr__(self):
return "MongoKeyValueStore{!r}<{!r}, {!r}>".format(
(self._data, self._parent, self._children, self._metadata),
self._fields,
self.inherited_settings
)
class CachingDescriptorSystem(MakoDescriptorSystem, EditInfoRuntimeMixin): # lint-amnesty, pylint: disable=abstract-method
"""
A system that has a cache of block json that it will use to load blocks
from, with a backup of calling to the underlying modulestore for more data
"""
def __repr__(self):
return "CachingDescriptorSystem{!r}".format((
self.modulestore,
str(self.course_id),
[str(key) for key in self.module_data.keys()],
self.default_class,
))
def __init__(self, modulestore, course_key, module_data, default_class, **kwargs):
"""
modulestore: the module store that can be used to retrieve additional blocks
course_key: the course for which everything in this runtime will be relative
module_data: a dict mapping Location -> json that was cached from the
underlying modulestore
default_class: The default_class to use when loading an
XModuleDescriptor from the module_data
resources_fs: a filesystem, as per MakoDescriptorSystem
error_tracker: a function that logs errors for later display to users
render_template: a function for rendering templates, as per
MakoDescriptorSystem
"""
id_manager = CourseLocationManager(course_key)
kwargs.setdefault('id_reader', id_manager)
kwargs.setdefault('id_generator', id_manager)
super().__init__(
load_item=self.load_item,
**kwargs
)
self.modulestore = modulestore
self.module_data = module_data
self.default_class = default_class
# cdodge: other Systems have a course_id attribute defined. To keep things consistent, let's
# define an attribute here as well, even though it's None
self.course_id = course_key
def load_item(self, location, for_parent=None): # lint-amnesty, pylint: disable=method-hidden
"""
Return an XBlock instance for the specified location
"""
assert isinstance(location, UsageKey)
if location.run is None:
# self.module_data is keyed on locations that have full run information.
# If the supplied location is missing a run, then we will miss the cache and
# incur an additional query.
# TODO: make module_data a proper class that can handle this itself.
location = location.replace(course_key=self.modulestore.fill_in_run(location.course_key))
json_data = self.module_data.get(location)
if json_data is None:
block = self.modulestore.get_item(location, using_descriptor_system=self)
return block
else:
# load the block and apply the inherited metadata
try:
category = json_data['location']['category']
class_ = self.load_block_type(category)
definition = json_data.get('definition', {})
metadata = json_data.get('metadata', {})
children = [
self._convert_reference_to_key(childloc)
for childloc in definition.get('children', [])
]
parent = None
if category not in DETACHED_XBLOCK_TYPES.union(['course']):
# try looking it up just-in-time (but not if we're working with a detached block).
parent = self.modulestore.get_parent_location(
as_published(location),
ModuleStoreEnum.RevisionOption.published_only if location.branch is None
else ModuleStoreEnum.RevisionOption.draft_preferred
)
data = definition.get('data', {})
if isinstance(data, str):
data = {'data': data}
mixed_class = self.mixologist.mix(class_)
if data: # empty or None means no work
data = self._convert_reference_fields_to_keys(mixed_class, location.course_key, data)
metadata = self._convert_reference_fields_to_keys(mixed_class, location.course_key, metadata)
kvs = MongoKeyValueStore(
data,
parent,
children,
metadata,
)
field_data = KvsFieldData(kvs)
scope_ids = ScopeIds(None, category, location, location)
block = self.construct_xblock_from_class(class_, scope_ids, field_data, for_parent=for_parent)
non_draft_loc = as_published(location)
metadata_inheritance_tree = self.modulestore._compute_metadata_inheritance_tree(self.course_id)
inherit_metadata(block, metadata_inheritance_tree.get(str(non_draft_loc), {}))
block._edit_info = json_data.get('edit_info')
# migrate published_by and published_on if edit_info isn't present
if block._edit_info is None:
block._edit_info = {}
raw_metadata = json_data.get('metadata', {})
# published_on was previously stored as a list of time components instead of a datetime
if raw_metadata.get('published_date'):
block._edit_info['published_date'] = datetime(
*raw_metadata.get('published_date')[0:6]
).replace(tzinfo=UTC)
block._edit_info['published_by'] = raw_metadata.get('published_by')
for wrapper in self.modulestore.xblock_field_data_wrappers:
block._field_data = wrapper(block, block._field_data) # pylint: disable=protected-access
# decache any computed pending field settings
block.save()
return block
except Exception: # pylint: disable=broad-except
log.warning("Failed to load descriptor from %s", json_data, exc_info=True)
return ErrorBlock.from_json(
json_data,
self,
location,
error_msg=exc_info_to_str(sys.exc_info())
)
def _convert_reference_to_key(self, ref_string):
"""
Convert a single serialized UsageKey string in a ReferenceField into a UsageKey.
"""
key = UsageKey.from_string(ref_string)
return key.replace(run=self.modulestore.fill_in_run(key.course_key).run)
def _convert_reference_fields_to_keys(self, class_, course_key, jsonfields): # lint-amnesty, pylint: disable=unused-argument
"""
Find all fields of type reference and convert the payload into UsageKeys
:param class_: the XBlock class
:param course_key: a CourseKey object for the given course
:param jsonfields: a dict of the jsonified version of the fields
"""
result = {}
for field_name, value in jsonfields.items():
field = class_.fields.get(field_name)
if field is None:
continue
elif value is None:
result[field_name] = value
elif isinstance(field, Reference):
result[field_name] = self._convert_reference_to_key(value)
elif isinstance(field, ReferenceList):
result[field_name] = [
self._convert_reference_to_key(ele) for ele in value
]
elif isinstance(field, ReferenceValueDict):
result[field_name] = {
key: self._convert_reference_to_key(subvalue) for key, subvalue in value.items()
}
else:
result[field_name] = value
return result
def lookup_item(self, location):
"""
Returns the JSON payload of the xblock at location.
"""
try:
json = self.module_data[location]
except KeyError:
json = self.modulestore._find_one(location)
self.module_data[location] = json
return json
def get_edited_by(self, xblock):
"""
See :class: cms.lib.xblock.runtime.EditInfoRuntimeMixin
"""
return xblock._edit_info.get('edited_by')
def get_edited_on(self, xblock):
"""
See :class: cms.lib.xblock.runtime.EditInfoRuntimeMixin
"""
return xblock._edit_info.get('edited_on')
def get_subtree_edited_by(self, xblock):
"""
See :class: cms.lib.xblock.runtime.EditInfoRuntimeMixin
"""
return xblock._edit_info.get('subtree_edited_by')
def get_subtree_edited_on(self, xblock):
"""
See :class: cms.lib.xblock.runtime.EditInfoRuntimeMixin
"""
return xblock._edit_info.get('subtree_edited_on')
def get_published_by(self, xblock):
"""
See :class: cms.lib.xblock.runtime.EditInfoRuntimeMixin
"""
return xblock._edit_info.get('published_by')
def get_published_on(self, xblock):
"""
See :class: cms.lib.xblock.runtime.EditInfoRuntimeMixin
"""
return xblock._edit_info.get('published_date')
def applicable_aside_types(self, block):
# "old" mongo does support asides yet
return []
# The only thing using this w/ wildcards is contentstore.mongo for asset retrieval
def location_to_query(location, wildcard=True, tag='i4x'):
"""
Takes a Location and returns a SON object that will query for that location by subfields
rather than subdoc.
