| David: for your book ? [message #21233] |
Tue, 15 August 2000 00:00  |
Martin Schultz
Messages: 515
Registered: August 1997
|
Senior Member |
|
|
Hi,
Testing IDL's capabilities is always dangerous: you might get more
than you ask for!
What would you expect to get out of the following:
test = [ 'dir1/', 'directory2/', 'another_directory/' ]
print,strmid(test,0,strlen(test)-1)
??
Well, I at least had expected to get the directory names without the
trailing '/'.
But, no!
help, strmid(test,0,strlen(test)-1)
reveals that the result is a 3x3 array!! To print it nicely, use a
format statement:
print,strmid(test,0,strlen(test)-1),format='(3(3(">",A,"< "),/))'
So, to achieve what I had in mind I either have to loop or extract the
matrix diagonale!
Does this make any sense, or is IDL simply overachieving here?
Cheers,
Martin
--
[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[ [[[[[[[
[[ Dr. Martin Schultz Max-Planck-Institut fuer Meteorologie [[
[[ Bundesstr. 55, 20146 Hamburg [[
[[ phone: +49 40 41173-308 [[
[[ fax: +49 40 41173-298 [[
[[ martin.schultz@dkrz.de [[
[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[ [[[[[[[
|
|
|
|
| Re: David: for your book ? [message #21276 is a reply to message #21233] |
Wed, 16 August 2000 00:00  |
landsman
Messages: 93
Registered: August 1991
|
Member |
|
|
In article <399937D2.8C0A52A2@dkrz.de>,
Martin Schultz <martin.schultz@dkrz.de> wrote:
>
> What would you expect to get out of the following:
>
> test = [ 'dir1/', 'directory2/', 'another_directory/' ]
> print,strmid(test,0,strlen(test)-1)
>
> ??
>
> Well, I at least had expected to get the directory names without the
> trailing '/'.
> But, no!
> help, strmid(test,0,strlen(test)-1)
> reveals that the result is a 3x3 array!!
When IDL added vector capabilities to STRMID in V5.3, they allowed for
both (1) the extraction of multiple substrings from each element of a
string, and (2) having the first character position and substring length
differ for each element of a string array. Unfortunately, adding all
this power has made the syntax awkward. According to the documentation
"If First_Character or Length is an array, the size of their first
dimension determines how many substrings are extracted from each element
of <expression>"
So the solution is to REFORM the string length (or position) array to
two dimensions:
IDL> test = [ 'dir1/', 'directory2/', 'another_directory/']
IDL> print,strmid(test,0,reform( strlen(test)-1, 1, 3) )
===>dir1 directory2 another_directory
By reforming the string length array into a 1 x 3 size, STRMID now knows
to extract only 1 substring from each element of test.
A simple program to extract a string array up to a specified character
using the vector capabilities of STRMID is available at
http://idlastro.gsfc.nasa.gov/ftp/v53/gettok.pro
--Wayne Landsman landsman@mpb.gsfc.nasa.gov
Sent via Deja.com http://www.deja.com/
Before you buy.
|
|
|
|
| Re: David: for your book ? [message #21303 is a reply to message #21233] |
Wed, 16 August 2000 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Martin Schultz (martin.schultz@dkrz.de) writes:
> BTW: any one out there who is still using earlier versions (prior to
> 5.2)?
Jeeze Louise, man! Have you checked with your systems administrator?
You are about to have a denial of service attack from all the
e-mail about to pile up in your account! :-)
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: David: for your book ? [message #21306 is a reply to message #21233] |
Wed, 16 August 2000 00:00 |
Martin Schultz
Messages: 515
Registered: August 1997
|
Senior Member |
|
|
Craig Markwardt wrote:
>
> Martin Schultz <martin.schultz@dkrz.de> writes:
>> As an aside, I recently found another problem in terms of backward
>> compability (I am
>> sure something similar was already mentioned in this newsgroup, but this
>> was the first
>> time I had this problem):
>> I tried to write a routine that takes advantage of the new StrMatch
>> and StRegex
>> functions in 5.3. In order to ensure backward compatibility, I enclosed
>> these function calls
>> with
>> IF !Version.Release GE 5.3 THEN BEGIN
>> ENDIF
>>
>> However, a colleague tells me: no chance! The program will report a
>> syntax error when run in
>> IDL 5.2! So, the problem is that IDL looks for a function of this name
>> during compile stage
>> and doesn't care if it will ever be used.
