[asdf-devel] Changes to ASDF for LW6

[Robert Goldman]

On 2/26/10 Feb 26 -10:52 AM, james anderson wrote:
> hmmm...
> 
> On 2010-02-26, at 17:42 , Robert Goldman wrote:
> 
>> On 2/26/10 Feb 26 -10:17 AM, james anderson wrote:
>>> good evening;
>>>
>>> as an aside, in light of these changes.
>>>
>>> ? is there some reason that they are not coded to expand at compile
>>> time into a second phase of operations in terms of strings.
>>> that two-phase method gets the symbols out of the image earlier as
>>> they're not needed in fasls. as a side-effect, #: idiom would not be
>>> necessary.
>>
>> I suspect that this may be necessary in order to make this work on ACL
>> in "modern," case-sensitive mode.  Using the symbol-name of an
>> uninterned symbol is usually the way I code for portability across  
>> mlisp
>> and other lisps.
>>
>> Note that you can't reliably just do this at compile-time, since  
>> you can
>> toggle the case mode in ACL.
> 
> you mean, you actually build with one case mode and load into a lisp  
> which is runnign with another.
> and the fasl loader adjust the symbol names?
> i have used "modern" acl's but, yes, that order of pain never  
> occurred to me.
> 
> when life gets boring, does one toggle the case-mode of running  
> production systems?

One certainly can, if one is the sort of person for whom hang-gliding
has gotten too boring.

"The functions set-case-mode and convert-mixed-case-symbols, and two
variables *current-case-mode* and *ignore-package-name-case* are
provided for controlling and sensing case modes." from the ACL docs....

best,
r