Fields in location that are None are ignored in the query.
If `wildcard` is True, then a None in a location is treated as a wildcard
query. Otherwise, it is searched for literally
"""
query = location.to_deprecated_son(prefix='_id.', tag=tag)
if wildcard:
for key, value in query.items():
# don't allow wildcards on revision, since public is set as None, so
# its ambiguous between None as a real value versus None=wildcard
if value is None and key != '_id.revision':
del query[key]
return query
def as_draft(location):
"""
Returns the Location that is the draft for `location`
If the location is in the DIRECT_ONLY_CATEGORIES, returns itself
"""
if location.block_type in DIRECT_ONLY_CATEGORIES:
return location
return location.replace(revision=MongoRevisionKey.draft)
def as_published(location):
"""
Returns the Location that is the published version for `location`
"""
return location.replace(revision=MongoRevisionKey.published)
class MongoBulkOpsMixin(BulkOperationsMixin):
"""
Mongo bulk operation support
"""
def _end_outermost_bulk_operation(self, bulk_ops_record, structure_key):
"""
The outermost nested bulk_operation call: do the actual end of the bulk operation.
"""
return True
def _is_in_bulk_operation(self, course_id, ignore_case=False): # lint-amnesty, pylint: disable=arguments-differ
"""
Returns whether a bulk operation is in progress for the given course.
"""
return super()._is_in_bulk_operation(
course_id.for_branch(None), ignore_case
)
class ParentLocationCache(dict):
"""
Dict-based object augmented with a more cache-like interface, for internal use.
"""
def has(self, key):
return key in self
def set(self, key, value):
self[key] = value
def delete_by_value(self, value):
keys_to_delete = [k for k, v in self.items() if v == value]
for key in keys_to_delete:
del self[key]
class MongoModuleStore(ModuleStoreDraftAndPublished, ModuleStoreWriteBase, MongoBulkOpsMixin):
"""
A Mongodb backed ModuleStore
"""
# If no name is specified for the asset metadata collection, this name is used.
DEFAULT_ASSET_COLLECTION_NAME = 'assetstore'
# TODO (cpennington): Enable non-filesystem filestores
# pylint: disable=invalid-name
# pylint: disable=attribute-defined-outside-init
def __init__(self, contentstore, doc_store_config, fs_root, render_template,
default_class=None,
error_tracker=null_error_tracker,
i18n_service=None,
fs_service=None,
user_service=None,
signal_handler=None,
retry_wait_time=0.1,
**kwargs):
"""
:param doc_store_config: must have a host, db, and collection entries. Other common entries: port, tz_aware.
"""
super().__init__(contentstore=contentstore, **kwargs)
def do_connection(
db, collection, host, port=27017, tz_aware=True, user=None, password=None, asset_collection=None, **kwargs
):
"""
Create & open the connection, authenticate, and provide pointers to the collection
"""
# Set a write concern of 1, which makes writes complete successfully to the primary
# only before returning. Also makes pymongo report write errors.
kwargs['w'] = 1
self.database = connect_to_mongodb(
db, host,
port=port, tz_aware=tz_aware, user=user, password=password,
retry_wait_time=retry_wait_time, **kwargs
)
self.collection = self.database[collection]
# Collection which stores asset metadata.
if asset_collection is None:
asset_collection = self.DEFAULT_ASSET_COLLECTION_NAME
self.asset_collection = self.database[asset_collection]
do_connection(**doc_store_config)
if default_class is not None:
module_path, _, class_name = default_class.rpartition('.')
try:
class_ = getattr(import_module(module_path), class_name)
except (ImportError, AttributeError):
fallback_module_path = "xmodule.hidden_block"
fallback_class_name = "HiddenBlock"
log.exception(
"Failed to import the default store class. "
f"Falling back to {fallback_module_path}.{fallback_class_name}"
)
class_ = getattr(import_module(fallback_module_path), fallback_class_name)
self.default_class = class_
else:
self.default_class = None
self.fs_root = path(fs_root)
self.error_tracker = error_tracker
self.render_template = render_template
self.i18n_service = i18n_service
self.fs_service = fs_service
self.user_service = user_service
self._course_run_cache = {}
self.signal_handler = signal_handler
def close_connections(self):
"""
Closes any open connections to the underlying database
"""
self.collection.database.client.close()
def _drop_database(self, database=True, collections=True, connections=True):
"""
A destructive operation to drop the underlying database and close all connections.
Intended to be used by test code for cleanup.
If database is True, then this should drop the entire database.
Otherwise, if collections is True, then this should drop all of the collections used
by this modulestore.
Otherwise, the modulestore should remove all data from the collections.
If connections is True, then close the connection to the database as well.
"""
# drop the assets
super()._drop_database(database, collections, connections)
connection = self.collection.database.client
if database:
connection.drop_database(self.collection.database.proxied_object)
elif collections:
self.collection.drop()
else:
self.collection.delete_many({})
if connections:
connection.close()
@autoretry_read()
def fill_in_run(self, course_key):
"""
In mongo some course_keys are used without runs. This helper function returns
a course_key with the run filled in, if the course does actually exist.
"""
if course_key.run is not None:
return course_key
cache_key = (course_key.org, course_key.course)
if cache_key not in self._course_run_cache:
matching_courses = list(self.collection.find(SON([
('_id.tag', 'i4x'),
('_id.org', course_key.org),
('_id.course', course_key.course),
('_id.category', 'course'),
])).limit(1))
if not matching_courses:
return course_key
self._course_run_cache[cache_key] = matching_courses[0]['_id']['name']
return course_key.replace(run=self._course_run_cache[cache_key])
def for_branch_setting(self, location):
"""
Returns the Location that is for the current branch setting.
"""
if location.block_type in DIRECT_ONLY_CATEGORIES:
return location.replace(revision=MongoRevisionKey.published)
if self.get_branch_setting() == ModuleStoreEnum.Branch.draft_preferred:
return location.replace(revision=MongoRevisionKey.draft)
return location.replace(revision=MongoRevisionKey.published)
def _get_parent_cache(self, branch):
"""
Provides a reference to one of the two branch-specific
ParentLocationCaches associated with the current request (if any).
