def create_node(data, parent=None):
"""
Recursive function for seting a Node and then processing its children.
"""
node = Node(title=data['title'])
node.save()
node.set_fields(data)
# Remove tree fields so they can be set by MPTT framework
for field in TREE_FIELDS:
setattr(node, field, None)
node.parent = parent
node.save()
pk_list.append(node.pk)
children = [x for x in json_list if (x['tree_id'] == data['tree_id'] and
x['lft'] > data['lft'] and
x['rght'] < data['rght'] and
x['level'] == data['level']+1)]
for child in children:
create_node(child, node)
def create_node(data, parent=None):
"""
Recursive function for seting a Node and then processing its children.
"""
node = Node(title=data['title'])
node.save()
node.set_fields(data)
# Remove tree fields so they can be set by MPTT framework
for field in TREE_FIELDS:
setattr(node, field, None)
node.parent = parent
node.save()
pk_list.append(node.pk)
children = [
x for x in json_list
if (x['tree_id'] == data['tree_id'] and x['lft'] > data['lft'] and
x['rght'] < data['rght'] and x['level'] == data['level'] + 1)
]
for child in children:
create_node(child, node)
class NodeView(APIView):
"""
API for interacting with Node objects. Unauthenticated requests
are permitted but do not alter the database. Several query parameters
have special significance, otherwise query parameters are treated as
filters:
- 'context=[integer]': Any query with context (even if it is null) is
treated as an actions-list. The query indicates which Context object
should be used.
- 'field_group=[string]': Allows for an alternate set of fields to
be returned.
- 'upcoming=[date-string]': Requests a list of Node objects that are
due soon. The date-string should be UTC and ISO formatted (YYYY-mm-dd).
"""
def get(self, request, *args, **kwargs):
"""Returns the details of the node as a json encoded object"""
SERIALIZERS = {
'default': NodeSerializer,
'actions_list': NodeListSerializer,
'outline': NodeOutlineSerializer,
'calendar': CalendarSerializer,
'calendar_deadlines': CalendarDeadlineSerializer,
}
import time
get_dict = request.GET.copy()
node_id = kwargs.get('pk')
# Look for the reserved query parameters
if get_dict.get('upcoming', None):
# Get Nodes with upcoming deadline
nodes = self.get_upcoming(request, *args, **kwargs)
default_serializer = 'actions_list'
elif get_dict.get('context', None):
# Context is given, so this is an actions list
nodes = self.get_actions_list(request, *args, **kwargs)
default_serializer = 'actions_list'
elif node_id is not None:
# A specific Node object is requested
nodes = get_object_or_404(Node, pk=node_id)
default_serializer = 'default'
else:
nodes = self.get_queryset(request, *args, **kwargs)
default_serializer = 'default'
# Check for alternate serializer
try:
field_group = get_dict.pop('field_group')[0]
except KeyError:
field_group = default_serializer
Serializer = SERIALIZERS[field_group]
# Serialize and return the queryset or object
is_many = isinstance(nodes, QuerySet)
serializer = Serializer(nodes, many=is_many, request=request)
return Response(serializer.data)
def get_queryset(self, request, *args, **kwargs):
"""
Return a queryset for regular GET queries. If a context
is given as a query parameter, then the get_action_list()
method is to be used instead.
"""
BOOLS = ('archived',) # Translate 'False' -> False for these fields
M2M = ['todo_state'] # For filtering on arrays
nodes = Node.objects.mine(request.user, get_archived=True)
get_dict = request.GET.copy()
# get_dict = dict(request.GET)
parent_id = get_dict.get('parent_id', None)
if parent_id == '0':
nodes = nodes.filter(parent=None)
get_dict.pop('parent_id')
# Apply each criterion to the queryset
for key in get_dict.keys():
if key in BOOLS:
query = {key: False if get_dict[key] == 'false' else True}
elif key in M2M:
# Convert to (param__in=[]) style list filtering
value_list = get_dict.getlist(key)
param = "{}__in".format(key)
query = {param: value_list}
else:
query = {key: get_dict[key]}
try:
nodes = nodes.filter(**query)
except FieldError:
pass
nodes = nodes.select_related('owner')
nodes = nodes.prefetch_related('users', 'focus_areas')
return nodes
def get_actions_list(self, request, *args, **kwargs):
"""
Fetches a queryset for the requested "Next Actions" list.
