| Arrays: strange feature ?! [message #19877] |
Wed, 03 May 2000 00:00  |
Nicolas Decoster
Messages: 34
Registered: March 2000
|
Member |
|
|
Hi.
More I use IDL, more I found strange features. The one I want to talk
about is the way you create multidimensional arrays. Look the following
IDL commands sequence:
IDL> print, [[1, 2], [3, 4]]
1 2
3 4
IDL> a = [1, 2]
IDL> b = [3, 4]
IDL> print, [a, b]
1 2 3 4
Why the same notation has two meanings? Why for the first print the
[something, something_else] notation creates an array of arrays, and for
the second one the same notation concatenates the arrays? I can't find
the logic behind this.
In fact I stopped on this because I want to create an array of images
using something like that: [image1, image2, image3]. What is the way to
do this without creating a new temporary variable array? Or is there a
way to create easily a list of anything.
I want to build a procedure that takes a list of images and displays
them, something like that:
myDisplayProcedure, myImagesList, parameter1, parameter2
where myImagesList is something like that:
{image1, image2, image3}
or anything else that can be create on the command line.
I hope I am clear enough.
Any comments? Any suggestions?
Nicolas.
--
T�l. : 00 (33) 5 62 88 11 16
Fax : 00 (33) 5 62 88 11 12
Nicolas.Decoster@Noveltis.fr
Noveltis
Parc Technologique du Canal
2, avenue de l'Europe
31520 Ramonville Saint Agne - France
|
|
|
|
|
|
| Re: Arrays: strange feature ?! [message #19969 is a reply to message #19877] |
Thu, 04 May 2000 00:00  |
Luca Fini
Messages: 2
Registered: March 2000
|
Junior Member |
|
|
On Wed, 3 May 2000, Nicolas Decoster wrote:
> More I use IDL, more I found strange features.
me too.
> The one I want to talk
> about is the way you create multidimensional arrays. Look the following
> IDL commands sequence:
>
> IDL> print, [[1, 2], [3, 4]]
> 1 2
> 3 4
> IDL> a = [1, 2]
> IDL> b = [3, 4]
> IDL> print, [a, b]
> 1 2 3 4
>
> Why the same notation has two meanings?
IDL is clearly a language grown by piling up things with no previous
design. Its syntax is crazy and some characteristics are very annoying.
It is very nice for quick and dirty programming, but reallya mess for
large scale projects.
l.f.
------------------------------------------------------------ --------------
-- ) Luca Fini Tel: +39 055 2752 307
___ |\ Osservatorio Astrofisico di Arcetri Fax: +39 055 2752 292
/ | | |-_ L.go E.Fermi, 5 +-----------------------------------------
(___|___//___) 50125 Firenze / WWW: http://www.arcetri.astro.it/~lfini
(_) (_) Italia / e-mail: lfini@arcetri.astro.it
-----------------------------+------------------------------ --------------
In a world without fences - who needs GATES?
|
|
|
|
| Re: Arrays: strange feature ?! [message #19972 is a reply to message #19877] |
Wed, 03 May 2000 00:00 |
rshill
Messages: 2
Registered: May 2000
|
Junior Member |
|
|
In article <3910324B.CDDD96FB@pet.mpin-koeln.mpg.de>,
Alex Schuster <alex@pet.mpin-koeln.mpg.de> wrote:
> You can create the array like this:
>
> image = [ [image1], [image2], [image3] ]
Not quite. In fact, the notation is remarkably flexible.