"""
if self.request_cache is not None:
return self.request_cache.data.setdefault(f'parent-location-{branch}', ParentLocationCache())
else:
return ParentLocationCache()
def _compute_metadata_inheritance_tree(self, course_id):
'''
Find all inheritable fields from all xblocks in the course which may define inheritable data
'''
# get all collections in the course, this query should not return any leaf nodes
course_id = self.fill_in_run(course_id)
query = SON([
('_id.tag', 'i4x'),
('_id.org', course_id.org),
('_id.course', course_id.course),
('_id.category', {'$in': BLOCK_TYPES_WITH_CHILDREN})
])
# if we're only dealing in the published branch, then only get published containers
if self.get_branch_setting() == ModuleStoreEnum.Branch.published_only:
query['_id.revision'] = None
# we just want the Location, children, and inheritable metadata
record_filter = {'_id': 1, 'definition.children': 1}
# just get the inheritable metadata since that is all we need for the computation
# this minimizes both data pushed over the wire
for field_name in InheritanceMixin.fields:
record_filter[f'metadata.{field_name}'] = 1
# call out to the DB
resultset = self.collection.find(query, record_filter)
# it's ok to keep these as deprecated strings b/c the overall cache is indexed by course_key and this
# is a dictionary relative to that course
results_by_url = {}
root = None
# now go through the results and order them by the location url
for result in resultset:
# manually pick it apart b/c the db has tag and we want as_published revision regardless
location = as_published(BlockUsageLocator._from_deprecated_son(result['_id'], course_id.run))
location_url = str(location)
if location_url in results_by_url:
# found either draft or live to complement the other revision
# FIXME this is wrong. If the child was moved in draft from one parent to the other, it will
# show up under both in this logic: https://openedx.atlassian.net/browse/TNL-1075
existing_children = results_by_url[location_url].get('definition', {}).get('children', [])
additional_children = result.get('definition', {}).get('children', [])
total_children = existing_children + additional_children
# use set to get rid of duplicates. We don't care about order; so, it shouldn't matter.
results_by_url[location_url].setdefault('definition', {})['children'] = set(total_children)
else:
results_by_url[location_url] = result
if location.block_type == 'course': # pylint: disable=no-member
root = location_url
# now traverse the tree and compute down the inherited metadata
metadata_to_inherit = {}
def _compute_inherited_metadata(url):
"""
Helper method for computing inherited metadata for a specific location url
"""
my_metadata = results_by_url[url].get('metadata', {})
# go through all the children and recurse, but only if we have
# in the result set. Remember results will not contain leaf nodes
for child in results_by_url[url].get('definition', {}).get('children', []):
if child in results_by_url:
new_child_metadata = copy.deepcopy(my_metadata)
new_child_metadata.update(results_by_url[child].get('metadata', {}))
results_by_url[child]['metadata'] = new_child_metadata
metadata_to_inherit[child] = new_child_metadata
_compute_inherited_metadata(child)
else:
# this is likely a leaf node, so let's record what metadata we need to inherit
metadata_to_inherit[child] = my_metadata.copy()
# WARNING: 'parent' is not part of inherited metadata, but
# we're piggybacking on this recursive traversal to grab
# and cache the child's parent, as a performance optimization.
# The 'parent' key will be popped out of the dictionary during
# CachingDescriptorSystem.load_item
metadata_to_inherit[child].setdefault('parent', {})[self.get_branch_setting()] = url
if root is not None:
_compute_inherited_metadata(root)
return metadata_to_inherit
def _clean_item_data(self, item):
"""
Renames the '_id' field in item to 'location'
"""
item['location'] = item['_id']
del item['_id']
@autoretry_read()
def _query_children_for_cache_children(self, course_key, items):
"""
Generate a pymongo in query for finding the items and return the payloads
"""
# first get non-draft in a round-trip
query = {
'_id': {'$in': [
UsageKey.from_string(item).map_into_course(course_key).to_deprecated_son() for item in items
]}
}
return list(self.collection.find(query))
def _cache_children(self, course_key, items, depth=0):
"""
Returns a dictionary mapping Location -> item data, populated with json data
for all descendents of items up to the specified depth.
(0 = no descendents, 1 = children, 2 = grandchildren, etc)
If depth is None, will load all the children.
This will make a number of queries that is linear in the depth.
"""
data = {}
to_process = list(items)
course_key = self.fill_in_run(course_key)
parent_cache = self._get_parent_cache(self.get_branch_setting())
while to_process and (depth is None or depth >= 0):
children = []
for item in to_process:
self._clean_item_data(item)
item_location = BlockUsageLocator._from_deprecated_son(item['location'], course_key.run)
item_children = item.get('definition', {}).get('children', [])
children.extend(item_children)
for item_child in item_children:
parent_cache.set(item_child, item_location)
data[item_location] = item
if depth == 0:
break
# Load all children by id. See
# http://www.mongodb.org/display/DOCS/Advanced+Queries#AdvancedQueries-%24or
# for or-query syntax
to_process = []
if children:
to_process = self._query_children_for_cache_children(course_key, children)
# If depth is None, then we just recurse until we hit all the descendents
if depth is not None:
depth -= 1
return data
def _load_item(self, course_key, item, data_cache,
using_descriptor_system=None, for_parent=None):
"""
Load an XModuleDescriptor from item, using the children stored in data_cache
Arguments:
course_key (CourseKey): which course to load from
item (dict): A dictionary with the following keys:
location: The serialized UsageKey for the item to load
data_dir (optional): The directory name to use as the root data directory for this XModule
data_cache (dict): A dictionary mapping from UsageKeys to xblock field data
(this is the xblock data loaded from the database)
using_descriptor_system (CachingDescriptorSystem): The existing CachingDescriptorSystem
to add data to, and to load the XBlocks from.
for_parent (:class:`XBlock`): The parent of the XBlock being loaded.
"""
course_key = self.fill_in_run(course_key)
location = BlockUsageLocator._from_deprecated_son(item['location'], course_key.run)
data_dir = getattr(item, 'data_dir', location.course)
root = self.fs_root / data_dir
resource_fs = _OSFS_INSTANCE.setdefault(root, OSFS(root, create=True))
if using_descriptor_system is None:
services = {}
if self.i18n_service:
services["i18n"] = self.i18n_service
if self.fs_service:
services["fs"] = self.fs_service
if self.user_service:
services["user"] = self.user_service
services["settings"] = SettingsService()
if self.request_cache:
services["request_cache"] = self.request_cache
services["partitions"] = PartitionService(course_key)
system = CachingDescriptorSystem(
modulestore=self,
course_key=course_key,
module_data=data_cache,
default_class=self.default_class,
resources_fs=resource_fs,
error_tracker=self.error_tracker,
render_template=self.render_template,
mixins=self.xblock_mixins,
select=self.xblock_select,
disabled_xblock_types=self.disabled_xblock_types,
services=services,
)
else:
system = using_descriptor_system
system.module_data.update(data_cache)
item = system.get_block(location, for_parent=for_parent)
# TODO Once TNL-5092 is implemented, we can remove the following line
# of code. Until then, set the course_version field on the block to be
# consistent with the Split modulestore. Since Mongo modulestore doesn't
# maintain course versions set it to None.
item.course_version = None
return item
def _load_items(self, course_key, items, depth=0, using_descriptor_system=None, for_parent=None):
"""
Load a list of xblocks from the data in items, with children cached up
to specified depth
"""
course_key = self.fill_in_run(course_key)
data_cache = self._cache_children(course_key, items, depth)
# if we are loading a course object, if we're not prefetching children (depth != 0) then don't
# bother with the metadata inheritance
return [
self._load_item(
course_key,
item,
data_cache,
using_descriptor_system=using_descriptor_system,
for_parent=for_parent,
)
for item in items
]
@autoretry_read()
def get_course_summaries(self, **kwargs):
"""
Returns a list of `CourseSummary`. This accepts an optional parameter of 'org' which
will apply an efficient filter to only get courses with the specified ORG
"""
def extract_course_summary(course):
"""
Extract course information from the course block for mongo.
