Python built-in sequence types overview
List deduction and builder expressions
List inference is a shortcut to building a list, and builder expressions can be used to create sequences of any other type
Example of a list deduction
for in range (3)]>>>test[0,2,4]
The general principle of using list derivation is that you can only list derivations to create new lists and keep them as short as possible.
There may be a variable leakage problem in list derivation in Python2, and Python3 solves the problem of variable leakage
Comparison of list derivation and Map/filter
Builder expression
Although you can also use list inference to initialize tuples, arrays, or other sequence types, a builder expression is a better choice. Because the iterator protocol is followed by the generator expression, you can produce elements individually, rather than creating a complete list, and then passing the list to a constructor. More Memory savings
Initializing tuples and arrays with generator expressions
The use of tuples
Tuples can also be used for records with no field names, in addition to being used as immutable lists
Tuples and records
Tuples are actually records of the data, and each tuple in the tuple holds the data for a field in the record, plus the location of the field.
Tuple unpacking
Tuple unpacking can be used * to help us focus on some elements of the tuple
Tuple unpacking can also be applied in nested structures
Named tuples
Collections.namedtuple is a factory function. Can be used to build a tuple with a field name and a class with a name
In addition to attributes that are integrated from a common tuple, named tuples have their own proprietary properties. _fields,_make,_asdict ()
Tuples for immutable lists
In addition to adding or subtracting elements related methods, tuples support all other methods of the list.
The methods and properties of a list or tuple
Object slices
S[A:B:C] for s between A and B with C as the interval value, C can be negative, reverse value
Multidimensional slices and ellipsis
Slice Assignment
Use + and * for sequences
During stitching, two sequences that are manipulated are not modified, and Python creates a new sequence that contains the same type of data as the result of stitching.
+ and * Follow this rule, not modify the original operand, but build a new sequence
Increment assignment of a sequence
The Special method behind + = is __iadd__ But if a class does not implement the method, Python calls __add__
The special method behind *= is that __imul__ is similar to + =
The lessons that have been learned
Do not put mutable objects in tuples
Incremental assignment is not an atomic operation
It's not hard to see Python bytecode, he's very helpful in understanding the mechanics behind the code.
List.sort methods and built-in functions sorted
List.sort The local sort, sorts the original list and returns none. Direct manipulation of the original list
Sorted is not, a new list is created, and the list is sorted in the original list.
Reverse parameter, true in descending order, false by default
Key parameter used to compare the keywords
Bisect managing sorted sequences
The Bisect module contains two main functions, bisect,insort, and two functions use a binary lookup algorithm to find or insert elements in an ordered sequence
Bisect.insort (Seq,item) Inserts the item into the SEQ and keeps the seq sorted in ascending order
Array
If we need a list that contains only numbers, then Array.array is more efficient than list, and arrays support all operations related to mutable sequences. including Pop,insert and extend. In addition, the array also provides a faster way to read and save files from a file frombytes and ToFile
Properties and methods for lists and arrays
Memory View Memoryview
SciPy and NumPy
bidirectional queues and other forms of queuing
The Collections.deque class is a thread-safe data type that can quickly add or remove elements from both ends.
Methods for list and bidirectional queues
Fluent Python the second chapter sequence consists of an array of learning records