> ...
>
> There is a way to do this without defining an external function, by
> using FORWARD_FUNCTION properly:
>
> FORWARD_FUNCTION stregex, strmatch
> IF !Version.Release GE 5.3 THEN BEGIN
> foo = stregex(bar)
> sm = strmatch(eg)
> ENDIF
>
> This will make sure that IDL will always recognize strmatch & stregex
> as functions during the compilation stage, and there will be no syntax
> errors in any IDL version. It is harmless in IDL 5.3. If you really
> want the warning message, then by all means go ahead and make your own
> strmatch definitions et al.
>
> Have fun,
> Craig
Thanks, Craig.
I didn't believe that this would cure it (I had thought it would just
change the
error message), but you are right, it works. So, maybe some people can
still live
with IDL 5.2 after all.
BTW: any one out there who is still using earlier versions (prior to
5.2)?
Maybe I shouldn't ask because I fear this will swamp the newsgroup -- so
please
reply only by email, subject "old idl", content '!version = ' then I
will
collect these messages for a while and report a summary later (after a
week
or two). I assure you that these messages will be kept strictly
confidental and
will be deleted after processing the version information.
Cheers,
Martin
--
[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[ [[[[[[[
[[ Dr. Martin Schultz Max-Planck-Institut fuer Meteorologie [[
[[ Bundesstr. 55, 20146 Hamburg [[
[[ phone: +49 40 41173-308 [[
[[ fax: +49 40 41173-298 [[
[[ martin.schultz@dkrz.de [[
[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[ [[[[[[[
|
|
|
|
| Re: David: for your book ? [message #21312 is a reply to message #21233] |
Wed, 16 August 2000 00:00 |
R.Bauer
Messages: 1424
Registered: November 1998
|
Senior Member |
|
|
Craig Markwardt wrote:
>
> Martin Schultz <martin.schultz@dkrz.de> writes:
>> As an aside, I recently found another problem in terms of backward
>> compability (I am
>> sure something similar was already mentioned in this newsgroup, but this
>> was the first
>> time I had this problem):
>> I tried to write a routine that takes advantage of the new StrMatch
>> and StRegex
>> functions in 5.3. In order to ensure backward compatibility, I enclosed
>> these function calls
>> with
>> IF !Version.Release GE 5.3 THEN BEGIN
>> ENDIF
>>
>> However, a colleague tells me: no chance! The program will report a
>> syntax error when run in
>> IDL 5.2! So, the problem is that IDL looks for a function of this name
>> during compile stage
>> and doesn't care if it will ever be used.
> ...
>
> There is a way to do this without defining an external function, by
> using FORWARD_FUNCTION properly:
>
> FORWARD_FUNCTION stregex, strmatch
> IF !Version.Release GE 5.3 THEN BEGIN
> foo = stregex(bar)
> sm = strmatch(eg)
> ENDIF
>
> This will make sure that IDL will always recognize strmatch & stregex
> as functions during the compilation stage, and there will be no syntax
> errors in any IDL version. It is harmless in IDL 5.3. If you really
> want the warning message, then by all means go ahead and make your own
> strmatch definitions et al.
>
> Have fun,
> Craig
>
> --
Don't write it yourself
Take the strmatch routine from the deutsch library
http://www.astro.washington.edu/deutsch/idl/htmlhelp/slibrar y24.html
Reimar
|
|
|
|
| Re: David: for your book ? [message #21322 is a reply to message #21233] |
Tue, 15 August 2000 00:00 |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
Martin Schultz <martin.schultz@dkrz.de> writes:
> As an aside, I recently found another problem in terms of backward
> compability (I am
> sure something similar was already mentioned in this newsgroup, but this
> was the first
> time I had this problem):
> I tried to write a routine that takes advantage of the new StrMatch
> and StRegex
> functions in 5.3. In order to ensure backward compatibility, I enclosed
> these function calls
> with
> IF !Version.Release GE 5.3 THEN BEGIN
> ENDIF
>
> However, a colleague tells me: no chance! The program will report a
> syntax error when run in
> IDL 5.2! So, the problem is that IDL looks for a function of this name
> during compile stage
> and doesn't care if it will ever be used.
...