"""
return {
field: course['metadata'][field]
for field in CourseSummary.course_info_fields
if field in course['metadata']
}
course_records = []
query = {'_id.category': 'course'}
course_org_filter = kwargs.get('org')
course_keys = kwargs.get('course_keys')
if course_keys:
course_queries = []
for course_key in course_keys:
course_query = {
f'_id.{value_attr}': getattr(course_key, key_attr)
for key_attr, value_attr in {'org': 'org', 'course': 'course', 'run': 'name'}.items()
}
course_query.update(query)
course_queries.append(course_query)
query = {'$or': course_queries}
elif course_org_filter:
query['_id.org'] = course_org_filter
course_records = self.collection.find(query, {'metadata': True})
courses_summaries = []
for course in course_records:
if not (course['_id']['org'] == 'edx' and course['_id']['course'] == 'templates'):
locator = CourseKey.from_string('/'.join(
[course['_id']['org'], course['_id']['course'], course['_id']['name']]
))
course_summary = extract_course_summary(course)
courses_summaries.append(
CourseSummary(locator, **course_summary)
)
return courses_summaries
@autoretry_read()
def get_courses(self, **kwargs):
'''
Returns a list of course descriptors. This accepts an optional parameter of 'org' which
will apply an efficient filter to only get courses with the specified ORG
'''
course_org_filter = kwargs.get('org')
if course_org_filter:
course_records = self.collection.find({'_id.category': 'course', '_id.org': course_org_filter})
else:
course_records = self.collection.find({'_id.category': 'course'})
base_list = sum(
[
self._load_items(
CourseKey.from_string('/'.join(
[course['_id']['org'], course['_id']['course'], course['_id']['name']]
)),
[course]
)
for course
# I tried to add '$and': [{'_id.org': {'$ne': 'edx'}}, {'_id.course': {'$ne': 'templates'}}]
# but it didn't do the right thing (it filtered all edx and all templates out)
in course_records
if not ( # TODO kill this
course['_id']['org'] == 'edx' and
course['_id']['course'] == 'templates'
)
],
[]
)
return [course for course in base_list if not isinstance(course, ErrorBlock)]
@autoretry_read()
def _find_one(self, location):
'''Look for a given location in the collection. If the item is not present, raise
ItemNotFoundError.
'''
assert isinstance(location, UsageKey)
item = self.collection.find_one(
{'_id': location.to_deprecated_son()}
)
if item is None:
raise ItemNotFoundError(location)
return item
def make_course_key(self, org, course, run):
"""
Return a valid :class:`~opaque_keys.edx.keys.CourseKey` for this modulestore
that matches the supplied `org`, `course`, and `run`.
This key may represent a course that doesn't exist in this modulestore.
"""
return CourseLocator(org, course, run, deprecated=True)
def make_course_usage_key(self, course_key):
"""
Return a valid :class:`~opaque_keys.edx.keys.UsageKey` for this modulestore
that matches the supplied course_key.
"""
return BlockUsageLocator(course_key, 'course', course_key.run)
def get_course(self, course_key, depth=0, **kwargs): # lint-amnesty, pylint: disable=arguments-differ
"""
Get the course with the given courseid (org/course/run)
"""
assert isinstance(course_key, CourseKey)
if not course_key.deprecated: # split course_key
# The supplied CourseKey is of the wrong type, so it can't possibly be stored in this modulestore.
raise ItemNotFoundError(course_key)
course_key = self.fill_in_run(course_key)
location = course_key.make_usage_key('course', course_key.run)
try:
return self.get_item(location, depth=depth)
except ItemNotFoundError:
return None
@autoretry_read()
def has_course(self, course_key, ignore_case=False, **kwargs): # lint-amnesty, pylint: disable=arguments-differ
"""
Returns the course_id of the course if it was found, else None
Note: we return the course_id instead of a boolean here since the found course may have
a different id than the given course_id when ignore_case is True.
If ignore_case is True, do a case insensitive search,
otherwise, do a case sensitive search
"""
assert isinstance(course_key, CourseKey)
if not course_key.deprecated: # split course_key
# The supplied CourseKey is of the wrong type, so it can't possibly be stored in this modulestore.
return False
if isinstance(course_key, LibraryLocator):
return None # Libraries require split mongo
course_key = self.fill_in_run(course_key)
location = course_key.make_usage_key('course', course_key.run)
if ignore_case:
course_query = location.to_deprecated_son('_id.')
for key in course_query.keys():
if isinstance(course_query[key], str):
course_query[key] = re.compile(r"(?i)^{}$".format(course_query[key]))
else:
course_query = {'_id': location.to_deprecated_son()}
course = self.collection.find_one(course_query, projection={'_id': True})
if course:
return CourseKey.from_string('/'.join([
course['_id']['org'], course['_id']['course'], course['_id']['name']]
))
else:
return None
def has_item(self, usage_key):
"""
Returns True if location exists in this ModuleStore.
"""
try:
self._find_one(usage_key)
return True
except ItemNotFoundError:
return False
def get_item(self, usage_key, depth=0, using_descriptor_system=None, for_parent=None, **kwargs): # lint-amnesty, pylint: disable=arguments-differ
"""
Returns an XModuleDescriptor instance for the item at location.
If any segment of the location is None except revision, raises
xmodule.modulestore.exceptions.InsufficientSpecificationError
If no object is found at that location, raises
xmodule.modulestore.exceptions.ItemNotFoundError
Arguments:
usage_key: a :class:`.UsageKey` instance
depth (int): An argument that some module stores may use to prefetch
descendents of the queried blocks for more efficient results later
in the request. The depth is counted in the number of
calls to get_children() to cache. None indicates to cache all descendents.
using_descriptor_system (CachingDescriptorSystem): The existing CachingDescriptorSystem
to add data to, and to load the XBlocks from.
