摘要：這種數(shù)據(jù)結(jié)構(gòu)包含以下幾種常見的操作向關(guān)聯(lián)數(shù)組添加鍵值對(duì)從關(guān)聯(lián)數(shù)組內(nèi)刪除鍵值對(duì)修改關(guān)聯(lián)數(shù)組內(nèi)的鍵值對(duì)根據(jù)已知的鍵尋找值字典問題是設(shè)計(jì)一種能夠具備關(guān)聯(lián)數(shù)組特性的數(shù)據(jù)結(jié)構(gòu)。

Python中有一個(gè)叫作dictionary的對(duì)象類型，翻譯過來就是“字典”，用dict表示。

>>> mydict = {}
>>> mydict
{}
>>> type(mydict)

>>> person={"name":"liu","sex":"male","age":10}
>>> person
{"age": 10, "name": "liu", "sex": "male"}

字典dict是以鍵值對(duì)的形式存儲(chǔ)數(shù)據(jù)。比如"name":"liu"前面的name叫作鍵（key），后面的liu是前面的鍵所對(duì)應(yīng)的值（value）。在一個(gè)字典中，鍵是唯一的，不能重復(fù)。值則對(duì)應(yīng)于鍵，值可以重復(fù)。

鍵值之間用冒號(hào)隔開，每一對(duì)鍵值之間用逗號(hào)隔開。

>>> person
{"age": 10, "name": "liu", "sex": "male"}
>>> person["hobby"]="reading" #增加鍵值對(duì)的方法
>>> person
{"hobby": "reading", "age": 10, "name": "liu", "sex": "male"}

>>> person["age"]=20
>>> person
{"hobby": "reading", "age": 20, "name": "liu", "sex": "male"}
>>> id(person)
49368784
>>> person["age"]=30
>>> person
{"hobby": "reading", "age": 30, "name": "liu", "sex": "male"}
>>> id(person)
49368784

結(jié)果可以看出，字典可以原地地修改，即它是可變的，并且不會(huì)創(chuàng)建新的對(duì)象。

>>> name = (["first","Google"],["second","Yahoo"])
>>> website=dict(name)
>>> website
{"second": "Yahoo", "first": "Google"}
>>> type(website)


或者這樣：

>>> ad = dict(name="liu",age= 26)
>>> ad
{"age": 26, "name": "liu"}
>>> type(ad)

>>> website={}.fromkeys(("third","forth"),("facebook","amazon"))
>>> website
{"forth": ("facebook", "amazon"), "third": ("facebook", "amazon")}
>>> website={}.fromkeys(("third","forth"),"facebook")
>>> website
{"forth": "facebook", "third": "facebook"}

特別注意的是，字典中的鍵必須是不可變對(duì)象，值可以是任意類的對(duì)象。

>>> dd = {(1,2):1}
>>> dd
{(1, 2): 1}
>>> type(dd)


>>> dd={[1,2]:1}

Traceback (most recent call last):
  File "", line 1, in 
    dd={[1,2]:1}
TypeError: unhashable type: "list"

字典類型的對(duì)象是以鍵值對(duì)的形式存儲(chǔ)數(shù)據(jù)的，所以只要知道鍵就能得到值，這在本質(zhì)上就是一種映射關(guān)系。

>>> person
{"hobby": "reading", "age": 30, "name": "liu", "sex": "male"}
>>> person["age"]
30
>>> person["hobby"]
"reading"

返回字典d中鍵值對(duì)的數(shù)量

>>> city_code = {"beijing":"010","shanghai":"021","guangzhou":"020"}
>>> len(city_code)
3

返回字典d中的鍵key的值

>>> city_code
{"beijing": "010", "shanghai": "021", "guangzhou": "020"}
>>> city_code["beijing"]
"010"

將值賦給字典d中的鍵key

>>> city_code
{"beijing": "010", "shanghai": "021", "guangzhou": "020"}
>>> city_code["beijing"]="01110"
>>> city_code
{"beijing": "01110", "shanghai": "021", "guangzhou": "020"}

刪除字典d中的鍵key值對(duì)

>>> city_code
{"beijing": "01110", "shanghai": "021", "guangzhou": "020"}
>>> del city_code["beijing"]
>>> city_code
{"shanghai": "021", "guangzhou": "020"}

