Support comments as part of multiline values in configparser

[dabuahoid]

Feature or enhancement

Proposal:

Today's configparser module does not consider lines being properly indented and starting with a comment_prefixes to be part of a multiline value. This can lead to unintended data loss during reading an INI file as:
config.ini

# I am a comment
    # I am also a comment
[dummy]
# I am a comment
key = I am a multiline
# I am a comment
    # multiline part which is not considered but should be
    being again part
# I am a comment

test.py

import configparser
config = configparser.ConfigParser(comment_prefixes='#', interpolation=None, inline_comment_prefixes=None)
config.read('config.ini')
print(config['dummy']['key'])
# output: 'I am a multiline\nbeing again part'
# expected output:  'I am a multiline\n# multiline part which is not considered but should be\nbeing again part''

Therefore, I would propose to add a new parameter allow_comment_in_value (default False) in the ConfigParser class to support such scenarios. Please note that in this scenario the comment_prefixes is set to '#' and any interpolation is disabled!

Thanks and best regards!

Has this already been discussed elsewhere?

No response given

Links to previous discussion of this feature:

No response

[donBarbos]

@dabuahoid Thank you for opening new issue, but could you provide use case like this:
config.ini:

[dummy]
key = i am a multiline
# value which is not considered
  and part of multiline

test.py:

import configparser
cofig = configparser.ConfigParser()
config.read('config.ini')
print(config['dummy']['key']) # I got 'i am a multiline\nand part of multiline' as expected

And as far as i know we can pass comment_prefixes or inline_comment_prefixes arguments to change the behavior in such scenarios

[dabuahoid]

Hi @donBarbos have updated the description accordingly.

[donBarbos]

cc @ambv as configparser expert, what you think about it?

EDIT: if we decide to add new flag I would like to implement this feature