"""
item = self._find_one(usage_key)
block = self._load_items(
usage_key.course_key,
[item],
depth,
using_descriptor_system=using_descriptor_system,
for_parent=for_parent,
)[0]
return block
@staticmethod
def _course_key_to_son(course_id, tag='i4x'):
"""
Generate the partial key to look up items relative to a given course
"""
return SON([
('_id.tag', tag),
('_id.org', course_id.org),
('_id.course', course_id.course),
])
@staticmethod
def _id_dict_to_son(id_dict):
"""
Generate the partial key to look up items relative to a given course
"""
return SON([
(key, id_dict[key])
for key in ('tag', 'org', 'course', 'category', 'name', 'revision')
])
@autoretry_read()
def get_items( # lint-amnesty, pylint: disable=arguments-differ
self,
course_id,
settings=None,
content=None,
key_revision=MongoRevisionKey.published,
qualifiers=None,
using_descriptor_system=None,
**kwargs
):
"""
Returns:
list of XModuleDescriptor instances for the matching items within the course with
the given course_id
NOTE: don't use this to look for courses
as the course_id is required. Use get_courses which is a lot faster anyway.
If you don't provide a value for revision, this limits the result to only ones in the
published course. Call this method on draft mongo store if you want to include drafts.
Args:
course_id (CourseKey): the course identifier
settings (dict): fields to look for which have settings scope. Follows same syntax
and rules as qualifiers below
content (dict): fields to look for which have content scope. Follows same syntax and
rules as qualifiers below.
key_revision (str): the revision of the items you're looking for.
MongoRevisionKey.draft - only returns drafts
MongoRevisionKey.published (equates to None) - only returns published
If you want one of each matching xblock but preferring draft to published, call this same method
on the draft modulestore with ModuleStoreEnum.RevisionOption.draft_preferred.
qualifiers (dict): what to look for within the course.
Common qualifiers are ``category`` or any field name. if the target field is a list,
then it searches for the given value in the list not list equivalence.
Substring matching pass a regex object.
For this modulestore, ``name`` is a commonly provided key (Location based stores)
This modulestore does not allow searching dates by comparison or edited_by, previous_version,
update_version info.
using_descriptor_system (CachingDescriptorSystem): The existing CachingDescriptorSystem
to add data to, and to load the XBlocks from.
"""
qualifiers = qualifiers.copy() if qualifiers else {} # copy the qualifiers (destructively manipulated here)
query = self._course_key_to_son(course_id)
query['_id.revision'] = key_revision
for field in ['category', 'name']:
if field in qualifiers:
qualifier_value = qualifiers.pop(field)
if isinstance(qualifier_value, list):
qualifier_value = {'$in': qualifier_value}
query['_id.' + field] = qualifier_value
for key, value in (settings or {}).items():
query['metadata.' + key] = value
for key, value in (content or {}).items():
query['definition.data.' + key] = value
if 'children' in qualifiers:
query['definition.children'] = qualifiers.pop('children')
query.update(qualifiers)
items = self.collection.find(
query,
sort=[SORT_REVISION_FAVOR_DRAFT],
)
blocks = self._load_items(
course_id,
list(items),
using_descriptor_system=using_descriptor_system
)
return blocks
def create_course(self, org, course, run, user_id, fields=None, **kwargs): # lint-amnesty, pylint: disable=arguments-differ
"""
Creates and returns the course.
Args:
org (str): the organization that owns the course
course (str): the name of the course
run (str): the name of the run
user_id: id of the user creating the course
fields (dict): Fields to set on the course at initialization
kwargs: Any optional arguments understood by a subset of modulestores to customize instantiation
Returns: a CourseBlock
Raises:
InvalidLocationError: If a course with the same org, course, and run already exists
"""
course_id = CourseKey.from_string('/'.join([org, course, run]))
# Check if a course with this org/course has been defined before (case-insensitive)
course_search_location = SON([
('_id.tag', 'i4x'),
('_id.org', re.compile(f'^{course_id.org}$', re.IGNORECASE)),
('_id.course', re.compile(f'^{course_id.course}$', re.IGNORECASE)),
('_id.category', 'course'),
])
courses = self.collection.find(course_search_location, projection={'_id': True})
try:
course = courses.next()
raise DuplicateCourseError(course_id, course['_id'])
except StopIteration:
pass
with self.bulk_operations(course_id):
xblock = self.create_item(user_id, course_id, 'course', course_id.run, fields=fields, **kwargs)
# create any other necessary things as a side effect
super().create_course(
org, course, run, user_id, runtime=xblock.runtime, **kwargs
)
return xblock
def create_xblock(
self, runtime, course_key, block_type, block_id=None, fields=None,
metadata=None, definition_data=None, **kwargs
):
"""
Create the new xblock but don't save it. Returns the new block.
:param runtime: if you already have an xblock from the course, the xblock.runtime value
:param fields: a dictionary of field names and values for the new xblock
"""
if metadata is None:
metadata = {}
if definition_data is None:
definition_data = {}
# @Cale, should this use LocalId like we do in split?
if block_id is None:
if block_type == 'course':
block_id = course_key.run
else:
block_id = '{}_{}'.format(block_type, uuid4().hex[:5])
if runtime is None:
services = {}
if self.i18n_service:
services["i18n"] = self.i18n_service
if self.fs_service:
services["fs"] = self.fs_service
if self.user_service:
services["user"] = self.user_service
services["partitions"] = PartitionService(course_key)
runtime = CachingDescriptorSystem(
modulestore=self,
module_data={},
course_key=course_key,
default_class=self.default_class,
resources_fs=None,
error_tracker=self.error_tracker,
render_template=self.render_template,
mixins=self.xblock_mixins,
select=self.xblock_select,
services=services,
)
xblock_class = runtime.load_block_type(block_type)
location = course_key.make_usage_key(block_type, block_id)
dbmodel = self._create_new_field_data(block_type, location, definition_data, metadata)
xblock = runtime.construct_xblock_from_class(
xblock_class,
# We're loading a descriptor, so student_id is meaningless
# We also don't have separate notions of definition and usage ids yet,
# so we use the location for both.
ScopeIds(None, block_type, location, location),
dbmodel,
for_parent=kwargs.get('for_parent'),
)
if fields is not None:
for key, value in fields.items():
setattr(xblock, key, value)
# decache any pending field settings from init
xblock.save()
return xblock
def create_item(self, user_id, course_key, block_type, block_id=None, **kwargs): # lint-amnesty, pylint: disable=arguments-differ
"""
Creates and saves a new item in a course.
Returns the newly created item.
Args:
user_id: ID of the user creating and saving the xblock
course_key: A :class:`~opaque_keys.edx.CourseKey` identifying which course to create
this item in
block_type: The typo of block to create
block_id: a unique identifier for the new item. If not supplied,
a new identifier will be generated
"""
if block_id is None:
if block_type == 'course':
block_id = course_key.run
else:
block_id = '{}_{}'.format(block_type, uuid4().hex[:5])
runtime = kwargs.pop('runtime', None)
xblock = self.create_xblock(runtime, course_key, block_type, block_id, **kwargs)
xblock = self.update_item(xblock, user_id, allow_not_found=True)
return xblock
def create_child(self, user_id, parent_usage_key, block_type, block_id=None, **kwargs): # lint-amnesty, pylint: disable=arguments-differ
"""
Creates and saves a new xblock that as a child of the specified block
Returns the newly created item.