檢查字典d中是否含有鍵為key的項(xiàng)

>>> city_code
{"shanghai": "021", "guangzhou": "020"}
>>> "shanghai" in city_code
True
>>> "beijing" in city_code
False

>>> city_code
{"shanghai": "021", "guangzhou": "020"}
>>> "Shanghai is a wonderful city,its area code is %(shanghai)s" % city_code
"Shanghai is a wonderful city,its area code is 021"

在計(jì)算機(jī)科學(xué)中，關(guān)聯(lián)數(shù)組又稱為映射（Map）、字典（Dictionary），是一個(gè)抽象的數(shù)據(jù)結(jié)構(gòu)，它包含著類似于鍵值的有序?qū)Α＿@種數(shù)據(jù)結(jié)構(gòu)包含以下幾種常見的操作：

向關(guān)聯(lián)數(shù)組添加鍵值對(duì)

從關(guān)聯(lián)數(shù)組內(nèi)刪除鍵值對(duì)

修改關(guān)聯(lián)數(shù)組內(nèi)的鍵值對(duì)

根據(jù)已知的鍵尋找值

字典問題是設(shè)計(jì)一種能夠具備關(guān)聯(lián)數(shù)組特性的數(shù)據(jù)結(jié)構(gòu)。解決字典問題的常用方法是散列表，但有些情況也可以直接使用有地址的數(shù)組、二叉樹，或者其他結(jié)構(gòu)。

散列表（hash table，也叫哈希表），是根據(jù)關(guān)鍵字而直接訪問在內(nèi)存存儲(chǔ)位置的數(shù)據(jù)結(jié)構(gòu)。即把鍵值通過一個(gè)函數(shù)的計(jì)算，映射到表中的一個(gè)位置來訪問記錄，加快了查找速度。這個(gè)映射函數(shù)稱為散列函數(shù)，存放記錄的數(shù)組稱為散列表。

淺拷貝：Python在所執(zhí)行的復(fù)制動(dòng)作中，如果是基本類型的數(shù)據(jù)，就在內(nèi)存中新建一個(gè)地址存儲(chǔ)，如果不是基本類型，就不會(huì)新建一個(gè)地址，而是用標(biāo)簽引用原來的對(duì)象。copy()實(shí)現(xiàn)的是淺拷貝

深拷貝：無論是基本數(shù)據(jù)類型還是其他類型都會(huì)新建一個(gè)地址來存儲(chǔ)。deepcopy()實(shí)現(xiàn)的是深拷貝

>>> ad={"name":"liu","age":20}
>>> id(ad)
49368640
>>> cd= ad.copy()
>>> cd
{"age": 20, "name": "liu"}
>>> id(cd)  #與ad是不同的對(duì)象
49225776
>>> cd["age"]=40  #修改cd沒有對(duì)ad造成影響
>>> cd
{"age": 40, "name": "liu"}
>>> ad
{"age": 20, "name": "liu"}

>>> x={"name":"liu","lang":["python","java","c"]}
>>> y = x.copy()
>>> y
{"lang": ["python", "java", "c"], "name": "liu"}
>>> id(x)
49226496
>>> id(y)
49228800

##y是從x拷貝過來的，兩個(gè)在內(nèi)存中是不同的對(duì)象

>>> y["lang"].remove("c")
>>> y
{"lang": ["python", "java"], "name": "liu"}
>>> x
{"lang": ["python", "java"], "name": "liu"}
>>> id(x)
49226496
>>> id(y)
49228800

y中刪除元素“c“后，x的鍵lang的值也發(fā)生了變化

>>> id(x["lang"])
49607424
>>> id(y["lang"])
49607424

x與y中的列表是同一個(gè)對(duì)象，但是作為字符串的那個(gè)鍵值對(duì)屬于不同的對(duì)象。

>>> x
{"lang": ["python", "java"], "name": "liu"}
>>> import copy #導(dǎo)入一個(gè)模塊
>>> z = copy.deepcopy(x) #深拷貝
>>> z
{"lang": ["python", "java"], "name": "liu"}
>>> id(x["lang"])
49607424
>>> id(z["lang"])
49607184

#此時(shí)x和z中的列表不屬于同個(gè)對(duì)象

>>>
>>> x["lang"].remove("java")
>>> x
{"lang": ["python"], "name": "liu"}
>>> z
{"lang": ["python", "java"], "name": "liu"}

