Re: [OctDev] Re: struct ("a",{1},"b",{1,2})

[Etienne G. <et...@cs...>]

  Hi all,

  maybe we could have an ostruct() with the old behavior? If stinckin'
matlab doesn't do structs that take lists as values, should we be
deprived of them? For now I rm'd struct.cc and use good'ol struct.m
that behaves as I want.

  Etienne

On Tue, Mar 30, 2004 at 07:36:03PM -0500, Paul Kienzle wrote:
# Yes, I did do a number of compatibility changes.   struct with these
# changes is now part of octave.
# 
# Sorry for breaking things...
# 
# Does the new interface do everything you need?  Or is there
# some way we can extend it to do what you need?
# 
# I don't remember why I didn't enable the obvious behaviour of
# struct() creating an empty struct.
# 
# - Paul
# 
# On Mar 30, 2004, at 3:26 PM, Etienne Grossmann wrote:
# 
# >
# >  Hi again, sorry to answer to myself. I guess the effect I described
# >in my previous mail are due to an attempt to imitate the other brand,
# >which behaves like:
# >======================================================================
# >>>a = struct('x',{1},'y',{1,2})
# >
# >a =
# >1x2 struct array with fields:
# >    x
# >    y
# >
# >>>a.x
# >
# >ans =
# >     1
# >ans =
# >     1
# >
# >>>a.y
# >
# >ans =
# >     1
# >ans =
# >     2
# >======================================================================
# >
# >  Hmm. So no more structs taking lists as values? I mean 'cells as
# >values', since lists seem to be doomed?
# >
# >  Etienne
# >
# >On Tue, Mar 30, 2004 at 02:51:12PM -0500, Etienne Grossmann wrote:
# >#
# >#   Hi all,
# >#
# ># I use, octave 2.1.53, cvs octave-forge, so my prob could be due to 
# >not
# ># using 2.1.57.
# >#
# >#   Is that the desired behavior of struct()? Are those bugs introduced
# ># between r 1.2 and r 1.3 of struct.cc?
# >#
# ># 
# >======================================================================
# >#   octave:3> struct ("a",{1},"b",{1,2})
# >#   ans =
# >#   {  a =
# >#       (,
# >#         [1] = 1
# >#         [2] = 1
# >#       ,)
# >#
# >#     b =
# >#       (,
# >#         [1] = 1
# >#         [2] = 2
# >#       ,)
# >#   }
# >#
# >#   octave:4> struct ("a",{1},"b",{1,2},"c",{})
# >#   error: struct: dimensions of parameter 4 do not match those of 
# >parameter 6
# >#   octave:4> struct ()
# >#
# >#   *** struct:
# >#
# >#   struct('field',value,'field',value,...)
# >#
# >#   Create a structure and initialize its value.
# >#
# >#   struct('field',{values},'field',{values},...)
# >#
# >#   Create a structure array and initialize its values.  The dimensions
# >#   of each array of values must match.  Singleton cells and non-cell 
# >values
# >#   are repeated so that they fill the entire array.
# >#
# >#   struct('field',{},'field',{},...)
# >#
# ># Create an empty structure array.
# ># 
# >======================================================================
# >#
# >#   Once upon a time, struct() returned an empty struct, without need 
# >to
# ># do struct([]). No field-value pair -> empty struct, right?
# >#
# >#   I don't recall that there's anything wrong with having struct 
# >values
# ># being lists, my code has relied on this for some time now.
# >#
# >#   Any tips welcome,
# >#
# >#   Etienne
# >#
# ># --
# ># Etienne Grossmann ------ http://www.cs.uky.edu/~etienne
# >
# >-- 
# >Etienne Grossmann ------ http://www.cs.uky.edu/~etienne
# >
# >
# >-------------------------------------------------------
# >This SF.Net email is sponsored by: IBM Linux Tutorials
# >Free Linux tutorial presented by Daniel Robbins, President and CEO of
# >GenToo technologies. Learn everything from fundamentals to system
# >administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
# >_______________________________________________
# >Octave-dev mailing list
# >Oct...@li...
# >https://lists.sourceforge.net/lists/listinfo/octave-dev
# >
# 

-- 
Etienne Grossmann ------ http://www.cs.uky.edu/~etienne