Args:
user_id: ID of the user creating and saving the xblock
parent_usage_key: a :class:`~opaque_key.edx.UsageKey` identifing the
block that this item should be parented under
block_type: The typo of block to create
block_id: a unique identifier for the new item. If not supplied,
a new identifier will be generated
"""
# attach to parent if given
parent = None
if parent_usage_key is not None:
parent = self.get_item(parent_usage_key)
kwargs.setdefault('for_parent', parent)
xblock = self.create_item(user_id, parent_usage_key.course_key, block_type, block_id=block_id, **kwargs)
if parent is not None and 'detached' not in xblock._class_tags:
# Originally added to support entrance exams (settings.FEATURES.get('ENTRANCE_EXAMS'))
if kwargs.get('position') is None:
parent.children.append(xblock.location)
else:
parent.children.insert(kwargs.get('position'), xblock.location)
self.update_item(parent, user_id, child_update=True) # lint-amnesty, pylint: disable=unexpected-keyword-arg
return xblock
def import_xblock(self, user_id, course_key, block_type, block_id, fields=None, runtime=None, **kwargs):
"""
Simple implementation of overwriting any existing xblock
"""
if block_type == 'course':
block_id = course_key.run
xblock = self.create_xblock(runtime, course_key, block_type, block_id, fields)
return self.update_item(xblock, user_id, allow_not_found=True)
def _get_course_for_item(self, location, depth=0):
'''
for a given XBlock, return the course that it belongs to
Also we have to assert that this block maps to only one course item - it'll throw an
assert if not
'''
return self.get_course(location.course_key, depth)
def _update_single_item(self, location, update, allow_not_found=False):
"""
Set update on the specified item, and raises ItemNotFoundError
if the location doesn't exist
"""
bulk_record = self._get_bulk_ops_record(location.course_key)
bulk_record.dirty = True
# See http://www.mongodb.org/display/DOCS/Updating for
# atomic update syntax
result = self.collection.update_one(
{'_id': location.to_deprecated_son()},
{'$set': update},
upsert=allow_not_found,
)
if result.matched_count == 0 and result.upserted_id is None:
raise ItemNotFoundError(location)
def _update_ancestors(self, location, update):
"""
Recursively applies update to all the ancestors of location
"""
parent = self._get_raw_parent_location(as_published(location), ModuleStoreEnum.RevisionOption.draft_preferred)
if parent:
self._update_single_item(parent, update)
self._update_ancestors(parent, update)
def update_item(self, xblock, user_id, allow_not_found=False, force=False, isPublish=False, # lint-amnesty, pylint: disable=arguments-differ
is_publish_root=True):
"""
Update the persisted version of xblock to reflect its current values.
xblock: which xblock to persist
user_id: who made the change (ignored for now by this modulestore)
allow_not_found: whether to create a new object if one didn't already exist or give an error
force: force is meaningless for this modulestore
isPublish: an internal parameter that indicates whether this update is due to a Publish operation, and
thus whether the item's published information should be updated.
is_publish_root: when publishing, this indicates whether xblock is the root of the publish and should
therefore propagate subtree edit info up the tree
"""
course_key = xblock.location.course_key
try:
definition_data = self._serialize_scope(xblock, Scope.content)
now = datetime.now(UTC)
payload = {
'definition.data': definition_data,
'metadata': self._serialize_scope(xblock, Scope.settings),
'edit_info': {
'edited_on': now,
'edited_by': user_id,
'subtree_edited_on': now,
'subtree_edited_by': user_id,
}
}
if isPublish:
payload['edit_info']['published_date'] = now
payload['edit_info']['published_by'] = user_id
elif 'published_date' in getattr(xblock, '_edit_info', {}):
payload['edit_info']['published_date'] = xblock._edit_info['published_date']
payload['edit_info']['published_by'] = xblock._edit_info['published_by']
if xblock.has_children:
children = self._serialize_scope(xblock, Scope.children)
payload.update({'definition.children': children['children']})
# Remove all old pointers to me, then add my current children back
parent_cache = self._get_parent_cache(self.get_branch_setting())
parent_cache.delete_by_value(xblock.location)
for child in xblock.children:
parent_cache.set(str(child), xblock.location)
self._update_single_item(xblock.scope_ids.usage_id, payload, allow_not_found=allow_not_found)
# update subtree edited info for ancestors
# don't update the subtree info for descendants of the publish root for efficiency
if not isPublish or (isPublish and is_publish_root):
ancestor_payload = {
'edit_info.subtree_edited_on': now,
'edit_info.subtree_edited_by': user_id
}
self._update_ancestors(xblock.scope_ids.usage_id, ancestor_payload)
# update the edit info of the instantiated xblock
xblock._edit_info = payload['edit_info']
# fire signal that we've written to DB
except ItemNotFoundError:
if not allow_not_found: # lint-amnesty, pylint: disable=no-else-raise
raise
elif not self.has_course(course_key):
raise ItemNotFoundError(course_key) # lint-amnesty, pylint: disable=raise-missing-from
return xblock
def _serialize_scope(self, xblock, scope):
"""
Find all fields of type reference and convert the payload from UsageKeys to deprecated strings
:param xblock: the XBlock class
:param jsonfields: a dict of the jsonified version of the fields
"""
jsonfields = {}
for field_name, field in xblock.fields.items():
if field.scope == scope and field.is_set_on(xblock):
if field.scope == Scope.parent:
continue
elif isinstance(field, Reference):
jsonfields[field_name] = str(field.read_from(xblock))
elif isinstance(field, ReferenceList):
jsonfields[field_name] = [
str(ele) for ele in field.read_from(xblock)
]
elif isinstance(field, ReferenceValueDict):
jsonfields[field_name] = {
key: str(subvalue) for key, subvalue in field.read_from(xblock).items()
}
else:
jsonfields[field_name] = field.read_json(xblock)
return jsonfields
def _get_non_orphan_parents(self, location, parents, revision):
"""
Extract non orphan parents by traversing the list of possible parents and remove current location
from orphan parents to avoid parents calculation overhead next time.
"""
non_orphan_parents = []
# get bulk_record once rather than for each iteration
bulk_record = self._get_bulk_ops_record(location.course_key)
for parent in parents:
parent_loc = BlockUsageLocator._from_deprecated_son(parent['_id'], location.course_key.run)
# travel up the tree for orphan validation
ancestor_loc = parent_loc
while ancestor_loc is not None:
current_loc = ancestor_loc
ancestor_loc = self._get_raw_parent_location(as_published(current_loc), revision)
if ancestor_loc is None:
bulk_record.dirty = True
# The parent is an orphan, so remove all the children including
# the location whose parent we are looking for from orphan parent
self.collection.update_one(
{'_id': parent_loc.to_deprecated_son()},
{'$set': {'definition.children': []}},
upsert=True,
)
elif ancestor_loc.block_type == 'course':
# once we reach the top location of the tree and if the location is not an orphan then the
# parent is not an orphan either
non_orphan_parents.append(parent_loc)
break
return non_orphan_parents
def _get_raw_parent_location(self, location, revision=ModuleStoreEnum.RevisionOption.published_only):
'''
Helper for get_parent_location that finds the location that is the parent of this location in this course,
but does NOT return a version agnostic location.