IDL> i1 = bytarr(10,20)
IDL> i2 = i1
IDL> i3 = i1
IDL> t = [i1,i2,i3]
IDL> help,t
T BYTE = Array[30, 20]
IDL> t = [[i1],[i2],[i3]]
IDL> help,t
T BYTE = Array[10, 60]
IDL> t = [[[i1]],[[i2]],[[i3]]]
IDL> help,t
T BYTE = Array[10, 20, 3]
IDL> t2 = [t,t]
IDL> help,t2
T2 BYTE = Array[20, 20, 3]
IDL> t2 = [[t],[t]]
IDL> help,t2
T2 BYTE = Array[10, 40, 3]
IDL> t2 = [[[t]],[[t]]]
IDL> help,t2
T2 BYTE = Array[10, 20, 6]
IDL> t2 = [[[[t]]],[[[t]]]]
t2 = [[[[t]]],[[[t]]]]
^
% Only eight levels of variable concatenation are allowed.
IDL>
As you can see, there is a limit. For three or fewer
dimensions, the rule seems to be that if the number of
outermost brackets in a row is 1, the arrays are
concatenated in X; if 2, they are concatenated in Y; and
if 3, they are concatenated in Z. Consider this further
example:
IDL> x = indgen(10)
IDL> help, [[[x]],[[x]],[[x]]]
<Expression> INT = Array[10, 1, 3]
YMMV, but I like it, myself.
Bob Hill
--
Robert S. Hill
Raytheon ITSS Code 681, NASA/GSFC, Greenbelt, MD 20771
Sent via Deja.com http://www.deja.com/
Before you buy.
|
|
|
|
|
|
| Re: Arrays: strange feature ?! [message #20002 is a reply to message #19877] |
Fri, 12 May 2000 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
I wrote the other day:
> Some of us find in IDL the same subtle intelligence
> that pervades the Universe. If it seems crazy to us,
> it is because we haven't evolved our consciousness
> enough to perceive the underlying plan.
And I can't believe the number of unsolicited e-mails
I got from the skeptics. You guys need to do a little
reading to open your minds. Here is a good web page
to get you in the proper mood for understanding the
journey...
http://www.thegreatillusion.com/
Cheers,
David
P.S. Speaking of strange features, I ran into a little one
that set me back for a couple of minutes in an IDL course
I was teaching last week. I was trying something new
(normally my spontaneous flights of fancy are well rehearsed,
but this one wasn't, although I really didn't expect any
problems with it), which was my first mistake.
I decided to write a widget program where we could click
on an image and print the image value back in a CW_Field
widget. (Normally, I disdain CW_Field widgets because the
text field doesn't look editable on PC platforms, but this
time that was exactly the effect I was going for.)
In any case, when we tried to put the image value into
the CW_Field widget:
Widget_Control, info.fieldID, Set_Value=imageValue
nothing would show up on the display! (No, I didn't have
the Modal keyword set, Dave.) This was very strange. I've
only written programs like this about a million times,
and the value *always* shows up where it is suppose to.
What in the world? Bug in IDL 5.3.1? Corrupted installation?
Coyote out of his teepee and loose in Milwaukee?
No, it turned out that the image value was an integer
value (which is what we had told CW_Field to expect),
but somehow in doing the conversion from integer to
string to get the value in the the text widget, the
integer was getting converted to blank characters.
I haven't had time to investigate the details of this.
(And after I thought about it I understood why it might
happen with bytes, but I still don't know why it
happened with integers.)
In any case, the problem was fixed by converting
the image value to a float before we tried to put it
into CW_Field. But it gave me a couple more gray hairs,
which I *certainly* didn't need. :-(
--
David Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438 E-Mail: davidf@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Arrays: strange feature ?! [message #20015 is a reply to message #19877] |
Fri, 12 May 2000 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Nicolas Decoster (Nicolas.Decoster@Noveltis.fr) writes:
> Just to be sure you understood my suggestion: the tip I'm talking about
> is the entrepenarial one, not the array-subscript one. But, of course, I
> guess this tip can't appear on a site which is part of the... marketing.
