james anderson | 1 Feb 2010 15:10
Picon
Gravatar

http://github.com/lisp/mcl

good afternoon;

i have accumulated over the past year several patch files for 5.2  
final. i have put them here:

   http://github.com/lisp/mcl

if anyone else has patches, please let me know.

_______________________________________________
info-mcl mailing list
info-mcl@...
http://clozure.com/mailman/listinfo/info-mcl

Sandro Pedrazzini | 1 Feb 2010 15:38
Favicon

Foreign functions from RMCL 5.1 to RMCL 5.2

Hi,

What is the difference in foreign functions' use from RMCL 5.1 to RMCL  
5.2?

I have a Carbon shared library  that I specify and use in RMCL 5.1 as  
follows:

(ccl::add-to-shared-library-search-path "CarbonRegex")

(define-entry-point ("match_pattern" ("CarbonRegex"))
   ((pat :string)
    (str :string))
   :long)

(defun c-match-pattern (pattern string)  ;;(print-db pattern)(print-db  
string)
   (with-cstrs ((pat pattern)
                (str string))
     (match_pattern pat str)))

;;  tests

(c-match-pattern ".*[ae]uer" "sauer")
=> 5

Everything is fine.

Trying the same thing on 5.2 I get following error, after calling the  
"c-match-pattern" function:
(Continue reading)

Terje Norderhaug | 1 Feb 2010 17:35

http://github.com/lisp/mcl

On Feb 1, 2010, at 6:10 AM, james anderson wrote:
> good afternoon;
>
> i have accumulated over the past year several patch files for 5.2
> final. i have put them here:
>
>    http://github.com/lisp/mcl
>
> if anyone else has patches, please let me know.

Many patches/fixes for RMCL 5.2.x are also relevant for MCL 5.2, and  
vice versa. See:

   http://code.google.com/p/mcl/issues/list

Thank you for sharing the patches!

-- Terje Norderhaug
   terje@...

_______________________________________________
info-mcl mailing list
info-mcl@...
http://clozure.com/mailman/listinfo/info-mcl

Rainer Joswig | 1 Feb 2010 17:56
Picon

http://github.com/lisp/mcl

http://github.com/lisp/mcl/blob/master/patches/patches%205.2/20090420-5.2.lisp

Above looks kind of strange...

Regards,

Rainer Joswig

Am 01.02.2010 um 15:10 schrieb james anderson:

> good afternoon;
> 
> i have accumulated over the past year several patch files for 5.2  
> final. i have put them here:
> 
>   http://github.com/lisp/mcl
> 
> if anyone else has patches, please let me know.
> 
> 
> _______________________________________________
> info-mcl mailing list
> info-mcl@...
> http://clozure.com/mailman/listinfo/info-mcl

Rainer Joswig, Hamburg, Germany
http://lispm.dyndns.org/
mailto:joswig@...

_______________________________________________
(Continue reading)

james anderson | 1 Feb 2010 18:26
Picon
Gravatar

http://github.com/lisp/mcl

the only thing i can guess - from looking at 'git diff' after an  
innocuous change, is that it didn't cope with the older eol form.
i've pushed another copy and it looks better.

On 2010-02-01, at 17:56 , Rainer Joswig wrote:

> http://github.com/lisp/mcl/blob/master/patches/patches% 
> 205.2/20090420-5.2.lisp
>
> Above looks kind of strange...
>
> Regards,
>
> Rainer Joswig
>
> Am 01.02.2010 um 15:10 schrieb james anderson:
>
>> good afternoon;
>>
>> i have accumulated over the past year several patch files for 5.2
>> final. i have put them here:
>>
>>   http://github.com/lisp/mcl
>>
>> if anyone else has patches, please let me know.
>>
>>
>> _______________________________________________
>> info-mcl mailing list
>> info-mcl@...
(Continue reading)

Terje Norderhaug | 1 Feb 2010 19:22

Re: Foreign functions from RMCL 5.1 to RMCL 5.2

On Feb 1, 2010, at 6:38 AM, Sandro Pedrazzini wrote:
> What is the difference in foreign functions' use from RMCL 5.1 to RMCL
> 5.2?

MCL 5.2 upgraded to by default use Mach-O instead of CFM. Try setting  
*dont-use-cfm* to NIL before defining the entry point for old school  
CFM shared libs.

-- Terje

> I have a Carbon shared library  that I specify and use in RMCL 5.1 as
> follows:
>
> (ccl::add-to-shared-library-search-path "CarbonRegex")
>
> (define-entry-point ("match_pattern" ("CarbonRegex"))
>    ((pat :string)
>     (str :string))
>    :long)
>
> (defun c-match-pattern (pattern string)  ;;(print-db pattern)(print-db
> string)
>    (with-cstrs ((pat pattern)
>                 (str string))
>      (match_pattern pat str)))
>
> ;;  tests
>
> (c-match-pattern ".*[ae]uer" "sauer")
> => 5
(Continue reading)

Rainer Joswig | 1 Feb 2010 19:38
Picon

http://github.com/lisp/mcl

Many thanks.

Regards,

Rainer Joswig

Am 01.02.2010 um 18:26 schrieb james anderson:

> the only thing i can guess - from looking at 'git diff' after an  
> innocuous change, is that it didn't cope with the older eol form.
> i've pushed another copy and it looks better.
> 
> On 2010-02-01, at 17:56 , Rainer Joswig wrote:
> 
>> http://github.com/lisp/mcl/blob/master/patches/patches% 
>> 205.2/20090420-5.2.lisp
>> 
>> Above looks kind of strange...
>> 
>> Regards,
>> 
>> Rainer Joswig
>> 
>> Am 01.02.2010 um 15:10 schrieb james anderson:
>> 
>>> good afternoon;
>>> 
>>> i have accumulated over the past year several patch files for 5.2
>>> final. i have put them here:
>>> 
(Continue reading)

peter | 2 Feb 2010 17:14

Re: Popup in RMCL 5.2.1

At 4:18 PM +0100 10/2/2, Sandro Pedrazzini wrote:
>Hi,
>
>Trying to migrate my application from RMCL 5.1 to 5.2.1 I am 
>encountering following problem with popup menus:
>
>- After fresh loading the application files on RMCL 5.2.1, the popup 
>menus are ok, those of my application, as well as those of RMCL.
>
>- After performing a "save-application" and restarting it through 
>the saved image, the popup menus are not drawn anymore. The 
>menu-items are there, you can select them pointing the mouse into 
>the popup menu area, but the menu itself is not visible (as shown in 
>the picture attached, showing the "Tools:Apropos..."  dialog).
>
>
>Did someone encounter a similar problem?

Yes. But I haven't begun to find out what's going on.
I've seen the same effect in simple static-text-dialog-item after 
just loading a file (rather than running a saved image). I wondered 
if somehow the layer ordering has messed up.  They seem to be there, 
know about their view-container, wptr, dimensions etc. and respond to 
clicks OK, just be invisible.

>It seems as I am loading something in the application that prevents 
>the popups to draw themselves, but only after a save-application. I 
>see the problem in 5.2 and 5.2.1. In 5.1 I did not have this problem.
>Where should I look?

(Continue reading)

Terje Norderhaug | 2 Feb 2010 17:30

Re: Popup in RMCL 5.2.1

On Feb 2, 2010, at 7:18 AM, Sandro Pedrazzini wrote:
> Did someone encounter a similar problem?
> It seems as I am loading something in the application that prevents  
> the popups to draw themselves, but only after a save-application. I  
> see the problem in 5.2 and 5.2.1. In 5.1 I did not have this problem.
> Where should I look?

I don't see this problem at all with saved MCL applications. Then  
again, that's after loading a number of patches.

You may try to load modern-mcl before saving the application and see  
whether that eliminates the issue.

-- Terje Norderhaug
   terje@...

_______________________________________________
info-mcl mailing list
info-mcl@...
http://clozure.com/mailman/listinfo/info-mcl

Sandro Pedrazzini | 2 Feb 2010 16:18
Favicon

Popup in RMCL 5.2.1

Hi,

Trying to migrate my application from RMCL 5.1 to 5.2.1 I am  
encountering following problem with popup menus:

- After fresh loading the application files on RMCL 5.2.1, the popup  
menus are ok, those of my application, as well as those of RMCL.

- After performing a "save-application" and restarting it through the  
saved image, the popup menus are not drawn anymore. The menu-items are  
there, you can select them pointing the mouse into the popup menu  
area, but the menu itself is not visible (as shown in the picture  
attached, showing the "Tools:Apropos..."  dialog).

Did someone encounter a similar problem?
It seems as I am loading something in the application that prevents  
the popups to draw themselves, but only after a save-application. I  
see the problem in 5.2 and 5.2.1. In 5.1 I did not have this problem.
Where should I look?

Thank you in advance for any hint.

Best regards
Sandro


(Continue reading)


Gmane