Splash-screen, elisp files compression, gdb usage

Splash-screen, elisp files compression, gdb usage

[ESCOUFLAIRE Pierre-Louis]

Hello,

While using emacs-22.1, I found the following problems:

   1- splash-screen '(setq inhibit-startup-message nil)':
       . calling emacs (no arguments):
           A splash-screen appears, which is perfectly normal.
       . calling emacs <filename>:
           A splash-screen appears, which is a regression from emacs-21.
           emacs waits for an input event to open associated buffer.

THIS IS NOT A BUG but RATHER A SUGGESTION.
I think that users requesting a file expect that file to be opened 
directly without having the splash-screen. I know that setting the 
'inhibit-startup-message' to 't' will avoid this problem but some 
configuration are using this splash-screen and can be annoying in 
certain cases (such as loading an elisp diff script in emacs from 
command line).
Newcomers to emacs are generally using 'emacs' with no arguments on 
their first use of it so they would see the splash-screen. Plus, they 
still can use the Help menu for tutorial. Thus, I do NOT think it is a 
good idea to display this splash-screen when given a <filename>.


   2- elisp files compression:
       . elisp source files are compressed on emacs 'make install'.
         Navigating through gzipped elisp source files is transparent.
	Byte-compiling gzipped elisp source files is not.

THIS HAPPENS ONLY FOR FILES WITH DEPENDENCIES.
THIS COULD BE AN ENHANCEMENT FOR THE NEXT RELEASE.
One could need to patch an emacs-provided elisp source file that has 
been gzipped and byte-compilation may fail if dependencies exist for 
this file.
Should it not be possible to have '--no_compress' (or look-alike) option 
to the 'configure' script so that problem for compilation transparency 
could be avoided?


   3- gdb
   3.1- gdb default behavior

       . GDB '--annotate=3' option is automatically added.
       . GDB finds the first executable in the current directory

THIS IS NOT A BUG but RATHER A SUGGESTION.
Default GDB behavior is to add the '--annotate=3' option that is unused 
by several GDB users. In our case, it has also made emacs freezing while 
typing a simple command such as 'help'.
Default GDB behavior is to find the first file matching 'execute' mode 
on in the current directory. The problem is that it also finds scripts 
(such as shell scripts) that are not 'debug-able'. I heard that emacs 
GDB had an enhancement to allow perl scripts debugging but it should not 
interfere with others. Perhaps an option/variable/... could help to 
disable this feature (e.g, variable gud-matching-executables-search).


   3.2- gdb navigation

       . GDB does not (always) navigate through source code.
         up/down/breakpoints/... do not switch to source code.

Here is a small/simple example:

-- main.c
   #include "pouet.h"
   int main () {
     pouet ();
   }

-- pouet.c
   #include "stdio.h"
   void pouet () {
     printf ("pouet-separate\n");
   }

-- pouet.h
   void pouet ();

--------------------

$ emacs
-> M-x gdb RET pouet RET
(gdb) b pouet
(gdb) r
(gdb) up
(gdb) do
(gdb)




Thanks a lot for your time and attention.

Regards,
Pierre-Louis ESCOUFLAIRE

____

This message and any files transmitted with it are legally privileged and intended for the sole use of the individual(s) or entity to whom they are addressed. If you are not the intended recipient, please notify the sender by reply and delete the message and any attachments from your system. Any unauthorised use or disclosure of the content of this message is strictly prohibited and may be unlawful.

Nothing in this e-mail message amounts to a contractual or legal commitment on the part of EUROCONTROL, unless it is confirmed by appropriately signed hard copy.

Any views expressed in this message are those of the sender.

Re: Splash-screen, elisp files compression, gdb usage

[Nick Roberts]

>   3- gdb
>    3.1- gdb default behavior
> 
>        . GDB '--annotate=3' option is automatically added.
>        . GDB finds the first executable in the current directory

> THIS IS NOT A BUG but RATHER A SUGGESTION.
> Default GDB behavior is to add the '--annotate=3' option that is unused 
> by several GDB users. 

The '--annotate=3' option instructs GDB to markup the output so that Emacs can
parse it.  It should be transparent to the user.  Emacs 21.1 uses as similar
option '--fullname' (equivalent to '--annotate=1') which gives the line number
and name of the source file where execution stopped.  However in this case
the option is not displayed in the mini-buffer.

>                        In our case, it has also made emacs freezing while 
> typing a simple command such as 'help'.

This shouldn't happen.  Can you please file a precise bug report with
M-x report-emacs-bug which will automatically include your configuration.


>   3.2- gdb navigation

>        . GDB does not (always) navigate through source code.
>          up/down/breakpoints/... do not switch to source code.

> Here is a small/simple example:

It should work.  Have you compiled with -g?

-- 
Nick                                           http://www.inet.net.nz/~nickrob

Re: Splash-screen, elisp files compression, gdb usage

[ESCOUFLAIRE Pierre-Louis]

Hello,

First of all, my apologies for sending an email with mixed stuff in it.
I will split it into different subjects to make it more readable and 
easier for you to dispatch to the good people.