'''
assert location.branch is None
assert revision == ModuleStoreEnum.RevisionOption.published_only \
or revision == ModuleStoreEnum.RevisionOption.draft_preferred # lint-amnesty, pylint: disable=consider-using-in
parent_cache = self._get_parent_cache(self.get_branch_setting())
if parent_cache.has(str(location)):
return parent_cache.get(str(location))
# create a query with tag, org, course, and the children field set to the given location
query = self._course_key_to_son(location.course_key)
query['definition.children'] = str(location)
# if only looking for the PUBLISHED parent, set the revision in the query to None
if revision == ModuleStoreEnum.RevisionOption.published_only:
query['_id.revision'] = MongoRevisionKey.published
def cache_and_return(parent_loc):
parent_cache.set(str(location), parent_loc)
return parent_loc
# query the collection, sorting by DRAFT first
parents = list(
self.collection.find(query, {'_id': True}, sort=[SORT_REVISION_FAVOR_DRAFT])
)
if len(parents) == 0:
# no parents were found
return cache_and_return(None)
if revision == ModuleStoreEnum.RevisionOption.published_only:
if len(parents) > 1:
non_orphan_parents = self._get_non_orphan_parents(location, parents, revision)
if len(non_orphan_parents) == 0:
# no actual parent found
return cache_and_return(None)
if len(non_orphan_parents) > 1: # lint-amnesty, pylint: disable=no-else-raise
# should never have multiple PUBLISHED parents
raise ReferentialIntegrityError(
"{} parents claim {}".format(len(parents), location)
)
else:
return cache_and_return(non_orphan_parents[0].replace(run=location.course_key.run))
else:
# return the single PUBLISHED parent
return cache_and_return(BlockUsageLocator._from_deprecated_son(parents[0]['_id'],
location.course_key.run))
else:
# there could be 2 different parents if
# (1) the draft item was moved or
# (2) the parent itself has 2 versions: DRAFT and PUBLISHED
# if there are multiple parents with version PUBLISHED then choose from non-orphan parents
all_parents = []
published_parents = 0
for parent in parents:
if parent['_id']['revision'] is None:
published_parents += 1
all_parents.append(parent)
# since we sorted by SORT_REVISION_FAVOR_DRAFT, the 0'th parent is the one we want
if published_parents > 1:
non_orphan_parents = self._get_non_orphan_parents(location, all_parents, revision)
return cache_and_return(non_orphan_parents[0].replace(run=location.course_key.run))
found_id = all_parents[0]['_id']
# don't disclose revision outside modulestore
return cache_and_return(BlockUsageLocator._from_deprecated_son(found_id, location.course_key.run))
def get_parent_location(self, location, revision=ModuleStoreEnum.RevisionOption.published_only, **kwargs):
'''
Find the location that is the parent of this location in this course.
Returns: version agnostic location (revision always None) as per the rest of mongo.
Args:
revision:
ModuleStoreEnum.RevisionOption.published_only
- return only the PUBLISHED parent if it exists, else returns None
ModuleStoreEnum.RevisionOption.draft_preferred
- return either the DRAFT or PUBLISHED parent,
preferring DRAFT, if parent(s) exists,
else returns None
'''
parent = self._get_raw_parent_location(location, revision)
if parent:
return parent
return None
def get_modulestore_type(self, course_key=None): # lint-amnesty, pylint: disable=arguments-differ, unused-argument
"""
Returns an enumeration-like type reflecting the type of this modulestore per ModuleStoreEnum.Type
Args:
course_key: just for signature compatibility
"""
return ModuleStoreEnum.Type.mongo
def get_orphans(self, course_key, **kwargs):
"""
Return an array of all of the locations for orphans in the course.
"""
course_key = self.fill_in_run(course_key)
detached_categories = [name for name, __ in XBlock.load_tagged_classes("detached")]
query = self._course_key_to_son(course_key)
query['_id.category'] = {'$nin': detached_categories}
all_items = self.collection.find(query)
all_reachable = set()
item_locs = set()
for item in all_items:
if item['_id']['category'] != 'course':
# It would be nice to change this method to return UsageKeys instead of the deprecated string.
item_locs.add(
str(as_published(BlockUsageLocator._from_deprecated_son(item['_id'], course_key.run)))
)
all_reachable = all_reachable.union(item.get('definition', {}).get('children', []))
item_locs -= all_reachable
return [UsageKey.from_string(item_loc).map_into_course(course_key) for item_loc in item_locs]
def get_courses_for_wiki(self, wiki_slug, **kwargs):
"""
Return the list of courses which use this wiki_slug
:param wiki_slug: the course wiki root slug
:return: list of course keys
"""
courses = self.collection.find(
{'_id.category': 'course', 'definition.data.wiki_slug': wiki_slug},
{'_id': True}
)
# the course's run == its name. It's the only xblock for which that's necessarily true.
return [
BlockUsageLocator._from_deprecated_son(course['_id'], course['_id']['name']).course_key
for course in courses
]
def _create_new_field_data(self, _category, _location, definition_data, metadata):
"""
To instantiate a new xblock which will be saved later, set up the dbModel and kvs
"""
kvs = MongoKeyValueStore(
definition_data,
None,
[],
metadata,
)
field_data = KvsFieldData(kvs)
return field_data
def _find_course_assets(self, course_key):
"""
Internal; finds (or creates) course asset info about all assets for a particular course
Arguments:
course_key (CourseKey): course identifier
Returns:
CourseAssetsFromStorage object, wrapping the relevant Mongo doc. If asset metadata
exists, other keys will be the other asset types with values as lists of asset metadata.
"""
# Using the course_key, find or insert the course asset metadata document.
# A single document exists per course to store the course asset metadata.
course_key = self.fill_in_run(course_key)
if course_key.run is None:
log.warning('No run found for combo org "{}" course "{}" on asset request.'.format(
course_key.org, course_key.course
))
course_assets = None
else:
# Complete course key, so query for asset metadata.
course_assets = self.asset_collection.find_one(
{'course_id': str(course_key)},
)
doc_id = None if course_assets is None else course_assets['_id']
if course_assets is None:
# Check to see if the course is created in the course collection.
if self.get_course(course_key) is None: # lint-amnesty, pylint: disable=no-else-raise
raise ItemNotFoundError(course_key)
else:
# Course exists, so create matching assets document.
course_assets = {'course_id': str(course_key), 'assets': {}}
doc_id = self.asset_collection.insert_one(course_assets).inserted_id
elif isinstance(course_assets['assets'], list):
# This record is in the old course assets format.
# Ensure that no data exists before updating the format.
assert len(course_assets['assets']) == 0
# Update the format to a dict.
self.asset_collection.update_one(
{'_id': doc_id},
{'$set': {'assets': {}}}
)
# Pass back wrapped 'assets' dict with the '_id' key added to it for document update purposes.
return CourseAssetsFromStorage(course_key, doc_id, course_assets['assets'])
def _make_mongo_asset_key(self, asset_type):
"""
Given a asset type, form a key needed to update the proper embedded field in the Mongo doc.
