如何在Python的yaml文件中配置多个字典键？

Question

I am trying to use a yaml file to store dictionary keys so I can parse various json payloads.

I have the following yaml file loaded into a variable called 'cfg':

my_table: metadata

When I run:

for i in json['resources']:
    print(i[cfg['my_table']])

I can see all values from my json's ['metadata'] keys just fine.

However, I would like to see all values from the ['metadata']['guid'] keys that exist under ['resources']. So when I change my yaml file to:

my_table: metadata, guid

It doesn't work. I've also tried these variations:

my_table: ['metadata', 'guid']  #tried ''.join("['{0}']".format(a) for a in cfg['my_table']) which gives me what I want, but when I shove this inside i[], I get a key error
my_table: ['metadata']['guid']  #yaml file doesn't parse properly

Any ideas?

Answer 1

So you want to store dictionary keys in a yaml file?

Just do

# keys.yaml
my_table: [metadata, guid]

And then in your python file

keys = yaml.safe_load('keys.yaml')['my_table']

for obj in json['resources']:  # don't call it json, that's a package name
    print(obj[keys[0]][keys[1]])

For a variable number of keys (not only two) do it like so:

def recursive_getitem(obj, keys):
   if len(keys) == 0:
      return obj
   return recursive_getitem(obj[keys[0]], keys[1:])

and then do

for obj in json['resources']:  
    print(recursive_getitem(obj, keys))

---

To answer your second question in the comments, you need to use a nested list

# keys.yaml
my_table:
    - [metadata, guid]
    - [entity, name]

.

# py file
keys = yaml.safe_load('keys.yaml')['my_table']

for obj in json['resources']:  # don't call it json, that's a package name
    print(obj[keys[0][0]][keys[0][1]])
    print(obj[keys[1][0]][keys[1][1]])