Only called if 'context' is passed as a GET query parameter (even if
it equals None).
"""
# Filter by parent
parent_id = request.GET.get('parent', None)
if parent_id is not None:
parent = Node.objects.get(pk=parent_id)
nodes = parent.get_descendants(include_self=True)
else:
nodes = Node.objects.all()
nodes = nodes.assigned(request.user).select_related('todo_state')
# Filter by todo state
final_Q = Q()
todo_states_params = request.GET.getlist('todo_state')
todo_string = ''
for todo_state in todo_states_params:
final_Q = final_Q | Q(todo_state=todo_state)
nodes = nodes.filter(final_Q)
# Filter by FocusArea
focus_area = request.GET.get('focus_area', None)
if focus_area:
nodes = nodes.filter(focus_areas=focus_area)
# Filter by context
context_id = request.GET.get('context', None)
if context_id == '0':
request.session['context_id'] = None
elif context_id != 'None' and context_id is not None:
context = Context.objects.get(id=context_id)
nodes = context.apply(nodes)
request.session['context_id'] = context_id
request.session['context_name'] = context.name
# DB optimization
nodes = nodes.select_related('owner')
nodes = nodes.prefetch_related('users', 'focus_areas')
return nodes
def get_upcoming(self, request, *args, **kwargs):
"""
Get QuerySet with deadlines coming up based on
'upcoming' query parameter.
"""
deadline_period = 7 # in days
all_nodes_qs = Node.objects.mine(request.user)
target_date = dt.datetime.strptime(request.GET['upcoming'],
'%Y-%m-%d').date()
# Determine query filters for "Upcoming Deadlines" section
undone_Q = Q(todo_state__closed = False) | Q(todo_state = None)
deadline = target_date + dt.timedelta(days=deadline_period)
upcoming_deadline_Q = Q(deadline_date__lte = deadline) # TODO: fix this
deadline_nodes = all_nodes_qs.filter(undone_Q, upcoming_deadline_Q)
deadline_nodes = deadline_nodes.order_by("deadline_date")
# DB optimization
deadline_nodes = deadline_nodes.select_related('owner')
deadline_nodes = deadline_nodes.prefetch_related('focus_areas')
return deadline_nodes
def post(self, request, pk=None, *args, **kwargs):
"""
Create a new Node, conducted through JSON format:
{
id: [node primary key],
title: 'Everything's shiny, captn',
todo_state: 2,
etc...
}
Ignores fields related to MPTT for new nodes as these get
set automatically based on the 'parent' attribute.
Returns: JSON object of all node fields, with changes.
"""
data = request.data.copy()
if pk is not None:
# Cannot POST if a node is specified by primary key
return HttpResponseNotAllowed(['GET', 'PUT'])
# Create new node
self.node = Node()
if not request.user.is_anonymous:
self.node.owner = request.user
self.node.save()
# Set fields (ignore mptt fields for new nodes)
for key in ('id', 'tree_id', 'lft', 'rght', 'level'):
try:
data.pop(key)
except KeyError:
pass
self.node.set_fields(data)
self.node.save()
# Return newly saved node as json
self.node = Node.objects.get(pk=self.node.pk)
serializer = NodeSerializer(self.node, request=request)
data = serializer.data
# Don't keep nodes sent via the public interface
if request.user.is_anonymous:
self.node.delete()
return Response(data)
def put(self, request, pk=None, *args, **kwargs):
"""
Edit existing nodes through JSON format:
{
id: [node primary key],
title: 'Everything's shiny, captn',
todo_state: 2,
etc...
