Python
Python Datatypes
Python supports Dynamic Type Bindings. That means, - we can declare variable with one datatype and change type of value on next line.
Example, -
myString = 10
print("Number value from myString variable - " + myString)
myString = "Ten"
print("String value from myString variable - " + Ten)
Pros of Dynamic Type Bindings:
- Very easy to work with
- Faster development time.
Cons of Dynamic Type Bindings:
- May result in bugs for unexcepted data types!
- you need to be aware of type()
Available data types:
Data type | Python syntax | Description |
---|---|---|
Integer | int | Whole number Example, - 10 100 300 |
Floating Point | float | Number with decimal points Example, - 11.50 100.02 |
String | str | Ordered sequence of characters. Example, - “Hello” ‘Hi’ “200”. Able to represent in single quote OR in double quote |
List | list | Ordered sequence of Objects. Example, - [10, ‘Hello’, 20] |
Dictionaries | dict | Unordered key-value paired. Example,- {“key1”:”value1”, “key2”:”value2”} |
Tuples | tup | Ordered immutable sequence of objects: {“hey”, 200, “Hello”}. Note: Once tupple is create, we can’t change its sequence at runtime. |
Sets | set | Unordered collection of unique objects: {“A”, “F”, “G”} |
Booleans | bool | logical value indicating true OR false |
NoneType | None | void object. |
Few datatype with explaination & examples:
Notes:
- type() method use to identfy datatype of given object.
Example,-
Input => my_var = 12 type(my_var) output => int
- ordereddict dictionary maintain key order also.
Back
python-documentation is maintained by ravaan-techky.