#此時(shí)，修改一個(gè)列表的值，就不會(huì)影響另外一個(gè)列表的值了。

clear的作用是將字典清空，得到的是空字典。del是將字典刪除，內(nèi)存中就沒有它了。

>>> x
{"lang": ["python"], "name": "liu"}
>>> x.clear()
>>> x
{}
>>> del a

>>> del x
>>> x  #x已刪除

Traceback (most recent call last):
  File "", line 1, in 
    x
NameError: name "x" is not defined
>>>

如果要清空一個(gè)字典，還能夠使用x = {}這種方法，但這種方法的本質(zhì)是將變量a的引用轉(zhuǎn)向了{(lán)}這個(gè)對(duì)象，那么原來的對(duì)象呢？原來的對(duì)象稱為了斷了線的風(fēng)箏，這樣的東西稱為在Python中稱為垃圾，而且Python能夠自動(dòng)將這樣的垃圾回收。

得到字典中的某個(gè)值

>>> z
{"lang": ["python", "java"], "name": "liu"}
>>> z["lang"].remove("java")
>>> z
{"lang": ["python"], "name": "liu"}
>>> 
>>> z.get("name")
"liu"

>>> print z.get("age") #返回None
None
>>> d["age"]  #拋出異常
Traceback (most recent call last):
  File "", line 1, in 
    d["age"]
NameError: name "d" is not defined

>>> z.get("age",30) #如果age不存在，則返回默認(rèn)值30
30

dict.get()和dict[key]的區(qū)別在于：如果鍵不在字典中，前者返回None，后者拋出異常

>>> z
{"lang": ["python"], "name": "liu"}
>>> z.setdefault("age",30)  #如果age不存在，則返回默認(rèn)值30，同時(shí)將這對(duì)鍵值對(duì)添加到字典中
30
>>> z
{"lang": ["python"], "age": 30, "name": "liu"}

>>> z.setdefault("web")
>>> z
{"lang": ["python"], "web": None, "age": 30, "name": "liu"}

>>> dd = {"name":"liu","age":40,"websit":"www.liuguoquan.com"}
>>> dd_kv = dd.items()
>>> dd_kv
[("websit", "www.liuguoquan.com"), ("age", 40), ("name", "liu")]
>>> type(dd_kv)

>>> list(dd_kv)
[("websit", "www.liuguoquan.com"), ("age", 40), ("name", "liu")]

>>> dd_iter=dd.iteritems()
>>> type(dd_iter)

>>> dd_iter

>>> list(dd_iter)  #必須用list()迭代遍歷
[("websit", "www.liuguoquan.com"), ("age", 40), ("name", "liu")]

>>> dd
{"websit": "www.liuguoquan.com", "age": 40, "name": "liu"}
>>> dd.keys() #列出鍵的列表
["websit", "age", "name"]

>>> dd.values() #列出值的列表
["www.liuguoquan.com", 40, "liu"]

刪除字典鍵值對(duì)

>>> dd
{"websit": "www.liuguoquan.com", "age": 40, "name": "liu"}
>>> dd.pop("name") #刪除指定鍵
"liu"

>>> dd
{"websit": "www.liuguoquan.com", "age": 40}
>>> dd.pop() #參數(shù)不能省略

Traceback (most recent call last):
  File "", line 1, in 
    dd.pop()
TypeError: pop expected at least 1 arguments, got 0


>>> dd
{"websit": "www.liuguoquan.com", "age": 40}
>>> dd.popitem() #隨機(jī)刪除一個(gè)鍵值對(duì)
("websit", "www.liuguoquan.com")

>>> dd
{"age": 40}
>>> dd.popitem()
("age", 40)

>>> dd
{}

更新字典

>>> d1={"name":"liu"}
>>> d2={"age":20}
>>> d1.update(d2) #將字典到d2添加到d1中，更新字典
>>> d1
{"age": 20, "name": "liu"}
>>> d2
{"age": 20}

>>> d2.update([("name","lee"),("lang","python")]) #更新字典
>>> d2
{"lang": "python", "age": 20, "name": "lee"}

判斷字典中是否存在某個(gè)鍵

>>> d2
{"lang": "python", "age": 20, "name": "lee"}
>>> d2.has_key("name")
True
>>> d2.has_key("web")
False