}
"""
if pk is None:
# Throw error response if user is trying to
# PUT without specifying a pk
return HttpResponseNotAllowed(['GET', 'POST'])
data = request.data.copy()
# Remove tree metadata from the request
TREE_FIELDS = ('lft', 'rght', 'level', 'tree_id')
for key in TREE_FIELDS:
try:
data.pop(key)
except KeyError:
pass
# Check the permissions of the Node
node = get_object_or_404(Node, pk=pk)
access = node.access_level(request.user)
if ((request.user.is_anonymous and node.owner is not None) or
(not request.user.is_anonymous and access != 'write')):
# Not authorized
return HttpResponse(
json.dumps({'status': 'failure',
'reason': 'unauthorized'}),
status=401)
# Update and return the Node
node.set_fields(data)
if not request.user.is_anonymous:
node.save()
node = Node.objects.get(pk=node.pk)
serializer = NodeSerializer(node, request=request)
return Response(serializer.data)
class NodeView(APIView):
"""
API for interacting with Node objects. Unauthenticated requests
are permitted but do not alter the database. Several query parameters
have special significance, otherwise query parameters are treated as
filters:
- 'context=[integer]': Any query with context (even if it is null) is
treated as an actions-list. The query indicates which Context object
should be used.
- 'field_group=[string]': Allows for an alternate set of fields to
be returned.
- 'upcoming=[date-string]': Requests a list of Node objects that are
due soon. The date-string should be UTC and ISO formatted (YYYY-mm-dd).
"""
def get(self, request, *args, **kwargs):
"""Returns the details of the node as a json encoded object"""
SERIALIZERS = {
'default': NodeSerializer,
'actions_list': NodeListSerializer,
'outline': NodeOutlineSerializer,
'calendar': CalendarSerializer,
'calendar_deadlines': CalendarDeadlineSerializer,
}
get_dict = request.GET.copy()
node_id = kwargs.get('pk')
# Look for the reserved query parameters
if get_dict.get('upcoming', None):
# Get Nodes with upcoming deadline
nodes = self.get_upcoming(request, *args, **kwargs)
default_serializer = 'actions_list'
elif get_dict.get('context', None):
# Context is given, so this is an actions list
nodes = self.get_actions_list(request, *args, **kwargs)
default_serializer = 'actions_list'
elif node_id is not None:
# A specific Node object is requested
nodes = get_object_or_404(Node, pk=node_id)
default_serializer = 'default'
else:
nodes = self.get_queryset(request, *args, **kwargs)
default_serializer = 'default'
# Check for alternate serializer
try:
field_group = get_dict.pop('field_group')[0]
except KeyError:
field_group = default_serializer
Serializer = SERIALIZERS[field_group]
# Serialize and return the queryset or object
is_many = isinstance(nodes, QuerySet)
serializer = Serializer(nodes, many=is_many, request=request)
return Response(serializer.data)
def get_queryset(self, request, *args, **kwargs):
"""
Return a queryset for regular GET queries. If a context
is given as a query parameter, then the get_action_list()
method is to be used instead.
"""
BOOLS = ('archived', ) # Translate 'False' -> False for these fields
M2M = ['todo_state'] # For filtering on arrays
nodes = Node.objects.mine(request.user, get_archived=True)
get_dict = request.GET.copy()
# get_dict = dict(request.GET)
parent_id = get_dict.get('parent_id', None)
if parent_id == '0':
nodes = nodes.filter(parent=None)
get_dict.pop('parent_id')
# Apply each criterion to the queryset
for key in get_dict.keys():
if key in BOOLS:
query = {key: False if get_dict[key] == 'false' else True}
elif key in M2M:
# Convert to (param__in=[]) style list filtering
value_list = get_dict.getlist(key)
param = "{}__in".format(key)
query = {param: value_list}
else:
query = {key: get_dict[key]}
try:
nodes = nodes.filter(**query)
except FieldError:
pass
nodes = nodes.select_related('owner')
nodes = nodes.prefetch_related('users', 'focus_areas')
return nodes
def get_actions_list(self, request, *args, **kwargs):
"""
Fetches a queryset for the requested "Next Actions" list.
Only called if 'context' is passed as a GET query parameter (even if
it equals None).
