Possible RFE: allow spaces in pathnames?

[ryran]

Obviously this isn't needed for targetcli itself, but I think it would be one little checkbox to increase the appeal of configshell as a general-use library ....

I haven't taken a crack at it yet -- before I do, could someone (ahem) comment on how feasible it would be? Is anyone violently opposed to this? (I.e., it would mean interpreting quotes and backslashes.)

[agrover]

Sure, this would be very nice to have. I think it might be a straightforward change to our pyparsing-based grammar. @cvubrugier ?

[cvubrugier]

Yes, we should support spaces in paths.

I see that pyparsing has a QuotedString class. I think that supporting backslashes may be a little bit more difficult though.

@ryan: a while ago, I wrote a test script when I replaced simpleparse with pyparsing to ensure that the new pyparsing-based parser worked like the old simpleparse-based parser:

https://gist.github.com/cvubrugier/6522643

Maybe you can use a similar approach to ensure that your modified pyparsing grammar does not introduce a regression. You can also add test cases with paths that contain spaces.