Errors on missing completion?

Errors on missing completion?

[Lennart Borgman (gmail)]

dabbrev-expand gives an error when it can't complete while 
lisp-complete-symbol just gives a message.

I can't see any meaning in that Emacs commands gives errors in cases 
like this. Is there anyone who can?

Re: Errors on missing completion?

[Juri Linkov]

> dabbrev-expand gives an error when it can't complete while
> lisp-complete-symbol just gives a message.
>
> I can't see any meaning in that Emacs commands gives errors in cases like
> this. Is there anyone who can?

Do you see any problem with this?

-- 
Juri Linkov
http://www.jurta.org/emacs/

Re: Errors on missing completion?

[Lennart Borgman (gmail)]

Juri Linkov wrote:
>> dabbrev-expand gives an error when it can't complete while
>> lisp-complete-symbol just gives a message.
>>
>> I can't see any meaning in that Emacs commands gives errors in cases like
>> this. Is there anyone who can?
> 
> Do you see any problem with this?

If you have debug-on-error set to t it gets very disturbing. It is 
rather similar to if isearch should raise an error on a search miss.

Re: Errors on missing completion?

[Juri Linkov]

>>> dabbrev-expand gives an error when it can't complete while
>>> lisp-complete-symbol just gives a message.
>>>
>>> I can't see any meaning in that Emacs commands gives errors in cases like
>>> this. Is there anyone who can?
>>
>> Do you see any problem with this?
>
> If you have debug-on-error set to t it gets very disturbing. It is rather
> similar to if isearch should raise an error on a search miss.

There is a special variable `debug-ignored-errors' with the list
of error messages to ignore, and as I see it already contains
"^No dynamic expansion for .* found$".  So when debug-on-error
is non-nil, `dabbrev-expand' is not disturbing due to this variable.
Do you have a test case where `debug-ignored-errors' gets ignored?

-- 
Juri Linkov
http://www.jurta.org/emacs/

Re: Errors on missing completion?

[Lennart Borgman (gmail)]

Juri Linkov wrote:
>>>> dabbrev-expand gives an error when it can't complete while
>>>> lisp-complete-symbol just gives a message.
>>>>
>>>> I can't see any meaning in that Emacs commands gives errors in cases like
>>>> this. Is there anyone who can?
>>> Do you see any problem with this?
>> If you have debug-on-error set to t it gets very disturbing. It is rather
>> similar to if isearch should raise an error on a search miss.
> 
> There is a special variable `debug-ignored-errors' with the list
> of error messages to ignore, and as I see it already contains
> "^No dynamic expansion for .* found$".  So when debug-on-error
> is non-nil, `dabbrev-expand' is not disturbing due to this variable.
> Do you have a test case where `debug-ignored-errors' gets ignored?

Thanks, I forgot about that variable.

No, I can't understand what happened before right now. I get no error 
now. I will try to actually read the error message next time ;-)

Re: Errors on missing completion?

[Juri Linkov]

>> There is a special variable `debug-ignored-errors' with the list
>> of error messages to ignore, and as I see it already contains
>> "^No dynamic expansion for .* found$".  So when debug-on-error
>> is non-nil, `dabbrev-expand' is not disturbing due to this variable.
>> Do you have a test case where `debug-ignored-errors' gets ignored?
>
> Thanks, I forgot about that variable.
>
> No, I can't understand what happened before right now. I get no error
> now. I will try to actually read the error message next time ;-)

Ah, I now see what is wrong.  When `dabbrev-check-other-buffers' is nil,
the error message is "No dynamic expansion for \"%s\" found in this-buffer"
that is not matched by `debug-ignored-errors'.  The following patch
fixes this.  All the rest messages are already covered by this list.

Index: lisp/dabbrev.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/dabbrev.el,v
retrieving revision 1.88
diff -c -r1.88 dabbrev.el
*** lisp/dabbrev.el	8 Jan 2008 20:44:44 -0000	1.88
--- lisp/dabbrev.el	21 Mar 2008 00:03:35 -0000
***************
*** 1011,1017 ****
  		(cons found-string dabbrev--last-table))
  	  result)))))
  
! (dolist (mess '("^No dynamic expansion for .* found$"
  		"^No further dynamic expansion for .* found$"
  		"^No possible abbreviation preceding point$"))
    (add-to-list 'debug-ignored-errors mess))
--- 1011,1017 ----
  		(cons found-string dabbrev--last-table))
  	  result)))))
  
! (dolist (mess '("^No dynamic expansion for .* found"
  		"^No further dynamic expansion for .* found$"
  		"^No possible abbreviation preceding point$"))
    (add-to-list 'debug-ignored-errors mess))

-- 
Juri Linkov
http://www.jurta.org/emacs/

Re: Errors on missing completion?

[Lennart Borgman (gmail)]

Juri Linkov wrote:
>>> There is a special variable `debug-ignored-errors' with the list
>>> of error messages to ignore, and as I see it already contains
>>> "^No dynamic expansion for .* found$".  So when debug-on-error
>>> is non-nil, `dabbrev-expand' is not disturbing due to this variable.
>>> Do you have a test case where `debug-ignored-errors' gets ignored?
>> Thanks, I forgot about that variable.
>>
>> No, I can't understand what happened before right now. I get no error
>> now. I will try to actually read the error message next time ;-)
> 
> Ah, I now see what is wrong.  When `dabbrev-check-other-buffers' is nil,
> the error message is "No dynamic expansion for \"%s\" found in this-buffer"
> that is not matched by `debug-ignored-errors'.  The following patch
> fixes this.  All the rest messages are already covered by this list.


Oh, thanks. But wouldn't it be easier to use catch - throw in cases like 
this instead. Is not that what the debug-ignored-errors in a way implements?