Re: [Chicken-hackers] [PATCH] bug in type-validation for "deprecated" declaration

[Peter Bex]

On Fri, Sep 07, 2012 at 11:51:14PM +0200, Felix wrote:
> The attached patch fixes a bug reported by Christian. The type-validation
> of "deprecated" type-specifiers returned an incorrect value. Also, a small
> bug in the type-database is fixed. 

I decided to add some regression tests for this and found that there's
a second bug.  Before, deprecated procedures which have a suggested
alternative would cause the compiler to crash, but with this patch
they do not get an error message shown, because their assigned type
is '(*) now, which breaks the global-result type dispatch.

Attached is a modified version of your patch which fixes the bug,
and contains a regression test.  I also removed the older deprecation
regression test because we probably want to eventually get rid of
get-environment-variable, so we can't keep using it in our tests.
This makes a bigger diff because all the line numbers moved and the
gensymed names are different.

> This should go into 4.8.0, I think.

I agree.  It's an annoying bug which causes breakage.

Cheers,
Peter
-- 
http://sjamaan.ath.cx
--
"The process of preparing programs for a digital computer
 is especially attractive, not only because it can be economically
 and scientifically rewarding, but also because it can be an aesthetic
 experience much like composing poetry or music."
                                                        -- Donald Knuth

0001-Type-validation-returned-incorrect-result-for-deprec.patch
Description: Text document