object
object
in the
Python
Glossary
object
in the
built-in
functions
in the
Python
Standard Library
In Variable number, we saw one way of passing a variable number of arguments to a function. In this example, we’ll see another way of doing it. But the machinery that we begin to introduce here will go far beyond this purpose.
An
object
is a big value that can contain smaller values inside of it.
Not every type of value can do this;
for example, an
int
cannot contain a smaller
int.
The data type of an object is called a
class.
For example, the object
mark
in
line
33
of
person.py
is an object of the class
Person
in
lines
9–10.
person
will therefore be able to contain smaller values inside of it.
The name of a class conventionally begins with an uppercase letter
(Person
in lines
9
and
15);
the name of an object of that class
conventionally begins with a lowercase letter
(mark
in
line
33,
person
in
line
13).
The smaller values contained in an object are called the
attributes
of the object.
The object
mark
is born without attributes because the
class
definition
in
lines
9–10
does not list any attributes.
It only has the word
pass,
which in this context means
“no attributes”.
We’ll talk about the
object
in
line
9
when we do
inheritance.
In the meantime, see
“Classes without an inheritance list”
in
Class
definitions.
Line
34
puts an attribute into the object
mark.
The name of the attribute is
lastName
and the value of the attribute is the
string
"Meretzky".
The
for
loop in
lines
39–40
loops through all the attributes of the object,
just to show you what attributes the object has at this point.
(See the dictionary
__dict__.)
In real life,
you’d probably print just the
individual attributes
you’re interested in, e.g.,
print(mark.lastName).
Before: lastName Meretzky age 64 favoriteSongs ['Help!', 'Day Tripper', 'Piggies'] After: lastName Meretzky favoriteSongs ['Help!', 'Day Tripper', 'Piggies', 'Revolution'] hair black
Why would we want to attach attributes to an object?
Well, it’s one way of grouping a set of variables into a family
by giving them a common last name:
mark.age,
mark.hair,
mark.favoriteSongs,
etc.
And now that they’re a family,
they can be passed simultaneously to a function
(line
42),
returned simultaneously from a function,
destroyed simultaneously (we haven’t talked about this yet),
etc.
I’m making an object sound like a dictionary. After all, a dictionary can be passed to a function, returned from a function, etc. So why create an object instead of a dictionary? It’s because an object can have more than just variables belonging to it. See the next example.
if hasattr(person, "age"):
del person.age
There is a slight danger in the
LBYL
style if your program is
multi-threaded:
another thread might delete the
age
attribute
at an instant between the
if
and the
del.
lastName.
Had we not known this,
we could not have written the
dot
and the
lastName
in that line.
How could we
not
know the name of the attribute when we’re writing
the program?
Well, the name of the attribute might come from the user’s input
as the program is running.
In that case,
we’d have to create the attribute with the
setattr
function.
Change
line
34
from
mark.lastName = "Meretzky"to
attributeName = input("What do you want to name the attribute? ") setattr(mark, attributeName, "Meretzky")
What do you want to name the attribute? apellido Before: apellido Meretzky age 64 favoriteSongs ['Help!', 'Day Tripper', 'Piggies'] After: apellido Meretzky favoriteSongs ['Help!', 'Day Tripper', 'Piggies', 'Revolution'] hair black
If we had written
print(mark.lastName)after line 34, we’d have to change it to
print(getattr(mark, attributeName))