On 11/27/2022 1:05 PM, damien thiriet via ntg-context wrote:
> Hi,
> 
> 
> I am used to giving very extensive feedbacks to my pupils.
> These are buffers with xtables flushing lua vectors.
> Here is their basic structures:
> 
> * an environment that stores xtables. Turned into MWE, look like:
> 
>     \startenvironment env_corrige.mkiv
> 
>     \startluacode
>        userdata = userdata or {}
>     function userdata.Critere (commentaire,note,notemax)
>        context.startxcell {"width=11cm","height=2.75cm"}
>           context(Eleve[commentaire])
>        context.stopxcell ()
>        context.startxcell {"width=1.5cm","align=flushright"}
>           context(Eleve[note])
>           context("/")
>           context(notemax)
>        context.stopxcell ()
>     end
>     \stopluacode
> 
>     \startbuffer [grille]
>     \startxtable
>        \startxrow
>           \startxcell
>              \ctxlua {userdata.Critere ("structure","NoteStructure",2)}
>           \stopxcell
>        \stopxrow
>     \stopxtable
>     \stopbuffer
>     \stopenvironment
> 
> * The main file is divided in subjects, each one of them
> being the feedback to one pupil. In a MWE fashion
> 
>     \useenvironment [env_corrige]
>     \starttext
>     \startsubject[title={Astérix}]
>     \startluacode
>     Eleve = {
>        structure = "analyse très fine",
>        NoteStructure = 2,
>     }
>     \stopluacode
>     \getbuffer [grille]
>     \stopsubject
> 
>     \startsubject[title={Obélix}]
>     Eleve = {
>        structure = "le détail doit être travaillé. Pensez à la finesse 
> de vos menhirs…"
>        NoteStructure = 1,
>     }
>     \stopsubject
>     \getbuffer [grille]
>     \stoptext
> 
> Since whatever I fill in those files is written
> inside luacode environment, I tried to switch to
> cld documents.
> 
> Here is my main file in cld version
> 
>     context.useenvironment({"env_corrige"})
> 
>     context.starttext()
> 
>     context.startsubject({title="Asterix"})
> 
>     Eleve = {
>        structure = "analyse très fine",
>        NoteStructure = 2,
>     }
>     context.getbuffer({"ExoCorrection"})
>     context.stopsubsubject()
> 
>     context.startsubject({title="Obélix"})
> 
>     Eleve = {
>        structure = "le détail doit être travaillé. Pensez à la finesse 
> de vos menhirs…"
>        NoteStructure = 1,
>     }
>     context.getbuffer({"ExoCorrection"})
>     context.stopsubsubject()
>     context.stoptext()
> 
> The troubles is that buffers are filled with
> the last version of Eleve table.
> I guess this is because the whole file
> is parsed and then filled in buffers.
> My understanding of the cld manuel is that
> context.tobuffer()
> and
> context.direct()
> 
> should be used, but couldn't figure out how.
> Tried several times, also with context.step but
> all failed.
> I think I didn't properly understand the "temp"
> and str of context.tobuffer description in manual.
> 
> What changes should I introduce to my cld files?
see attached ... it's more that you overload Eleve ... an alternative is 
to make a function that gets that table as (then local) argument but I 
don't want to touch your code too much

the function wrapper sort of delays (immediate evaluation is a general 
problem, of function driven typesetting, which is why in the end a pure 
function driven one is not better / easier than tex)


Hans

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------