Re: Internal visibility

[ludo@gnu.org (Ludovic Courtès)]

Hi,

"Neil Jerram" <neiljerram@googlemail.com> writes:

> 2008/6/1 Ludovic Courtès <ludo@gnu.org>:
>>
>> It's just that it's convenient and equivalent to 1.6's `SCM_STRING_CHARS ()',
>> so people have come to use it...  According to Google's codesearch,
>> `scm_i_string_chars ()' is used by Mailutils, Lilypond, AutoGen, SND and
>> a few others.
>
> IMO that amounts to a strong case for making it an official API.  I
> believe the argument against doing that is something to do with future
> multi- and variable-byte string encodings - but perhaps we can handle
> that as a transitional issue once those encodings are in place?

I did not mark them as internal so as to leave them in this
"semi-official" state.  But it was always clear from 1.8.0 that the `i'
in `scm_i_' means "internal" and that applications could only use it at
their own risk, so I wouldn't want to go as far as documenting or
encouraging it.

As Clinton said, there were good reasons for these functions to be
internal in the first place, and it seems more reasonable to keep it
this way.  We *will* support Unicode eventually, right?  :-)

Thanks,
Ludovic.