python 递归实例(recursion in real world using python)
现有数据，结构如下：
recipes = {
'apple_pie': {
'ingredients': ['flour', 'sugar', 'eggs', 'shortening', 'apples', 'cinnamon'],
'cook_time': 60,
'prep_time': 30,
'ratings': {'john': 10, 'emily': 8, 'mom': 4}
},
'peanut_butter_cookies': {
'ingredients': ['flour', 'sugar', 'peanut_butter', 'butter', 'eggs'],
'cook_time': 45,
'prep_time': 20,
'allergens': ['peanuts']
},
'clam_chowder': {
'ingredients': ['flour', 'milk', 'clams', 'vegetable_stock', 'salt'],
'cook_time': 120,
'prep_time': 30,
'allergens': ['shellfish']
}
}
如果对Python中的基本数据类型--字典类型比较熟悉的话，可以很容易发现这是个“大字典”。而字典里面又包含字典，这样组织数据复合人类的逻辑，但是编程时候需要解析这些数据。比如我们需要解析成这样的数据：
{'apple_pie/cook_time': 60,
'apple_pie/ingredients': ['flour',
'sugar',
'eggs',
'shortening',
'apples',
'cinnamon'],
'apple_pie/prep_time': 30,
'apple_pie/ratings/emily': 8,
'apple_pie/ratings/john': 10,
'apple_pie/ratings/mom': 4,
'clam_chowder/allergens': ['shellfish'],
'clam_chowder/cook_time': 120,
'clam_chowder/ingredients': ['flour',
'milk',
'clams',
'vegetable_stock',
'salt'],
'clam_chowder/prep_time': 30,
'peanut_butter_cookies/allergens': ['peanuts'],
'peanut_butter_cookies/cook_time': 45,
'peanut_butter_cookies/ingredients': ['flour',
'sugar',
'peanut_butter',
'butter',
'eggs'],
'peanut_butter_cookies/prep_time': 20}仔细查看数据，发现字典里面不再有小字典。通俗来时就是把里面的大块数据（字典）打碎，然后碾平。
那么我们如何实现这样的解析功能呢？
看来用递归不错哟。
现在使用递归方法将上面的字典数据一层层地”碾平“
def flatten(obj):
if isinstance(obj, list):
# if object is a list object, flatten every sub e.ement
result = []
for elem in obj:
flat_elem = flatten(elem)
result.append(flat_elem)
elif isinstance(obj, dict):
# if object is a dictionary, flatten it on level
result = {}
for key, value in obj.items():
if isinstance(value, dict):
flat_value = flatten(value)
for subkey, subvalue in flat_value.items():
flat_key = '/'.join([key, subkey])
result[flat_key] = subvalue
else:
result[key] = value
else:
# otherwise just return the value
result = obj
return result
现在使用递归方法将上面的字典数据一层层地”碾平“
def flatten(obj):
if isinstance(obj, list):
# if object is a list object, flatten every sub e.ement
result = []
for elem in obj:
flat_elem = flatten(elem)
result.append(flat_elem)
elif isinstance(obj, dict):
# if object is a dictionary, flatten it on level
result = {}
for key, value in obj.items():
if isinstance(value, dict):
flat_value = flatten(value)
for subkey, subvalue in flat_value.items():
flat_key = '/'.join([key, subkey])
result[flat_key] = subvalue
else:
result[key] = value
else:
# otherwise just return the value
result = obj
return result
输出结果：