"G. Milde" <[EMAIL PROTECTED]> writes:
| On 9.06.05, Lars Gullik Bj�nnes wrote:
>> [EMAIL PROTECTED] writes:
>>
>> | On Thu, 9 Jun 2005, G. Milde wrote:
>> >
>> >> On 8.06.05, Jean-Marc Lasgouttes wrote:
>> >> > >>>>> "G" == G Milde <[EMAIL PROTECTED]> writes:
>> >>
>> >> > G> ... all the lfuns should have some usage doc (telling the
>> >> > G> expected arguments).
>
>> | Going off on a slight tangent, what if we put some cursory documentation
>> | of the lfuns in the *source* code using wiki markup?
>
>> | Especially since we'd avoid maintaining separate documentation of the
>> | lfuns, and anything that makes it easier for the developers to document
>> | changes to the lfun ought to be a good thing...
>>
>> You made it harder... now we have to know wiki markup.
>
| Actually, we need only basic data, markup can be added by the script.
>
| Currently, in LyXAction.C are lines like
>
| { LFUN_WRITEAS, "buffer-write-as", ReadOnly },
| { LFUN_CANCEL, "cancel", NoBuffer },
| { LFUN_INSET_CAPTION, "caption-insert", Noop },
| { LFUN_LEFT, "char-backward", ReadOnly | NoUpdate},
Yeah... I don't know why we removed the cocumantaion from the array...
| If these lines where completed by comments as e.g.
ugly bugly *smile*
| // { LFUN, "public name", flags }, // (argument(s)), Synopsis (optional)
>
| { LFUN_WRITEAS, "buffer-write-as", ReadOnly }, // (filename), write buffer
to new file
| { LFUN_CANCEL, "cancel", NoBuffer }, // (), cancel action
| { LFUN_INSET_CAPTION, "caption-insert", Noop },
| { LFUN_LEFT, "char-backward", ReadOnly | NoUpdate}, // ()
No... we really should encode this information in the lfun definition.
IMHO both a short textual description, and what arguments to expect
and their type.
If we have this information attached to the lfun a doc tool could
extract it easily. but just adding comments... no not in favour.
And all of this is 1.5 stuff anyway.
--
Lgb
