allobjects · allobjectvars · hoc_pointer_ · name_declared · object_id · object_pop · object_push · symbols
Debugging and Internals Access
type = h.name_declared("name")
type = h.name_declared("name", 1)
type = h.name_declared("name", 2)
Return 0 if the name is not in the NEURON/HOC symbol table. The first form looks for names in the top level symbol table. The second form looks in the current object context. The last form also looks in the top level symbol table but is useful in Python to distinguish subtypes of variables which appear as doubles in HOC but internally are really not doubles and so cannot be pointed to by double*, eg. h.secondorder which is <type ‘int’> or h.nseg which returns either
TypeError: Section access unspecifiedor
nseg not a USERPROPERTY that can be pointed to
If the name exists return
2 if an objref
3 if a Section
4 if a strdef
5 if a scalar or double variable. (if second arg is not 2)
if second arg is 2
5 if a scalar double
6 if a double array
7 if an integer
8 if a section property
This function checks the NEURON/HOC symbol table; Python objects are handled separately.
To test if a simple name is a local variable in Python, use:
if 'soma' in locals(): # do something
dir()are also often useful.
If the name is known in advance, use a
exceptblock and catch NameError and AttributeError:
try: h.soma.squiggle except (NameError, AttributeError): print('Name does not exist')
Combining this with an
evalcan allow testing arbitrary names, but is potentially unsafe as it allows execution of arbitrary code.
- symbols — type the names of HOC functions and variables
- Types a list of functions and variable names defined in HOC. Dimensions of arrays are also indicated.
No longer works. The nearest replacement is
Returns 0 if the object reference does not point to an object instance. (Otherwise returns the pointer cast to a double, not a very useful number, except that this is equal to the value returned by Python’s
If the second argument is 1, it returns the index of the object name. Returns -1 if the object is the NULLObject.
from neuron import h a, b, c = h.List(), h.List(), h.Vector() print(h.object_id(a)) # displays a double; equal to hash(a) print(h.object_id(a, 1)) # 0 since a == h.List print(h.object_id(b, 1)) # 1 since b == h.List print(h.object_id(c, 1)) # 0 since c == h.Vector
- Prints all the HOC object references (objref variables) that have been declared along with the class type of the object they reference and the number of references. Objects created via Python and not assigned to a HOC objref
>>> h('objref foo') 1 >>> h.foo = h.Vector() >>> h.allobjectvars() obp hoc_obj_ -> NULL obp hoc_obj_ -> NULL obp foo -> Vector with 1 refs. 0.0 >>> banana = h.foo >>> h.allobjectvars() obp hoc_obj_ -> NULL obp hoc_obj_ -> NULL obp foo -> Vector with 2 refs. 0.0
nref = h.allobjects(objectref)
Prints the internal names of all class instances (objects) available from the interpreter along with the number of references to them.
With a templatename the list is restricted to objects of that class.
With an object variable, nothing is printed but the reference count is returned. The count is too large by one if the argument was of the form templatename[index] since a temporary reference is created while the object is on the stack during the call.
>>> v = h.Vector() >>> foo = h.List() >>> h.allobjects() List with 1 refs Vector with 1 refs 0.0 >>> h.allobjects('Vector') Vector with 1 refs 0.0 >>> h.allobjects(foo) 2.0
- Enter the context of the object referenced by objref. In this context you can directly access any variables or call any functions, even those not declared as public. Do not attempt to create any new symbol names! This function is generally used by the object itself to save its state in a session.
- Pop the last object from an