Then, the problems have been solved thanks to your concern.

Thanks for the information on the '--annotate' option.
I would like to notice that the Ctrl-D binding is behaving as RET on 
end-of-buffer. I do not know if this was expected or not but that can be 
rather annoying for those used to this binding to quit GDB.

Otherwise, for the navigation, it seems it is not an error from GDB. In 
fact, the buffer is opened but not displayed and we suspect a problem in 
elisp (usage of text-properties, a bug report should be sent soon).

Finally, for the last bug, we cannot reproduce it. The configuration 
might have been a bit altered in that particular moment.


In any case, sorry for the inconvenience and thanks a lot for your time 
and attention.


Regards,
Pierre-Louis ESCOUFLAIRE

____

This message and any files transmitted with it are legally privileged and intended for the sole use of the individual(s) or entity to whom they are addressed. If you are not the intended recipient, please notify the sender by reply and delete the message and any attachments from your system. Any unauthorised use or disclosure of the content of this message is strictly prohibited and may be unlawful.

Nothing in this e-mail message amounts to a contractual or legal commitment on the part of EUROCONTROL, unless it is confirmed by appropriately signed hard copy.

Any views expressed in this message are those of the sender.

Re: Splash-screen, elisp files compression, gdb usage

[Nick Roberts]

 > Thanks for the information on the '--annotate' option.
 > I would like to notice that the Ctrl-D binding is behaving as RET on 
 > end-of-buffer. I do not know if this was expected or not but that can be 
 > rather annoying for those used to this binding to quit GDB.

Hmm, I did try to deal with this but it doesn't seem to do the right thing
now.  Does this patch improve things for you?

-- 
Nick                                           http://www.inet.net.nz/~nickrob


*** gdb-ui.el	26 Jul 2007 09:28:01 +1200	1.206.2.2
--- gdb-ui.el	11 Aug 2007 15:56:56 +1200	
*************** This filter may simply queue input for a
*** 1129,1138 ****
  	  (let ((item (concat string "\n")))
  	    (if gdb-enable-debug (push (cons 'send item) gdb-debug-log))
  	    (process-send-string proc item)))
!       (if (and (string-match "\\\\$" string)
! 	       (not comint-input-sender-no-newline)) ;;Try to catch C-d.
  	  (setq gdb-continuation (concat gdb-continuation string "\n"))
! 	(let ((item (concat gdb-continuation string "\n")))
  	  (gdb-enqueue-input item)
  	  (setq gdb-continuation nil)))))
  
--- 1129,1138 ----
  	  (let ((item (concat string "\n")))
  	    (if gdb-enable-debug (push (cons 'send item) gdb-debug-log))
  	    (process-send-string proc item)))
!       (if (string-match "\\\\\\'" string)
  	  (setq gdb-continuation (concat gdb-continuation string "\n"))
! 	(let ((item (concat gdb-continuation string
! 			 (if (not comint-input-sender-no-newline) "\n"))))
  	  (gdb-enqueue-input item)
  	  (setq gdb-continuation nil)))))

Re: Splash-screen, elisp files compression, gdb usage

[ESCOUFLAIRE Pierre-Louis]

Yes, it does. :)

Thanks for your concern.

Regards,
Pierre-Louis ESCOUFLAIRE


Nick Roberts wrote:
>  > Thanks for the information on the '--annotate' option.
>  > I would like to notice that the Ctrl-D binding is behaving as RET on 
>  > end-of-buffer. I do not know if this was expected or not but that can be 
>  > rather annoying for those used to this binding to quit GDB.
> 
> Hmm, I did try to deal with this but it doesn't seem to do the right thing
> now.  Does this patch improve things for you?

____

This message and any files transmitted with it are legally privileged and intended for the sole use of the individual(s) or entity to whom they are addressed. If you are not the intended recipient, please notify the sender by reply and delete the message and any attachments from your system. Any unauthorised use or disclosure of the content of this message is strictly prohibited and may be unlawful.

Nothing in this e-mail message amounts to a contractual or legal commitment on the part of EUROCONTROL, unless it is confirmed by appropriately signed hard copy.

Any views expressed in this message are those of the sender.

Re: Splash-screen, elisp files compression, gdb usage

[Nick Roberts]

 > Yes, it does. :)

Thanks for testing it.  I've committed this change to EMACS_22_BASE, it
should filter through to the trunk shortly.

-- 
Nick                                           http://www.inet.net.nz/~nickrob

 > Thanks for your concern.
 > 
 > Regards,
 > Pierre-Louis ESCOUFLAIRE
 > 
 > 
 > Nick Roberts wrote:
 > >  > Thanks for the information on the '--annotate' option.
 > >  > I would like to notice that the Ctrl-D binding is behaving as RET on 
 > >  > end-of-buffer. I do not know if this was expected or not but that can be 
 > >  > rather annoying for those used to this binding to quit GDB.
 > > 
 > > Hmm, I did try to deal with this but it doesn't seem to do the right thing
 > > now.  Does this patch improve things for you?