没有新行的Yaml转储

Question

I am not sure if rules for Yaml dictates new line as a must have, But I am having an issue with the new line characters in the Yaml serialized string. Here is what I am trying to do:

yaml.dump(my_python_dict)

Returns:

"---\nasset_data:\n- {assetDefaultScenePath: 'C:\\\\data\\\\my_newfile.ma',\n  assetName: new_asset}\n- {assetDefaultScenePath: 'C:\\\\assetA.ma',\n  assetName: assetA.ma}\naudio_dir: ''\nbg_dir: ''\ndestination: ''\nepisode: '00'\nprod: test\n"

The issue is the YAML string is generated from Ruby and is passed through an intermediate scripting language (MEL) before it reaches Python. The intermediate language cannot understand Yaml and computes all the "\\n" and fails. I was wondering if there is a way to dump Yaml without a newline character? I am guessing that would cause issues at load. I am currently out of idea's and any recommendation would be helpful.

Here is the detail on how things are tied up. The ruby code will build a Windows Command line and execute it.

"Mayabatch" -command "source \"C:/cygwin/home/workspace/bpt-maya/src/bd_lay/generate.mel\";generate(\”{SERIALIZED_DICT}\")"

1. SERIALIZED DICT is where the YAML data needs to be added.
2. generate.mel has a python call with SERIALIZED DICT as argument
3. The Mayabatch does not support direct Python execution thats why it needs to be wrapped. JSON will fail here due to the double quotes that's why I need some other serialization.

Answer 1

YAML is a super-set of JSON. So, a valid JSON is also a valid YAML (but valid YAML isn't necessarily valid JSON).

This fact may allow one to get around MEL's YAML-parsing limitations

>>> import yaml
>>> import json

>>> orig_yaml_txt = "---\nasset_data:\n- {assetDefaultScenePath: 'C:\\\\data\\\\my_newfile.ma',\n  assetName: new_asset}\n- {assetDefaultScenePath: 'C:\\\\assetA.ma',\n  assetName: assetA.ma}\naudio_dir: ''\nbg_dir: ''\ndestination: ''\nepisode: '00'\nprod: test\n"
>>> dict_1 = yaml.load(orig_yaml_txt)
>>> intermediate_json_txt = json.dumps(dict_1)
>>> dict_2 = yaml.load(intermediate_json_txt)
>>> dict_1 == dict_2
True

In the above snippet, I take your YAML-serialized text, load it into a python dictionary, then dump that to a json-serialized text, and then load it into a python dictionary using YAML to parse.

In your question, you state that the original YAML is generated from Ruby. Perhaps, instead of dumping YAML, you could dump JSON from Ruby, or add another step in the pipeline that converts the YAML to JSON before being fed to MEL.