DUEL support

DUEL support

[Sergei Golubchik]

Hi.

Is there an interest in having DUEL support in GDB ?

DUEL is a high-level data exploring language, originally written as a
gdb patch in 1993, for a PhD project.

Apparently, forgotten and abandoned since then, although it made into
dbx on irix (looks like SGI simply applied it, as the patch is public
domain) - that's where I first saw the feature.

I found that patch yesterday, ported to 6.6, fixed a few bugs (added
support for bool, long long, typedefs, references, fixed type parsing,
etc). Looks like working.  I didn't spend too much time on it though,
the patch definitely needs polishing.

Anybody's interested ?

This is what it looks like:

(gdb) dl
Supported DUEL commands:
duel help      - give basic help (shortcut: dl ?)
duel longhelp  - give a longer help (dl ??)
duel examples  - show useful usage examples (dl ex)
duel operators - operators summary (dl ops)
duel aliases   - show current aliases (dl alias)
duel clear     - clear all aliases
duel debug     - toggle duel debug mode
(gdb) dl ex
x[10..20,22,24,40..60]    display x[i] for the selected indexes
x[9..0]                   display x[i] backwards
x[..100] >? 5 <? 10       display x[i] if 5<x[i]<10
x[0..99]=>if(_>5 && _<10) _   same
val[..50].if(is_dx) x else y   val[i].x or val[i].y depending on val[i].is_dx
emp[..50].if(is_m) _      return emp[i] if emp[i].is_m
x[i:=..100]=y[i] ;        assign y[i] to x[i]
x[i:=..100] >? x[i+1]     check if x[i] is not sorted
(x[..100] >? 0)[[2]]      return the 3rd positive x[i]
argv[0..]@0               argv[0] argv[1] .. until first null
emp[0..]@(code==0)        emp[0]..emp[n-1] where emp[n].code==0
head-->next->val          val of each element in a linked list
*head-->next[[20]]        element 20 of list, '*' display struct
w/fields
#/head-->next             count elements on a linked list
#/(head-->next-val>?5)    count those over 5
head-->(next!=?head)      expand cyclic linked list (tail->head)
T mytype x ;              declare var for user defined type (need 'T')
int i ; for(i=0 ;i<5 ..   declare variable, use C construct.
(gdb)

Regards,
Sergei

Re: DUEL support

[Jim Blandy]

Sergei Golubchik <sergii at pisem.net> writes:
> Hi.
>
> Is there an interest in having DUEL support in GDB ?
>
> DUEL is a high-level data exploring language, originally written as a
> gdb patch in 1993, for a PhD project.
>
> Apparently, forgotten and abandoned since then, although it made into
> dbx on irix (looks like SGI simply applied it, as the patch is public
> domain) - that's where I first saw the feature.
>
> I found that patch yesterday, ported to 6.6, fixed a few bugs (added
> support for bool, long long, typedefs, references, fixed type parsing,
> etc). Looks like working.  I didn't spend too much time on it though,
> the patch definitely needs polishing.
>
> Anybody's interested ?

DUEL is insanely cool, but last I knew, its author was very hostile to
the GPL, and refused to assign his changes to the FSF.

Re: DUEL support

[Sergei Golubchik]

Hi!

On Nov 18, Jim Blandy wrote:
> Sergei Golubchik <sergii at pisem.net> writes:
> > Hi.
> >
> > Is there an interest in having DUEL support in GDB ?
> >
> > DUEL is a high-level data exploring language, originally written as a
> > gdb patch in 1993, for a PhD project.
> >
> > Apparently, forgotten and abandoned since then, although it made into
> > dbx on irix (looks like SGI simply applied it, as the patch is public
> > domain) - that's where I first saw the feature.
> >
> > I found that patch yesterday, ported to 6.6, fixed a few bugs (added
> > support for bool, long long, typedefs, references, fixed type parsing,
> > etc). Looks like working.  I didn't spend too much time on it though,
> > the patch definitely needs polishing.
> >
> > Anybody's interested ?
> 
> DUEL is insanely cool, but last I knew, its author was very hostile to
> the GPL, and refused to assign his changes to the FSF.

Yes, according to the "GPL vs. public domain" file in the distribution
he was (though it was in 1993). But why would it matter, he explicitly
put his patch under public domain ?
 
Regards / Mit vielen Grüssen,
Sergei

-- 
   __  ___     ___ ____  __
  /  |/  /_ __/ __/ __ \/ /   Sergei Golubchik <serg@mysql.com>
 / /|_/ / // /\ \/ /_/ / /__  Principal Software Developer
/_/  /_/\_, /___/\___\_\___/  MySQL GmbH, Dachauer Str. 37, D-80335 München
       <___/                  Geschäftsführer: Kaj Arnö - HRB München 162140

Re: DUEL support

[Robert Dewar]

Sergei Golubchik wrote:

> Yes, according to the "GPL vs. public domain" file in the distribution
> he was (though it was in 1993). But why would it matter, he explicitly
> put his patch under public domain ?

One potential problem is that in somme countries, for sure France, it is
not possible for an author to entirely disclaim copyright moral rights,
so putting something in the public dommain can be a legally dubious 
notion. And how do we know the patch is in the public domain even in
the US without a signed document. A notice on a file means nothing
legally.

Re: DUEL support

[Sergei Golubchik]

Hi!

On Nov 19, Robert Dewar wrote:
> Sergei Golubchik wrote:
> 
> >Yes, according to the "GPL vs. public domain" file in the
> >distribution he was (though it was in 1993). But why would it matter,
> >he explicitly put his patch under public domain ?
> 
> One potential problem is that in some countries, for sure France, it
> is not possible for an author to entirely disclaim copyright moral
> rights, so putting something in the public domain can be a legally
> dubious notion.

Not really. An author cannot disclaim all rights, but he can disclaim
material rights, which is enough to re-release a software as GPL.

> And how do we know the patch is in the public domain even in the US
> without a signed document. A notice on a file means nothing legally.

There're signed documents too - I mean publications and PhD thesis where
the author says plain and clear that "Duel is public-domain software".

Regards
Sergei