"""
return f'assets.{asset_type}'
def _save_asset_metadata_list(self, asset_metadata_list, user_id, import_only):
"""
Internal; saves the info for a particular course's asset.
Arguments:
asset_metadata_list (list(AssetMetadata)): list of data about several course assets
user_id (int|long): user ID saving the asset metadata
import_only (bool): True if edited_on/by data should remain unchanged.
"""
course_key = asset_metadata_list[0].asset_id.course_key
course_assets = self._find_course_assets(course_key)
assets_by_type = self._save_assets_by_type(course_key, asset_metadata_list, course_assets, user_id, import_only)
# Build an update set with potentially multiple embedded fields.
updates_by_type = {}
for asset_type, assets in assets_by_type.items():
updates_by_type[self._make_mongo_asset_key(asset_type)] = list(assets)
# Update the document.
self.asset_collection.update_one(
{'_id': course_assets.doc_id},
{'$set': updates_by_type}
)
return True
def save_asset_metadata(self, asset_metadata, user_id, import_only=False):
"""
Saves the info for a particular course's asset.
Arguments:
asset_metadata (AssetMetadata): data about the course asset data
user_id (int|long): user ID saving the asset metadata
import_only (bool): True if importing without editing, False if editing
Returns:
True if info save was successful, else False
"""
return self._save_asset_metadata_list([asset_metadata, ], user_id, import_only)
def save_asset_metadata_list(self, asset_metadata_list, user_id, import_only=False):
"""
Saves the asset metadata for each asset in a list of asset metadata.
Optimizes the saving of many assets.
Args:
asset_metadata (AssetMetadata): data about the course asset data
user_id (int|long): user ID saving the asset metadata
import_only (bool): True if importing without editing, False if editing
Returns:
True if info save was successful, else False
"""
return self._save_asset_metadata_list(asset_metadata_list, user_id, import_only)
def copy_all_asset_metadata(self, source_course_key, dest_course_key, user_id):
"""
Copy all the course assets from source_course_key to dest_course_key.
If dest_course already has assets, this removes the previous value.
It doesn't combine the assets in dest.
Arguments:
source_course_key (CourseKey): identifier of course to copy from
dest_course_key (CourseKey): identifier of course to copy to
"""
source_assets = self._find_course_assets(source_course_key)
dest_assets = {'assets': source_assets.asset_md.copy(), 'course_id': str(dest_course_key)}
self.asset_collection.delete_many({'course_id': str(dest_course_key)})
# Update the document.
self.asset_collection.insert_one(dest_assets)
def set_asset_metadata_attrs(self, asset_key, attr_dict, user_id): # lint-amnesty, pylint: disable=arguments-differ
"""
Add/set the given dict of attrs on the asset at the given location. Value can be any type which pymongo accepts.
Arguments:
asset_key (AssetKey): asset identifier
attr_dict (dict): attribute: value pairs to set
Raises:
ItemNotFoundError if no such item exists
AttributeError is attr is one of the build in attrs.
"""
course_assets, asset_idx = self._find_course_asset(asset_key)
if asset_idx is None:
raise ItemNotFoundError(asset_key)
# Form an AssetMetadata.
all_assets = course_assets[asset_key.asset_type]
md = AssetMetadata(asset_key, asset_key.path)
md.from_storable(all_assets[asset_idx])
md.update(attr_dict)
# Generate a Mongo doc from the metadata and update the course asset info.
all_assets[asset_idx] = md.to_storable()
self.asset_collection.update_one(
{'_id': course_assets.doc_id},
{"$set": {self._make_mongo_asset_key(asset_key.asset_type): all_assets}}
)
def delete_asset_metadata(self, asset_key, user_id):
"""
Internal; deletes a single asset's metadata.
Arguments:
asset_key (AssetKey): key containing original asset filename
Returns:
Number of asset metadata entries deleted (0 or 1)
"""
course_assets, asset_idx = self._find_course_asset(asset_key)
if asset_idx is None:
return 0
all_asset_info = course_assets[asset_key.asset_type]
all_asset_info.pop(asset_idx)
# Update the document.
self.asset_collection.update_one(
{'_id': course_assets.doc_id},
{'$set': {self._make_mongo_asset_key(asset_key.asset_type): all_asset_info}}
)
return 1
def delete_all_asset_metadata(self, course_key, user_id): # lint-amnesty, pylint: disable=unused-argument
"""
Delete all of the assets which use this course_key as an identifier.
Arguments:
course_key (CourseKey): course_identifier
"""
# Using the course_id, find the course asset metadata document.
# A single document exists per course to store the course asset metadata.
try:
course_assets = self._find_course_assets(course_key)
self.asset_collection.delete_many({'_id': course_assets.doc_id})
except ItemNotFoundError:
# When deleting asset metadata, if a course's asset metadata is not present, no big deal.
pass
def heartbeat(self):
"""
Check that the db is reachable.
"""
try:
# The ismaster command is cheap and does not require auth.
self.database.client.admin.command('ismaster')
return {ModuleStoreEnum.Type.mongo: True}
except pymongo.errors.ConnectionFailure:
raise HeartbeatFailure(f"Can't connect to {self.database.name}", 'mongo') # lint-amnesty, pylint: disable=raise-missing-from
def ensure_indexes(self):
"""
Ensure that all appropriate indexes are created that are needed by this modulestore, or raise
an exception if unable to.
This method is intended for use by tests and administrative commands, and not
to be run during server startup.
"""
# Because we often query for some subset of the id, we define this index:
create_collection_index(
self.collection,
[
('_id.tag', pymongo.ASCENDING),
('_id.org', pymongo.ASCENDING),
('_id.course', pymongo.ASCENDING),
('_id.category', pymongo.ASCENDING),
('_id.name', pymongo.ASCENDING),
('_id.revision', pymongo.ASCENDING),
],
background=True
)
# Because we often scan for all category='course' regardless of the value of the other fields:
create_collection_index(self.collection, '_id.category', background=True)
# Because lms calls get_parent_locations frequently (for path generation):
create_collection_index(self.collection, 'definition.children', sparse=True, background=True)
# To allow prioritizing draft vs published material
create_collection_index(self.collection, '_id.revision', background=True)
# Some overrides that still need to be implemented by subclasses
def convert_to_draft(self, location, user_id):
raise NotImplementedError()
def delete_item(self, location, user_id, **kwargs): # lint-amnesty, pylint: disable=arguments-differ
raise NotImplementedError()
def has_changes(self, xblock):
raise NotImplementedError()
def has_published_version(self, xblock):
raise NotImplementedError()
def publish(self, location, user_id):
raise NotImplementedError()
def revert_to_published(self, location, user_id):
raise NotImplementedError()
def unpublish(self, location, user_id):
raise NotImplementedError()