Well, there is marketing, and then there is marketing. Let's
just say we all have to make a living somehow. :-)
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438 E-Mail: davidf@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Arrays: strange feature ?! [message #20019 is a reply to message #19877] |
Fri, 12 May 2000 00:00 |
Nicolas Decoster
Messages: 34
Registered: March 2000
|
Member |
|
|
David Fanning wrote:
>
> Nicolas Decoster (Nicolas.Decoster@Noveltis.fr) writes:
>
>> Well, in fact I am in consulting business. And I was before I had to
>> meet IDL. IDL come to me as a tool to do my consulting business. I guess
>> the people I work for do not mind I know well the quote-double
>> quote-octal thing of IDL. They do not ever mind I use IDL or not. They
>> want me to correctly analyse and treat their data. It is my business to
>> do it not too slowly. But you are right, perhaps I can convince them IDL
>> _is_ the ultimate tool to treat their data, and they absolutly need idl
>> scripts and applications to handle their data efficiently at home and to
>> exploit my results, and IDL is very powerful but require quite complex
>> programming but I know the holy words (i.e. the "quote-double quote"
>> thing) to built such complex-bu-powerful-applications... hum... Then ?
>> $$$ Kling ! $$$ Money !!!
>>
>> Yeaaah !
>
> Now you've got the idea. It's all a matter of marketing. :-)
>
>> Thanks for the tip, David. I suggest you replace the tip of the year on
>> your site by this one. :-)
>
> Yeah, I have been looking for something new. I haven't been
> asked a color question in ages. I'm writing the color section
> for the 2nd Edition of the book now, but maybe I should do us
> all a favor and just delete this section and save room
> for more array subscribing features. :-)
Just to be sure you understood my suggestion: the tip I'm talking about
is the entrepenarial one, not the array-subscript one. But, of course, I
guess this tip can't appear on a site which is part of the... marketing.
Later.
Nicolas.
--
T�l. : 00 (33) 5 62 88 11 16
Fax : 00 (33) 5 62 88 11 12
Nicolas.Decoster@Noveltis.fr
Noveltis
Parc Technologique du Canal
2, avenue de l'Europe
31520 Ramonville Saint Agne - France
|
|
|
|
| Re: Arrays: strange feature ?! [message #20020 is a reply to message #19877] |
Fri, 12 May 2000 00:00 |
dmarshall
Messages: 23
Registered: December 1997
|
Junior Member |
|
|
Ah, yes.
Which is why progammers get Christmas and Halloween mixed up since
oct31=dec25
(Christmas falls on December 25th and Halloween on October 31)
Dave.
In article <391A76A6.8C74B2B0@mpipsykl.mpg.de>, Benno Puetz <puetz@mpipsykl.mpg.de> writes:
> marc wrote:
>
>> But another more funny 'feature' is this:
>>
>> IDL> print,'abc'
>> abc
>> IDL> print,"abc"
>> abc
>> IDL> print,'777'
>> 777
>> IDL> print,"777"
>>
>> print,"777"
>> ^
>> % Syntax error.
>>
>> Cheers,
>> :-) marc
>
> It's a real feature
>
> IDL> print,"777
> 511
>
> (check the online help for 'octal')
>
> --
> Benno Puetz
> Kernspintomographie
> Max-Planck-Institut f. Psychiatrie Tel.: +49-89-30622-413
> Kraepelinstr. 10 Fax : +49-89-30622-520
> 80804 Muenchen, Germany
>
>
>
|
|
|
|
| Re: Arrays: strange feature ?! [message #20022 is a reply to message #19877] |
Fri, 12 May 2000 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Nicolas Decoster (Nicolas.Decoster@Noveltis.fr) writes:
> Well, in fact I am in consulting business. And I was before I had to
> meet IDL. IDL come to me as a tool to do my consulting business. I guess
> the people I work for do not mind I know well the quote-double
> quote-octal thing of IDL. They do not ever mind I use IDL or not. They
> want me to correctly analyse and treat their data. It is my business to
> do it not too slowly. But you are right, perhaps I can convince them IDL
> _is_ the ultimate tool to treat their data, and they absolutly need idl
> scripts and applications to handle their data efficiently at home and to
> exploit my results, and IDL is very powerful but require quite complex
> programming but I know the holy words (i.e. the "quote-double quote"
> thing) to built such complex-bu-powerful-applications... hum... Then ?
