On Thu, Apr 23, 2015 at 11:09 AM, David Nalesnik <[email protected]>
wrote:

>
>
> On Thu, Apr 23, 2015 at 11:07 AM, Carl Sorensen <[email protected]>
> wrote:
>
>>
>>
>> On 4/23/15 7:55 AM, "H. S. Teoh" <[email protected]> wrote:
>>
>> >On Thu, Apr 23, 2015 at 12:34:32PM +0000, Carl Sorensen wrote:
>> >>
>> >>
>> >> On 4/22/15 10:57 PM, "Super-User" <[email protected]> wrote:
>> >>
>> >> >Seheme, as a dialect of Lisp, can be annotated with ;; so that our
>> >> >Scheme codes implementing Jianpu can be documented as well.
>> >>
>> >>But the Internals Reference (IR) is automatically generated from the
>> >> C++ source code of the engravers.  And we don't have any automatic
>> >> tools to generate IR references from Scheme engravers.
>> >[...]
>> >
>> >It should be possible, right? At the very least, we could use a fixed
>> >format, say for the comment header in the .scm file of the engraver, and
>> >use a script to extract and format that into documentation.
>>
>> Yes, it's certainly possible.  And it can be done automatically (not from
>> comments) because the information is in Scheme structures with known
>> syntax.
>
>
I think there would need to be an enhancement of the mechanism for creating
Scheme engravers, though.  C++ engravers have an expression at their ends
such as the following (in lily/slur-engraver.cc):

ADD_TRANSLATOR (Slur_engraver,
                /* doc */
                "Build slur grobs from slur events.",

                /* create */
                "Slur ",

                /* read */
                "slurMelismaBusy "
                "doubleSlurs ",

                /* write */
                ""
               );

Checking the documentation for Slur_engraver shows how this is used.  See:

http://www.lilypond.org/doc/v2.18/Documentation/internals/slur_005fengraver

--David
_______________________________________________
lilypond-user mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/lilypond-user

Reply via email to