Read and dump [bracket, list] from and to yaml with python

If you want to load, then dump (maybe after modifying some values), PyYAML is not the right tool, as it will mangle many things in the syntactic representation.

It will drop flow style as you noticed, but also drop comments, anchor/alias names, specific integer formats (octal, hex, binary), etc.

There is little control over the flow- vs block-style of the output in PyYAML. You can have all-block; node collections as flow,
and you can have all-flow, using the default_flow_style parameter to safe_dump().

You'd be better of using ruamel.yaml (disclaimer: I am the author of that library), as it supports the now 10 year old YAML 1.2 standard (where PyYAML only handles the outdated YAML 1.1) and will get you output which is much closer, and often identical to your YAML input.

from ruamel.yaml import YAML

yaml = YAML()
with open(system_bsc_path) as f:
    system_bsc_dict = yaml.load(f)
with open(system_bsc_path, "w") as f:
    yaml.dump(system_bsc_dict, f)

If you are Python 3, you can use:

from pathlib import Path
yaml_file = Path(system_bsc_path)
system_bsc_dict = yaml.load(yaml_file)
yaml.dump(system_bsc_dict, yaml_file)

By default any new lists (and dicts) will be block style, if you want to add a flow-style list, then you can use yaml.default_flow_style = True to set all those lists, or using fine control by setting the flow attribute (.fa) on the special internal representation:

def FSlist(l):  # concert list into flow-style (default is block style)
    from ruamel.yaml.comments import CommentedSeq
    cs = CommentedSeq(l)
    cs.fa.set_flow_style()
    return cs

system_bsc_dict['existing_field'] = FSlist(["Boston Maestro 4000"])

As mentioned here, using default_flow_style should do the trick.

But seems that you actually need it set to True.

I.e.

>>> d = {'chs_per_cath': [[[10, 11, 12, 13], [13000, 13100, 13200, 13300]],
  [[16, 17, 18, 19, 20, 21, 22, 23, 24, 25],
   [13400, 13500, 13600, 13700, 13800, 13900, 14000, 14100, 14200, 14300]],
  [[32, 33, 34, 35, 36, 37, 38, 39, 40, 41],
   [13400, 13500, 13600, 13700, 13800, 13900, 14000, 14100, 14200, 14300]]]}

>>> print(yaml.dump(d, default_flow_style=True))
{chs_per_cath: [[[10, 11, 12, 13], [13000, 13100, 13200, 13300]], [[16, 17, 18, 19,
        20, 21, 22, 23, 24, 25], [13400, 13500, 13600, 13700, 13800, 13900, 14000,
        14100, 14200, 14300]], [[32, 33, 34, 35, 36, 37, 38, 39, 40, 41], [13400,
        13500, 13600, 13700, 13800, 13900, 14000, 14100, 14200, 14300]]]}

Please take a look at default_flow_style argument for yaml.dump().

By default, PyYAML chooses the style of a collection depending on whether it has nested collections. If a collection has nested collections, it will be assigned the block style. Otherwise it will have the flow style.

If you want collections to be always serialized in the block style, set the parameter default_flow_style of dump() to False.

https://pyyaml.org/wiki/PyYAMLDocumentation#DumpingYAML