Parse yaml wihtout any assumptions

[awdm]

I'm using this software to automatically duplicate and modify yaml-files to setup Kubernetes resources. However, I noticed that these files are really inconsistent. Some examples:

• Sometimes things just need to be quoted, no matter the actual data type, e.g. when you are declaring environment variables for a new container. But there is no way to recognize this through the syntax, only through the semantics.
• In some tutorials, ports get their port number as a name. If people follow these tutorials they end up writing yaml which can't be parsed correctly

So what I'm looking for is a parser/serializer that makes absoluteley zero assumptions about the data type it is processing. If it reads a string containing a number, it gets parsed into a string. If it's told to write a string that contains a boolean, it writes a string (with qoutation marks).

[MariuszKogut]

I have the same problem with a docker-compose.yml.

Input file:

version: '3.7'
...

Output after deserialize and serialize to same file

version: 3.7
...

Version is no longer a string and the schema validation during calling docker-compose fails.

@awdm Did you found a workaround for your problem?

[EdwardCooke]

This issue was just recently fixed and went out with the v12 version about 5 days ago. When a string is passed in to the serializer, it will handle it and quote as needed (numbers/bools/etc).

I'm closing the issue for now. If you still see problems with it let me know and I'll get them resolved.