[Caml-list] serious type confusion

[Caml-list] serious type confusion

[briand]

OK, this is what I get for coding too long in scheme :-)

observe the following:

# let a=ref [];;
val a : '_a list ref = {contents = []}
# a := {x=0.;y=0.;z=0.} :: [];;
- : unit = ()
# !a;;
- : point list = [{x = 0.; y = 0.; z = 0.}]
# a := {x=0.;y=0.;z=0.} :: !a;;
- : unit = ()
# !a
  ;;
- : point list = [{x = 0.; y = 0.; z = 0.}; {x = 0.; y = 0.; z = 0.}]
# 

Exactly as expected !  types aren't so bad after all..  until here :

              current_polyline := {x=wx; y=wy; z=wz} :: !current_polyline;

This expression has type point but is here used with type point list

Huh?  The first example worked.  Oh I see, it's operator precedence,
it's assigning the point before doing the concatenation.  Oh but wait,
it worked in the previous example.  I'll try and fix it anyway...

              current_polyline := ({x=wx; y=wy; z=wz} :: !current_polyline);

This expression has type point but is here used with type point list

Huh ???  Isn't that impossible ?  I've triple checked and
current_polyline is always consistently used as a list, and more
importantly a point list...

Any hints ?

Thank You

Brian

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners

[Caml-list] serious type confusion

[briand]

I am in awe of type inferencing.

Sure enough I checked the code and I was using current_polyline in a
manner inconsistent with a point list.

never mind...

Brian

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners

Re: [Caml-list] serious type confusion

[Matt Gushee]

On Sat, Mar 27, 2004 at 09:31:00AM -0800, briand@aracnet.com wrote:
> 
> OK, this is what I get for coding too long in scheme :-)
> 
> observe the following:
> 
> # let a=ref [];;
> val a : '_a list ref = {contents = []}
> # a := {x=0.;y=0.;z=0.} :: [];;
> - : unit = ()
> # !a;;
> - : point list = [{x = 0.; y = 0.; z = 0.}]
> # a := {x=0.;y=0.;z=0.} :: !a;;
> - : unit = ()
> # !a
>   ;;
> - : point list = [{x = 0.; y = 0.; z = 0.}; {x = 0.; y = 0.; z = 0.}]
> # 
> 
> Exactly as expected !  types aren't so bad after all..  until here :
> 
>               current_polyline := {x=wx; y=wy; z=wz} :: !current_polyline;
> 
> This expression has type point but is here used with type point list
> 
> Huh?  The first example worked.  Oh I see, it's operator precedence,
> it's assigning the point before doing the concatenation.  Oh but wait,
> it worked in the previous example.  I'll try and fix it anyway...
> 
>               current_polyline := ({x=wx; y=wy; z=wz} :: !current_polyline);
> 
> This expression has type point but is here used with type point list
> 
> Huh ???  Isn't that impossible ?  I've triple checked and
> current_polyline is always consistently used as a list, and more
> importantly a point list...
> 
> Any hints ?

Hints, yes. Hard to say for sure without seeing more of your code, but
two things that occur to me offhand are:

* Is current_polyline defined within the same scope where the error
  occurs, or is it received as an argument? If it is the latter, and
  somewhere before the error location there is an expression that *uses*
  !current_polyline as a point, that would explain your error.  If your
  function is too complex to easily identify the problem by sight, an
  easy way to check for this would be to put a type constraint on the
  parameter:

    let myfun (current_polyline:point list ref) =
      ....

  That way you'll get an error on entry into the function if
  current_polyline is wrongly defined in the calling scope. If, on the
  other hand, there was an expression that tried to use the value as a
  point, it will now fail.

* Pay attention to *exactly* which expression is causing the error--
  i.e., not just the line number, but also the position in the line.
  Is it possible that !current_polyline is actually a 'point list list'
  rather than a 'point list'? If so, then of course the new element you
  add to it must be a 'point list'.

Hope this helps.

-- 
Matt Gushee                 When a nation follows the Way,
Englewood, Colorado, USA    Horses bear manure through
mgushee@havenrock.com           its fields;
http://www.havenrock.com/   When a nation ignores the Way,
                            Horses bear soldiers through
                                its streets.
                                
                            --Lao Tzu (Peter Merel, trans.)

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners

Re: [Caml-list] serious type confusion

[briand]

>>>>> "Matt" == Matt Gushee <mgushee@havenrock.com> writes:

  Matt> * Is current_polyline defined within the same scope where the error
  Matt>   occurs, or is it received as an argument? If it is the latter, and
  Matt>   somewhere before the error location there is an expression that *uses*
  Matt>   !current_polyline as a point, that would explain your error.  If your
  Matt>   function is too complex to easily identify the problem by sight, an
  Matt>   easy way to check for this would be to put a type constraint on the
  Matt>   parameter:

  Matt>     let myfun (current_polyline:point list ref) =
  Matt>       ....

That was it exactly (see my followup).

However that's a very nice trick for future debuggin effort.

Thanks

Brian

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners