`gnus-bound-and-true-p'

`gnus-bound-and-true-p'

[Reiner Steib]

On Tue, Nov 01 2011, Ted Zlatanov wrote:
> -(eval-when-compile
> -  ;; This is unnecessary in the compiled version as it is a macro.
> -  (if (fboundp 'bound-and-true-p)
> -      (defalias 'gnus-bound-and-true-p 'bound-and-true-p)
> -    (defmacro gnus-bound-and-true-p (var)
> -      "Return the value of symbol VAR if it is bound, else nil."
> -      `(and (boundp (quote ,var)) ,var))))
> -
> -(defun gnus-bound-and-true-dumber-p (sym)
> +;; simple check, can be a macro but this way, although slow, it's really clear
> +(defun gnus-bound-and-true-p (sym)
>    (and (boundp sym) (symbol-value sym)))

On Wed, Nov 02 2011, Ted Zlatanov wrote:
> -;; simple check, can be a macro but this way, although slow, it's really clear
> +;; Simple check: can be a macro but this way, although slow, it's really clear.
> +;; We don't use `bound-and-true-p' because it's not in XEmacs.
>  (defun gnus-bound-and-true-p (sym)
>    (and (boundp sym) (symbol-value sym)))

Can't you use an alias for Emacs and a macro for XEmacs?

(if (fboundp 'bound-and-true-p)
    (defalias 'gnus-bound-and-true-p 'bound-and-true-p)
  (defmacro gnus-bound-and-true-p (var)
    "Return the value of symbol VAR if it is bound, else nil."
    `(and (boundp (quote ,var)) ,var)))

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: `gnus-bound-and-true-p'

[Robert Pluim]

Reiner Steib <reinersteib+gmane@imap.cc> writes:

> On Tue, Nov 01 2011, Ted Zlatanov wrote:
>> -(eval-when-compile
>> -  ;; This is unnecessary in the compiled version as it is a macro.
>> -  (if (fboundp 'bound-and-true-p)
>> -      (defalias 'gnus-bound-and-true-p 'bound-and-true-p)
>> -    (defmacro gnus-bound-and-true-p (var)
>> -      "Return the value of symbol VAR if it is bound, else nil."
>> -      `(and (boundp (quote ,var)) ,var))))
>> -
>> -(defun gnus-bound-and-true-dumber-p (sym)
>> +;; simple check, can be a macro but this way, although slow, it's really clear
>> +(defun gnus-bound-and-true-p (sym)
>>    (and (boundp sym) (symbol-value sym)))
>
> On Wed, Nov 02 2011, Ted Zlatanov wrote:
>> -;; simple check, can be a macro but this way, although slow, it's really clear
>> +;; Simple check: can be a macro but this way, although slow, it's really clear.
>> +;; We don't use `bound-and-true-p' because it's not in XEmacs.
>>  (defun gnus-bound-and-true-p (sym)
>>    (and (boundp sym) (symbol-value sym)))
>
> Can't you use an alias for Emacs and a macro for XEmacs?
>
> (if (fboundp 'bound-and-true-p)
>     (defalias 'gnus-bound-and-true-p 'bound-and-true-p)
>   (defmacro gnus-bound-and-true-p (var)
>     "Return the value of symbol VAR if it is bound, else nil."
>     `(and (boundp (quote ,var)) ,var)))

Especially since in XEmacs 21.5:

`bound-and-true-p' is a compiled Lisp macro
  -- loaded from "/build/xemacs-edge/lisp/subr.elc"
(bound-and-true-p VAR)

Documentation:
Return the value of symbol VAR if it is bound, else nil.

And:

(fboundp 'bound-and-true-p) => t

Robert

Re: `gnus-bound-and-true-p'

[Ted Zlatanov]

On Wed, 02 Nov 2011 22:09:32 +0100 Robert Pluim <rpluim@gmail.com> wrote: 

RP> Reiner Steib <reinersteib+gmane@imap.cc> writes:

>> Can't you use an alias for Emacs and a macro for XEmacs?
>> 
>> (if (fboundp 'bound-and-true-p)
>> (defalias 'gnus-bound-and-true-p 'bound-and-true-p)
>> (defmacro gnus-bound-and-true-p (var)
>> "Return the value of symbol VAR if it is bound, else nil."
>> `(and (boundp (quote ,var)) ,var)))

That was failing in XEmacs according to the Gnus Buildbot.  See
http://www.randomsample.de:4456/grid, older builds.  It may have been
the `eval-when-compile'.

RP> Especially since in XEmacs 21.5:

RP> `bound-and-true-p' is a compiled Lisp macro
RP>   -- loaded from "/build/xemacs-edge/lisp/subr.elc"
RP> (bound-and-true-p VAR)

RP> Documentation:
RP> Return the value of symbol VAR if it is bound, else nil.

RP> And:

RP> (fboundp 'bound-and-true-p) => t

I've spend too many hours dealing with this issue.  So if you guys or
anyone else can provide a patch that works in Emacs and XEmacs, please
do and let's be done.  Please keep in mind this is used very rarely, so
not using a macro works perfectly well as far as performance, but I
agree that it's nice to rely on the built-in functionality.

Thanks
Ted

Re: `gnus-bound-and-true-p'

[Dave Abrahams]

on Wed Nov 02 2011, Ted Zlatanov <tzz-AT-lifelogs.com> wrote:

> I've spend too many hours dealing with this issue.  So if you guys or
> anyone else can provide a patch that works in Emacs and XEmacs, please
> do and let's be done.  Please keep in mind this is used very rarely, so
> not using a macro works perfectly well as far as performance, but I
> agree that it's nice to rely on the built-in functionality.

I'm sorry, but I think all this focus on bound-and-true-p is
misdirected.  IIUC, the fix for the problem is in this patch:

  http://lists.gnu.org/archive/html/emacs-devel/2011-11/msg00000.html

I simply didn't realize that (defvar <symbol>) would leave <symbol>
unbound.

-- 
Dave Abrahams
BoostPro Computing
http://www.boostpro.com

Re: `gnus-bound-and-true-p'

[Ted Zlatanov]

On Wed, 02 Nov 2011 14:48:16 -0800 Dave Abrahams <dave@boostpro.com> wrote: 

DA> on Wed Nov 02 2011, Ted Zlatanov <tzz-AT-lifelogs.com> wrote:

>> I've spend too many hours dealing with this issue.  So if you guys or
>> anyone else can provide a patch that works in Emacs and XEmacs, please
>> do and let's be done.  Please keep in mind this is used very rarely, so
>> not using a macro works perfectly well as far as performance, but I
>> agree that it's nice to rely on the built-in functionality.

DA> I'm sorry, but I think all this focus on bound-and-true-p is
DA> misdirected.  IIUC, the fix for the problem is in this patch:

DA>   http://lists.gnu.org/archive/html/emacs-devel/2011-11/msg00000.html

DA> I simply didn't realize that (defvar <symbol>) would leave <symbol>
DA> unbound.

The macro was failing on XEmacs even after I fixed the defvar, according
to the Buildbot.  That's why I got rid of it.  So I don't think it was
your fault :)

Ted