"""
# Filter by parent
parent_id = request.GET.get('parent', None)
if parent_id is not None:
parent = Node.objects.get(pk=parent_id)
nodes = parent.get_descendants(include_self=True)
else:
nodes = Node.objects.all()
nodes = nodes.assigned(request.user).select_related('todo_state')
# Filter by todo state
final_Q = Q()
todo_states_params = request.GET.getlist('todo_state')
todo_string = ''
for todo_state in todo_states_params:
final_Q = final_Q | Q(todo_state=todo_state)
nodes = nodes.filter(final_Q)
# Filter by FocusArea
focus_area = request.GET.get('focus_area', None)
if focus_area:
nodes = nodes.filter(focus_areas=focus_area)
# Filter by context
context_id = request.GET.get('context', None)
if context_id == '0':
request.session['context_id'] = None
elif context_id != 'None' and context_id is not None:
context = Context.objects.get(id=context_id)
nodes = context.apply(nodes)
request.session['context_id'] = context_id
request.session['context_name'] = context.name
# DB optimization
nodes = nodes.select_related('owner')
nodes = nodes.prefetch_related('users', 'focus_areas')
return nodes
def get_upcoming(self, request, *args, **kwargs):
"""
Get QuerySet with deadlines coming up based on
'upcoming' query parameter.
"""
deadline_period = 7 # in days
all_nodes_qs = Node.objects.mine(request.user)
target_date = dt.datetime.strptime(request.GET['upcoming'],
'%Y-%m-%d').date()
# Determine query filters for "Upcoming Deadlines" section
undone_Q = Q(todo_state__closed=False) | Q(todo_state=None)
deadline = target_date + dt.timedelta(days=deadline_period)
upcoming_deadline_Q = Q(deadline_date__lte=deadline)
scheduled_Q = ~Q(scheduled_date__gt=target_date) | Q(
deadline_date__lte=target_date)
deadline_nodes = all_nodes_qs.filter(undone_Q, upcoming_deadline_Q,
scheduled_Q)
deadline_nodes = deadline_nodes.order_by("deadline_date")
# DB optimization
deadline_nodes = deadline_nodes.select_related('owner')
deadline_nodes = deadline_nodes.prefetch_related('focus_areas')
return deadline_nodes
def post(self, request, pk=None, *args, **kwargs):
"""
Create a new Node, conducted through JSON format:
{
id: [node primary key],
title: 'Everything's shiny, captn',
todo_state: 2,
etc...
}
Ignores fields related to MPTT for new nodes as these get
set automatically based on the 'parent' attribute.
Returns: JSON object of all node fields, with changes.
"""
data = request.data.copy()
if pk is not None:
# Cannot POST if a node is specified by primary key
return HttpResponseNotAllowed(['GET', 'PUT'])
# Create new node
self.node = Node()
if not request.user.is_anonymous:
self.node.owner = request.user
self.node.save()
# Set fields (ignore mptt fields for new nodes)
for key in ('id', 'tree_id', 'lft', 'rght', 'level'):
try:
data.pop(key)
except KeyError:
pass
self.node.set_fields(data)
self.node.save()
# Return newly saved node as json
self.node = Node.objects.get(pk=self.node.pk)
serializer = NodeSerializer(self.node, request=request)
data = serializer.data
# Don't keep nodes sent via the public interface
if request.user.is_anonymous:
self.node.delete()
return Response(data)
def put(self, request, pk=None, *args, **kwargs):
"""
Edit existing nodes through JSON format:
{
id: [node primary key],
title: 'Everything's shiny, captn',
todo_state: 2,
etc...
}
"""
if pk is None:
# Throw error response if user is trying to
# PUT without specifying a pk
return HttpResponseNotAllowed(['GET', 'POST'])
data = request.data.copy()
# Remove tree metadata from the request
TREE_FIELDS = ('lft', 'rght', 'level', 'tree_id')
for key in TREE_FIELDS:
try:
data.pop(key)
except KeyError:
pass
# Check the permissions of the Node
node = get_object_or_404(Node, pk=pk)
access = node.access_level(request.user)
if ((request.user.is_anonymous and node.owner is not None)
or (not request.user.is_anonymous and access != 'write')):
# Not authorized
return HttpResponse(json.dumps({
'status': 'failure',
'reason': 'unauthorized'
}),
status=401)
# Update and return the Node
node.set_fields(data)
if not request.user.is_anonymous:
node.save()
node = Node.objects.get(pk=node.pk)
serializer = NodeSerializer(node, request=request)
return Response(serializer.data)
