Fix CLN service startup failure by trimming trailing spaces in config parameters

[maxrantil]

fixes #7132

I would really appreciate some friendly feedback on this, as it is my first PR to the project and I might have misunderstood or missed important details.

I did find a similar function strip_trailing_whitespace in plugins/bcli.c, but from what I understand, having void trim_trailing_spaces(char *str); in ccan would still be beneficial, right?

[vincenzopalazzo]

Thanks for this, looks good to me

I was wondering if it is possible move this trim function inside tal string https://github.com/rustyrussell/ccan/blob/master/ccan/tal/str/str.h

[maxrantil]

I was wondering if it is possible move this trim function inside tal string https://github.com/rustyrussell/ccan/blob/master/ccan/tal/str/str.h

Do you mean to also change the function to use tal for allocating memory for the lines then? I thought it was not necessary since they were already allocated, but I can of course fix that if it is desirable. @vincenzopalazzo

[vincenzopalazzo]

Do you mean to also change the function to use tal for allocating memory for the lines then?

No I mean to have like tal_strtrim or strtrim (not in tal), at the end of the day your code is implementing a subset of functionality to trim an origin string.

[maxrantil]

Pardon me, but I don't fully understand what you mean. Could you elaborate on your suggestion? To my understanding, the tal functions you are linking to, @vincenzopalazzo , relate to string functions involving memory allocation. If I were to implement tal_strtrim(), from my experience, those implementations remove spaces from the beginning and end and allocate a new string. Is that what you are aiming for here? The current placement of the function trim_trailing_spaces() in ccan/ccan/str/str.c made sense to me because it manipulates an already allocated string.
Like I said at the beginning, I am new here and would appreciate a more thorough explanation if you have time for that.

[endothermicdev]

Nice PR! It's a neat solution to the problem. I agree with Vincenzo - so long as a ccan library is being modified, it would be nice to go ahead and trim the leading whitespace too and resize the buffer (which would make this a tal/str operation). It's not a significant difference in this case, but it could be more useful for the library. I'm okay with either approach.

[maxrantil]

Thanks for the comment @endothermicdev, I'll start doing a strtrim function in tal tomorrow if you both think that is more useful then 👍

[maxrantil]

@vincenzopalazzo & @endothermicdev , I've now changed to using tal_strtrim, is this what you had in mind?

[maxrantil]

@endothermicdev I am having some issues with tests on my local machine using tal_resize like this.
Is there anything in the way I am using it now, that you can see, is not working like it should? I get some out of memory errors from within the tal_resize() on my machine but the tests seems to run fine here in github-actions...
I am working on writing a test for it that I will push with the PR.

[maxrantil]

The local test I have tried is to add a space after a config varible in the startup_regtest.sh file and then sourced it and start_ln 2.

[maxrantil]

If someone could have a look at the pytest and tell me if something is wrong in it, that would be appreciated.