Black Screen: lift (TFA) (unitype.icl)

[Vag Vagoff]

For once something new:

module test

Start = test

:: A i = A i
:: B :== A.i: A i

test :: [B]
test = []

gives

    lift (TFA) (unitype.icl)

on dreadful black screen.
_______________________________________________
clean-list mailing list
clean-list@...
http://mailman.science.ru.nl/mailman/listinfo/clean-list

[Carlos Aya]

Likewise, on the command line I got
------------
$bin/clm test
Compiling test
lift (TFA) (unitype.icl)clm: reading compiler result failed: No such file or directory
$
-------------

And the message is right, I could find such unitype.icl file around (I'm suing the Linux distribution).

regards
Carlos

--- On Tue, 11/8/09, clean-list-request@... <clean-list-request@...> wrote:

Message: 1
Date: Mon, 10 Aug 2009 21:15:26 +0300
From: Vag <vag.vagoff@...>
Subject: [clean-list] Black Screen: lift (TFA) (unitype.icl)
To: clean-list <clean-list@...>
Message-ID:
    <b4e8fb960908101115s30801acdta11c3c065727aef5@...>
Content-Type: text/plain; charset=ISO-8859-1

For once something new:

module test

Start = test

:: A i = A i
:: B :== A.i: A i

test :: [B]
test = []

gives

    lift (TFA) (unitype.icl)

on dreadful black screen.

---

Find local businesses and services in your area with Yahoo!7 Local. Get started.
_______________________________________________
clean-list mailing list
clean-list@...
http://mailman.science.ru.nl/mailman/listinfo/clean-list

[Vag Vagoff]

> $bin/clm test
> Compiling test
> lift (TFA) (unitype.icl)clm: reading compiler result failed: No such file or
> directory

Actually, you got two messages: one from compiler executable (cocl)
called by clm:

    lift (TFA) (unitype.icl

and one from clm itself:

    clm: reading compiler result failed: No such file or directory

complaining that compiler died and left no output files.

unitype.icl is a part of Clean compiler frontend. It lives at
compiler/frontend/unitype.icl relative to CVS root. It contains lines

        lift modules cons_vars (TFA vars type) subst ls
                = abort "lift (TFA) (unitype.icl)"

and that is his post mortem message.
_______________________________________________
clean-list mailing list
clean-list@...
http://mailman.science.ru.nl/mailman/listinfo/clean-list

[John van Groningen]

>For once something new:
>
>module test
>
>Start = test
>
>:: A i = A i
>:: B :== A.i: A i
>
>test :: [B]
>test = []
>
>gives
>
>    lift (TFA) (unitype.icl)
>
>on dreadful black screen.

A for all quantifier (A. var: ) can only be used at the beginning
of the type of a function argument, constructor argument or field
of a record. Not in a type synonym.

Unfortunately the parser accepts the quantifier almost everywhere
in a type, and this may cause a compiler crash.

Kind regards,

John van Groningen
_______________________________________________
clean-list mailing list
clean-list@...
http://mailman.science.ru.nl/mailman/listinfo/clean-list

[Vag Vagoff]

John van Groningen wrote:
> A for all quantifier (A. var: ) can only be used at the beginning
> of the type of a function argument, constructor argument or field
> of a record. Not in a type synonym.
>
>  
Why then the following code causes same error message (TFA):

module SelfAppArgument

Start = a \ a = a

a :: (A.a:a -> a) -> (A.a:a -> a)
a a = a a

?
_______________________________________________
clean-list mailing list
clean-list@...
http://mailman.science.ru.nl/mailman/listinfo/clean-list

[John van Groningen]

Vag wrote:

>John van Groningen wrote:
>>A for all quantifier (A. var: ) can only be used at the beginning
>>of the type of a function argument, constructor argument or field
>>of a record. Not in a type synonym.
>>
>>
>Why then the following code causes same error message (TFA):
>
>module SelfAppArgument
>
>Start = a \ a = a
>
>a :: (A.a:a -> a) -> (A.a:a -> a)
>a a = a a

Because you are using a for all quantifier at the beginning of the
result of a function type specification, and not at the beginning of
an argument of a function type specification.

Kind regards,

John van Groningen

_______________________________________________
clean-list mailing list
clean-list@...
http://mailman.science.ru.nl/mailman/listinfo/clean-list