I had a look on the toml library source code and I think I understand that this encoder should format dictionnaries in one line instead of classic dictionnary toml format.
My expected behavior is the following :
<code>import toml
data = {'a': 1, 'b': 2, 'c': 3}
# Default Encoder
toml_string = toml.dumps(data)
print(toml_string)
# Output:
# a = 1
# b = 2
# c = 3
# TomlPreserveInlineDictEncoder
toml_string = toml.dumps(data, encoder=toml.TomlPreserveInlineDictEncoder())
print(toml_string)
# Output:
# {a = 1, b = 2, c = 3}
</code>
<code>import toml
data = {'a': 1, 'b': 2, 'c': 3}
# Default Encoder
toml_string = toml.dumps(data)
print(toml_string)
# Output:
# a = 1
# b = 2
# c = 3
# TomlPreserveInlineDictEncoder
toml_string = toml.dumps(data, encoder=toml.TomlPreserveInlineDictEncoder())
print(toml_string)
# Output:
# {a = 1, b = 2, c = 3}
</code>
import toml
data = {'a': 1, 'b': 2, 'c': 3}
# Default Encoder
toml_string = toml.dumps(data)
print(toml_string)
# Output:
# a = 1
# b = 2
# c = 3
# TomlPreserveInlineDictEncoder
toml_string = toml.dumps(data, encoder=toml.TomlPreserveInlineDictEncoder())
print(toml_string)
# Output:
# {a = 1, b = 2, c = 3}
But, in fact, changing encoder doesn’t change anything to the dumped string. (I also tried with nested dictionnaries)
Did I miss something ? Is this functionnality suppose to work ?
(I tested it in Python 3.8 and 3.10)
toml.TomlPreserveInlineDictEncoder is meant to preserve the formatting of dicts that are originally formatted inline when parsed by toml‘s own decoder.
For example:
<code>import toml
data = toml.loads('''
A = {a = 1, b = 2, c = 3}
B = {a = 1, b = 2, c = 3}
''')
toml_string = toml.dumps(data, encoder=toml.TomlPreserveInlineDictEncoder())
print(toml_string)
# Output:
# A = { a = 1, b = 2, c = 3 }
# B = { a = 1, b = 2, c = 3 }
</code>
<code>import toml
data = toml.loads('''
A = {a = 1, b = 2, c = 3}
B = {a = 1, b = 2, c = 3}
''')
toml_string = toml.dumps(data, encoder=toml.TomlPreserveInlineDictEncoder())
print(toml_string)
# Output:
# A = { a = 1, b = 2, c = 3 }
# B = { a = 1, b = 2, c = 3 }
</code>
import toml
data = toml.loads('''
A = {a = 1, b = 2, c = 3}
B = {a = 1, b = 2, c = 3}
''')
toml_string = toml.dumps(data, encoder=toml.TomlPreserveInlineDictEncoder())
print(toml_string)
# Output:
# A = { a = 1, b = 2, c = 3 }
# B = { a = 1, b = 2, c = 3 }
Since toml‘s encoder determines if a dict should be formatted inline by checking if the dict is an instance of toml.decoder.InlineTableDict, you can manually format a section of the dict as inline with a subclass of dict and InlineTableDict:
<code>import toml
class InlineDict(dict, toml.decoder.InlineTableDict):
pass
data = {'A': InlineDict({'a': 1, 'b': 2, 'c': 3})}
toml_string = toml.dumps(data, encoder=toml.TomlPreserveInlineDictEncoder())
print(toml_string)
# Output:
# A = { a = 1, b = 2, c = 3 }
</code>
<code>import toml
class InlineDict(dict, toml.decoder.InlineTableDict):
pass
data = {'A': InlineDict({'a': 1, 'b': 2, 'c': 3})}
toml_string = toml.dumps(data, encoder=toml.TomlPreserveInlineDictEncoder())
print(toml_string)
# Output:
# A = { a = 1, b = 2, c = 3 }
</code>
import toml
class InlineDict(dict, toml.decoder.InlineTableDict):
pass
data = {'A': InlineDict({'a': 1, 'b': 2, 'c': 3})}
toml_string = toml.dumps(data, encoder=toml.TomlPreserveInlineDictEncoder())
print(toml_string)
# Output:
# A = { a = 1, b = 2, c = 3 }
2