Are regex flags supported?

[spacether]

Is your feature request related to a problem? Please describe. Are regex flags supported? I would like them to be used when generating samples.

Per:

• https://www.ecma-international.org/ecma-262/5.1/#sec-15.10.7.2
• https://docs.python.org/3/library/re.html#re.compile Flags can be global, ignore_case and multi_line

Describe the solution you'd like One where we can pass in the flags in the regex input string.

Describe alternatives you've considered N/A

Additional context This came up when reviewing https://github.com/OpenAPITools/openapi-generator/pull/8054#pullrequestreview-555587102

[curious-odd-man]

Hello @spacether .

At the moment flags are not supported.

The ignore_case (case insensitive generation) can be turned on using configuration object and this option.

The generated values are always multiline - you can specify ^ and $ character to mark start/end of string and \n or \r\n or \r of new line.

As for global flag - I don't see how it fits in generation of text by pattern. Please let me know if you want some specific behavior with/without global flag.

In general I will implement parsing of the ECMA format and parsing flags, though as far as I see I will ignore g and m flags. And i flag will set the parameter for case insensitive generation.

This will be done as part of #56 issue.

Please add details/examples if you expect some specific behavior (different behavior) for g and m flags or close the ticket if you agree with what I wrote above.

[spacether]

Thank you for explaining that. Glad to see that we can use the ignore_case option. I agree that global has no impact so it makes sense to omit it. It looks like when not in multiline mode then ^ will only match the beginning of the passed in string and $ will only match the end of the string. When in multiline mode ^ will additionally match right after all newlines and $ will additionally match right before newlines. Do you agree with that interpretation? Could a multiline True/False option be added to your Configuration?

Here is a sample of what could be generated with MULTILINE=True/False pattern = '^a'

MULTILINE=False

• generated_examples = ['a']

MULTILINE=TRUE

• generated_examples = ['a', '\na', '\n\na']

[curious-odd-man]

Another issue is created for multiline generation - #58 . Case insensitive generation is supported by configuration option and will be supported with flags when #56 is done.

[spacether]

Thank you!