> $$$ Kling ! $$$ Money !!!
>
> Yeaaah !
Now you've got the idea. It's all a matter of marketing. :-)
> Thanks for the tip, David. I suggest you replace the tip of the year on
> your site by this one. :-)
Yeah, I have been looking for something new. I haven't been
asked a color question in ages. I'm writing the color section
for the 2nd Edition of the book now, but maybe I should do us
all a favor and just delete this section and save room
for more array subscribing features. :-)
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438 E-Mail: davidf@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Arrays: strange feature ?! [message #20024 is a reply to message #19877] |
Fri, 12 May 2000 00:00 |
Nicolas Decoster
Messages: 34
Registered: March 2000
|
Member |
|
|
David Fanning wrote:
>
> Nicolas Decoster (Nicolas.Decoster@Noveltis.fr) writes:
>
>> It seems that IDL likes the one-notation-two-actions thing. I must say
>> that such a thing is confusing me. I hope I will not loose too much time
>> in trying to understand all the IDL's "strange feature".
>
> You are focused on the wrong thing. Instead of
> worrying about "losing too much time", you should
> see this as an opportunity to learn something no one
> else is going to be able to figure out, thereby making
> it possible for you to charge big bucks for your
> specialized knowledge. :-)
>
> Cheers,
>
> David
>
> P.S. Let's just say IDL makes it real easy for someone
> to get into the consulting business if they have that
> entrepreneurial state of mind.
>
Well, in fact I am in consulting business. And I was before I had to
meet IDL. IDL come to me as a tool to do my consulting business. I guess
the people I work for do not mind I know well the quote-double
quote-octal thing of IDL. They do not ever mind I use IDL or not. They
want me to correctly analyse and treat their data. It is my business to
do it not too slowly. But you are right, perhaps I can convince them IDL
_is_ the ultimate tool to treat their data, and they absolutly need idl
scripts and applications to handle their data efficiently at home and to
exploit my results, and IDL is very powerful but require quite complex
programming but I know the holy words (i.e. the "quote-double quote"
thing) to built such complex-bu-powerful-applications... hum... Then ?
$$$ Kling ! $$$ Money !!!
Yeaaah !
Thanks for the tip, David. I suggest you replace the tip of the year on
your site by this one. :-)
Cheers,
Nicolas.
--
T�l. : 00 (33) 5 62 88 11 16
Fax : 00 (33) 5 62 88 11 12
Nicolas.Decoster@Noveltis.fr
Noveltis
Parc Technologique du Canal
2, avenue de l'Europe
31520 Ramonville Saint Agne - France
|
|
|
|
| Re: Arrays: strange feature ?! [message #20026 is a reply to message #19877] |
Fri, 12 May 2000 00:00 |
marc schellens[1]
Messages: 183
Registered: January 2000
|
Senior Member |
|
|
Benno Puetz wrote:
>
> marc wrote:
>
>> But another more funny 'feature' is this:
>>
>> IDL> print,'abc'
>> abc
>> IDL> print,"abc"
>> abc
>> IDL> print,'777'
>> 777
>> IDL> print,"777"
>>
>> print,"777"
>> ^
>> % Syntax error.
>>
>> Cheers,
>> :-) marc
>
> It's a real feature
>
> IDL> print,"777
> 511
>
> (check the online help for 'octal')
I know, I know...
nevertheless found it funny
one more thing:
IDL> print,'70
70
IDL> print,"70
56
(strings don't need trailing delimiters)
:-) marc
|
|
|
|
| Re: Arrays: strange feature ?! [message #20044 is a reply to message #19877] |
Thu, 11 May 2000 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Nicolas Decoster (Nicolas.Decoster@Noveltis.fr) writes:
> It seems that IDL likes the one-notation-two-actions thing. I must say
> that such a thing is confusing me. I hope I will not loose too much time
> in trying to understand all the IDL's "strange feature".
