How can I write data in YAML format in a file?

Question

I need to write the below data to yaml file using Python:

{"A": "a", "B": {"C": "c", "D": "d", "E": "e"}} 

i.e., dictionary in a dictionary. How can I achieve this?

Horseflesh

See also: How can I parse a YAML file in Python – Emblazon

Answer 1

import yaml

data = dict(
    A = 'a',
    B = dict(
        C = 'c',
        D = 'd',
        E = 'e',
    )
)

with open('data.yml', 'w') as outfile:
    yaml.dump(data, outfile, default_flow_style=False)

The default_flow_style=False parameter is necessary to produce the format you want (flow style), otherwise for nested collections it produces block style:

A: a
B: {C: c, D: d, E: e}

Agnail

default_flow_style=True does the opposite as stated in the answer below! – Weslee

Is there a benefit of using with here as opposed to the one-liner? – Fresco

@Fresco with ensures the file is closed properly when it is no longer needed – Regionalism

See Munch, #52571369 import yaml; from munch import munchify; f = munchify(yaml.safe_load(…));print(f.B.C) – Hennebery

Answer 2

Link to the PyYAML documentation showing the difference for the default_flow_style parameter. To write it to a file in block mode (often more readable):

d = {'A':'a', 'B':{'C':'c', 'D':'d', 'E':'e'}}
with open('result.yml', 'w') as yaml_file:
    yaml.dump(d, yaml_file, default_flow_style=False)

produces:

A: a
B:
  C: c
  D: d
  E: e

Balls

See Munch, #52571369 import yaml; from munch import munchify; f = munchify(yaml.safe_load(…));print(f.B.C) – Hennebery

Answer 3

As of 2024, I had to use sort_keys=False to maintain the specific order in the dictionary, i.e.:

import yaml

data = {
    "mode" : 'separate-outputs',
    "parent" : 'd93fca33ec026010b521b3587beb5897dff43189a8e1b4da078846c40d680dcei0',
    "postage" : 690,
    "inscriptions" : [
        {"file": "mango.avif", "metadata": {"title": "test"}},
        {"file": "mango2.avif", "metadata": {"title": "test"}},
    ]
}

with open('data.yml', 'w') as outfile:
    yaml.dump(data, outfile, default_flow_style=False, sort_keys=False)

data.yml

mode: separate-outputs
parent: d93fca55ec026010b521b3587beb5897dff43189a8e1b4da078846c40d680dcei0
postage: 690
inscriptions:
- file: mango.avif
  metadata:
    title: test
- file: mango2.avif
  metadata:
    title: test

Meit

Answer 4

At the time this question was asked in 2012, the default flow style was None but since PyYaml 5.1, it's been False, so for recent versions of PyYaml (e.g. PyYaml 6.0.1), the nested objects are written in block style by default, i.e. the following code:

import yaml
d = {'A': 'a', 'B': {'C': 'c', 'D': 'd', 'E': 'e'}}

with open('data.yaml', 'w') as f:
    yaml.dump(d, f)

would produce a file that looks like

A: a
B:
  C: c
  D: d
  E: e

Also, if you don't pass a file handle, dump() returns the data as a string, which can be written into a file as well.

s = yaml.dump(d)

with open("data.yaml", "w") as f:
    f.write(s)

which again would produce the same file as above.

---

default_flow_style=None lets you mix styles:

print(yaml.dump(d, default_flow_style=None))
                   
A: a
B: {C: c, D: d, E: e}

while default_flow_style=True is straight up flow style:

print(yaml.dump(d, default_flow_style=True))
                   
{A: a, B: {C: c, D: d, E: e}}

Also, if you want to see valid kwargs of dump but help(yaml.dump) is not all too helpful, try help(yaml.dump_all) instead because dump is just a wrapper for dump_all.

Lupine