Expanded objects in collection

[Farzad Sehat]

Damien, could you please isolate a minimal use case which when executed
crashes. And then pulish this usecase to alioth bug tracker.
Indeed, Ben email's doesn't work since a long time a he could not see you
bug. And moreover he realy wants us to make minimal
usecase for bug in Alioth.

M. Farzad Sehat
farzad.sehat at gmail.com


2009/11/13 Damien Bouvarel <damien.bouvarel at gmail.com>

> Hello,
>
> i'm restoring the opengl stuff and i'm facing the following error:
>
> Depending pass: .....--SEMANTIC---------
> Type `OBJECT' is invalid with `Expanded INTEGER'.
> Line 112 column 34 in
> OBJECT(/home/dams/Lisaac/lisaac_0.39_beta/lib/standard/kernel/object.li):
>   - Self:SELF '=='  Right 60 other:OBJECT :BOOLEAN <- ( deferred; FALSE);
>                                   ^
> Line 64 column 9 in
> HASHED_DICTIONARY(/home/dams/Lisaac/lisaac_0.39_beta/lib/standard/collection/
> hashed_dictionary.li):
>   - has k:K :BOOLEAN <-
>
> Triggered by the code:
>
> keys := HASHED_DICTIONARY(INTEGER,INTEGER).create;
> keys.add k_backspace to (`XK_BackSpace`:INTEGER);
> ...
>
>  keys.has key.if {         <<<
>   ...
> }
>
> So it is no more possible to store Expanded object like INTEGER in
> collections? It used to work..
> I think the '==' profil changed, and by the way what is the semantic of
> '~=' operator? i saw it somewhere.
>
> I tried the REFERENCE(E) on a simple example and it segfaults:
> + i:INTEGER;
> + ref:REFERENCE(INTEGER);
>
> ref := i;     <<<
>
> It should be ok because REFERENCE(E) imports from E? Nevertheless it
> shouldn't segfault  :)
>
> Damien
>
>
>
>
> _______________________________________________
> Lisaac-devel mailing list
> Lisaac-devel at lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/lisaac-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/lisaac-devel/attachments/20091114/1d0a1e3c/attachment.htm>