[plt-scheme] DrScheme buglet?

I now think that there shouldn't be a yellow warning.

Instead, if there is an error, drscheme just runs a program that only
contains the teachpacks, so the teachpacks are always available in the
REPL.

Robby

On 8/31/07, Matthias Felleisen <matthias at ccs.neu.edu> wrote:
> I am missing the yellow warning. (I did read the messages out of order.)
>
> On Aug 31, 2007, at 12:31 PM, Robby Findler wrote:
>
> > Did you try the current drscheme? :)
> >
> > On 8/31/07, Matthias Felleisen <matthias at ccs.neu.edu> wrote:
> >> I second Ryan's proposal.
> >>
> >> On Aug 29, 2007, at 11:17 AM, Ryan Culpepper wrote:
> >>
> >>> On Wed, 2007-08-29 at 08:52 -0500, Robby Findler wrote:
> >>>> This only happens when there are syntax errors, right? Not runtime
> >>>> errors?
> >>>>
> >>>> I think that's just a property of the way teachpacks work in the
> >>>> current world. I'm not sure it is possible to get around that.
> >>>>
> >>>> Roughly, what's happening underneath the hood, is that the
> >>>> teachpack
> >>>> is (an invisible) part of your program and, when your program is
> >>>> syntactically malformed, no part of your program is available in
> >>>> the
> >>>> REPL. For example, in this program:
> >>>>
> >>>>   (define x 1)
> >>>>   unbound-id
> >>>>
> >>>> You won't be able to use "x" in the REPL. But, in this one, you
> >>>> will:
> >>>>
> >>>>   (define x 1)
> >>>>   (first empty)
> >>>>
> >>>> I know that's not real help, but I hope it is at least explanatory.
> >>>>
> >>>> And maybe this will inspire someone (Matthew?) to think of a way
> >>>> to do
> >>>> better, I'm not sure.
> >>>
> >>> Would it make sense to make that explanation part of DrScheme? That
> >>> is:
> >>>
> >>> When a syntax error occurs, print a message to the effect of "No
> >>> definitions executed due to syntax error" in the interactions
> >>> window and
> >>> lock the interactions editor.
> >>>
> >>> Or instead of locking the interactions, DrScheme could instead
> >>> display
> >>> one of those yellow highlighted messages (like "definitions out of
> >>> sync"
> >>> is displayed now) before every interaction. That would let students
> >>> use
> >>> the interactions window to stumble around trying out syntaxes until
> >>> they
> >>> find the right one.
> >>>
> >>> Ryan
> >>>
> >>>
> >>>> On 8/29/07, Todd O'Bryan <toddobryan at mac.com> wrote:
> >>>>> In v371 of DrScheme, it appears that teachpacks don't "stick."
> >>>>>
> >>>>> When you first open a DrScheme window, even though it says that a
> >>>>> teachpack is loaded, it's unavailable until you hit Run.
> >>>>>
> >>>>> That's not a big problem, but the teachpacks disappear if
> >>>>> there's an
> >>>>> error during a Run. In other words, if students are developing a
> >>>>> function, hit run, and have an error, if they drop down to the
> >>>>> interactions window to experiment, the functions provided by the
> >>>>> teachpack are no longer available.
> >>>>>
> >>>>> The workaround is to comment out offending code in the definitions
> >>>>> window, but this is kind of a pain.
> >>>>>
> >>>>> Todd
> >>>>>
> >>>>> _________________________________________________
> >>>>>   For list-related administrative tasks:
> >>>>>   http://list.cs.brown.edu/mailman/listinfo/plt-scheme
> >>>>>
> >>>> _________________________________________________
> >>>>   For list-related administrative tasks:
> >>>>   http://list.cs.brown.edu/mailman/listinfo/plt-scheme
> >>>
> >>> _________________________________________________
> >>>   For list-related administrative tasks:
> >>>   http://list.cs.brown.edu/mailman/listinfo/plt-scheme
> >>
> >>
>
>