Re: File extension when saving

["Alan E. Davis" <aedavis eccomm com>]

Thank you for your response:

I may actually take the time to file a bug report; however, you state
that this is a well reasoned feature, not a bug.  Other programs do
automatically add the extension (eg, The Gimp), without complaining
about a dot in the file name.  What's the big deal, if it isn't a nod
to the dark side?  

Forgive my (lack of) humor.  

And I hope you will not take it as an unkindness if I suggest that
perhaps a check of the file's internal content would provide a more
reliable indication of the type of file involved?  I may be missing
something here.  

I apologize if my statement seemed to be over-the-top, or in any way
critical of the excellent work of the volunteers who have been
developing this great (and improving!) package.  Thanks and Kudo's to
everyone.  In my humble, user's opinion, and suspicion, it would be
possible to do what the developers have in mind, without penalizing the
use of any sort of legal filename, whatsoever.  Unless this is
necessary to accede to a M$ designed behavior, most of which that I
have encountered seem to be idiosyncratically flawed.  I may be wrong
about all or any of this.

Thanks again.  Your answer did help.  Some.

Alan Davis





On Fri, 25 Mar 2005 23:26:59 -0500
Adrian Custer <acuster nature berkeley edu> wrote:

> Hello,
>
> Sorry you are having an unpleasant experience.
>
> What I understand of your email, Gnumeric seems to be acting
> correctly. 
>
> The first file appears to have an extension, that is the file ends
> with a dot and followed by a small number of characters (one).
> Gnumeric sees the dot-three, takes it to be an extension, recognizes
> that such an extension is unusual for the xml format gnumeric uses,
> and therefore warns the user. 
>
> The second file does not appear to have any extension and therefore
> Gnumeric adds the correct extension for this file type. This
> behaviour has been examined extensively and the developers
> consistently have felt that having gnumeric silently add the
> extension is the best compromise.
>
> If you feel differently, please feel free to voice your opinion in
> the best place to bring about change. File a bug in
> bugzilla.gnome.org, but add some detail as to what you would expect
> in each situation. If you also discuss other corner cases, your bug
> is more likely to generate useful discussion. 
>
> However, please realize that the developers have thought extensively
> about this situation, and have come to agree upon the current
> compromise. They are therefore unlikely to change their mind without
> a well presented, logical, and comprehensive bug report. Also, I'd
> encourage you to drop the "this is the most broken behaviour I've
> ever seen" tone since that's likely simply to turn off those who make
> the decisions.
>
> best of luck,
> adrian
>
>
>
> On Sat, 2005-03-26 at 12:49 +1000, Alan E. Davis wrote:
> > Gnumeric exhibits a troubling behavior when saving files.  The
> > following two files received different treatment:
> >
> >     1.  masterfile.3
> >     2.  masterfile-3
> >
> > File #1: received a message that the extension does not match the
> > file type.
> >
> > File #2: saved with the .gnumeric extension,without complaint or
> > message.
> >
> > Is this a gnome behavior?  I think not, because other programs do
> > not do this.  This seems to me to be a derrier-garde bug, not a
> > feature. Reminds me of years gone by when I used windoze.  
> >
> > I really like to use various ploys in writing long filenames that I
> > can remember, usually with multiple suffixes.  In NO OTHER program,
> > in the ten years of using GNU/Linux, has this behavior been
> > encountered.  May I request either assistance and explanation, or,
> > better, to do away with this terrible "bug".
> >
> > Alan Davis
> > aedavis eccomm com
>
> _______________________________________________
> gnumeric-list mailing list
> gnumeric-list gnome org
> http://mail.gnome.org/mailman/listinfo/gnumeric-list


-- 
    A non-free program systematically denies the users the freedom to
    cooperate; it is the basis of an antisocial scheme to dominate
    people.     --Richard M. Stallman