You are focused on the wrong thing. Instead of
worrying about "losing too much time", you should
see this as an opportunity to learn something no one
else is going to be able to figure out, thereby making
it possible for you to charge big bucks for your
specialized knowledge. :-)
Cheers,
David
P.S. Let's just say IDL makes it real easy for someone
to get into the consulting business if they have that
entrepreneurial state of mind.
--
David Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438 E-Mail: davidf@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Arrays: strange feature ?! [message #20045 is a reply to message #19877] |
Thu, 11 May 2000 00:00 |
Nicolas Decoster
Messages: 34
Registered: March 2000
|
Member |
|
|
Benno Puetz wrote:
>
> marc wrote:
>
>> But another more funny 'feature' is this:
>>
>> IDL> print,'abc'
>> abc
>> IDL> print,"abc"
>> abc
>> IDL> print,'777'
>> 777
>> IDL> print,"777"
>>
>> print,"777"
>> ^
>> % Syntax error.
>>
>> Cheers,
>> :-) marc
>
> It's a real feature
Let me connect these words to the subject:
"It's a real _strange_ feature"
These "strange feature" is much more confusing with the following lines:
IDL> print, "877"
877
IDL> print, "777"
print, "777"
^
% Syntax error.
It seems that IDL likes the one-notation-two-actions thing. I must say
that such a thing is confusing me. I hope I will not loose too much time
in trying to understand all the IDL's "strange feature".
Later.
Nicolas.
--
T�l. : 00 (33) 5 62 88 11 16
Fax : 00 (33) 5 62 88 11 12
Nicolas.Decoster@Noveltis.fr
Noveltis
Parc Technologique du Canal
2, avenue de l'Europe
31520 Ramonville Saint Agne - France
|
|
|
|
| Re: Arrays: strange feature ?! [message #20054 is a reply to message #19877] |
Thu, 11 May 2000 00:00 |
Benno Puetz
Messages: 16
Registered: March 2000
|
Junior Member |
|
|
marc wrote:
> But another more funny 'feature' is this:
>
> IDL> print,'abc'
> abc
> IDL> print,"abc"
> abc
> IDL> print,'777'
> 777
> IDL> print,"777"
>
> print,"777"
> ^
> % Syntax error.
>
> Cheers,
> :-) marc
It's a real feature
IDL> print,"777
511
(check the online help for 'octal')
--
Benno Puetz
Kernspintomographie
Max-Planck-Institut f. Psychiatrie Tel.: +49-89-30622-413
Kraepelinstr. 10 Fax : +49-89-30622-520
80804 Muenchen, Germany
|
|
|
|
| Re: Arrays: strange feature ?! [message #20057 is a reply to message #19877] |
Thu, 11 May 2000 00:00 |
marc schellens[1]
Messages: 183
Registered: January 2000
|
Senior Member |
|
|
David Fanning wrote:
>
> Luca Fini (lfini@arcetri.astro.it) writes:
>
>> IDL is clearly a language grown by piling up things with no previous
>> design. Its syntax is crazy and some characteristics are very annoying.
>
> Some of us find in IDL the same subtle intelligence
> that pervades the Universe. If it seems crazy to us,
> it is because we haven't evolved our consciousness
> enough to perceive the underlying plan.
>
I agree, but the universe perhaps has to be this way, while a
programming language *could* be for sure more clearly structured that
IDL.
Even this array stuff Nicolas complains about isn't a bug but a feature
(i.e. the syntax of IDL) like Alex pointed out.
But another more funny 'feature' is this:
IDL> print,'abc'
abc
IDL> print,"abc"
abc
IDL> print,'777'
777
IDL> print,"777"
print,"777"
^
% Syntax error.
Cheers,
:-) marc
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