There is a way to do this without defining an external function, by
using FORWARD_FUNCTION properly:
FORWARD_FUNCTION stregex, strmatch
IF !Version.Release GE 5.3 THEN BEGIN
foo = stregex(bar)
sm = strmatch(eg)
ENDIF
This will make sure that IDL will always recognize strmatch & stregex
as functions during the compilation stage, and there will be no syntax
errors in any IDL version. It is harmless in IDL 5.3. If you really
want the warning message, then by all means go ahead and make your own
strmatch definitions et al.
Have fun,
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@cow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
| Re: David: for your book ? [message #21323 is a reply to message #21233] |
Tue, 15 August 2000 00:00 |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Martin Schultz (martin.schultz@dkrz.de) writes:
> So, the problem is that IDL looks for a function of this name
> during compile stage and doesn't care if it will ever be used.
Yes. I thought I was going crazy recently, because I don't
remember IDL ever complaining about not being able to
find a function until run-time. But clearly it is
looking for this functions during compile time.
As you suggest, this makes it awfully hard to
write version-independent code. Perhaps it is
done as an impetus to upgrade to the latest version
of IDL. :-)
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: David: for your book ? [message #21324 is a reply to message #21233] |
Tue, 15 August 2000 00:00 |
Martin Schultz
Messages: 515
Registered: August 1997
|
Senior Member |
|
|
Craig Markwardt wrote:
>
> Martin Schultz <martin.schultz@dkrz.de> writes:
>> Hi,
>>
>> Testing IDL's capabilities is always dangerous: you might get more
>> than you ask for!
>>
>> What would you expect to get out of the following:
>>
>> test = [ 'dir1/', 'directory2/', 'another_directory/' ]
>> print,strmid(test,0,strlen(test)-1)
>
> Hmm, I got an error message!
>
> IDL> test = [ 'dir1/', 'directory2/', 'another_directory/' ]
> IDL> print,strmid(test,0,strlen(test)-1)
> % STRMID: Expression must be a scalar in this context: <LONG Array>.
> % Execution halted at: $MAIN$
>
> IDL> print, !version
> { alpha OSF unix 5.2 Oct 30 1998}
>
> Is this something new in IDL 5.3?
>
Apparently so! What I get as a result is a 3x3 matrix containing in the
first column
all directory namews truncated to a length of 4, in the second column
10, and in the
third column StrLen(test[2]-1). But I am glad you point this out.
As an aside, I recently found another problem in terms of backward
compability (I am
sure something similar was already mentioned in this newsgroup, but this
was the first
time I had this problem):
I tried to write a routine that takes advantage of the new StrMatch
and StRegex
functions in 5.3. In order to ensure backward compatibility, I enclosed
these function calls
with
IF !Version.Release GE 5.3 THEN BEGIN
ENDIF
However, a colleague tells me: no chance! The program will report a
syntax error when run in
IDL 5.2! So, the problem is that IDL looks for a function of this name
during compile stage
and doesn't care if it will ever be used. Only help is to write two
little dummy functions
named stregex.pro and strmatch.pro which do nothing but print a warning
message. And so I
did. Works nicely, and you don't even have to test for the release
number because these two
functions are internal in 5.3 and hence they are discovered before
anything on disk can be
found.
Just a little bedtime story ;-)
Cheers,
Martin
--
[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[ [[[[[[[
[[ Dr. Martin Schultz Max-Planck-Institut fuer Meteorologie [[
[[ Bundesstr. 55, 20146 Hamburg [[
[[ phone: +49 40 41173-308 [[
[[ fax: +49 40 41173-298 [[
[[ martin.schultz@dkrz.de [[
[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[[ [[[[[[[
|
|
|
|
| Re: David: for your book ? [message #21325 is a reply to message #21233] |
Tue, 15 August 2000 00:00 |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
Martin Schultz <martin.schultz@dkrz.de> writes:
> Hi,
>
> Testing IDL's capabilities is always dangerous: you might get more
> than you ask for!
>
> What would you expect to get out of the following:
>
> test = [ 'dir1/', 'directory2/', 'another_directory/' ]
> print,strmid(test,0,strlen(test)-1)
Hmm, I got an error message!
IDL> test = [ 'dir1/', 'directory2/', 'another_directory/' ]
IDL> print,strmid(test,0,strlen(test)-1)
% STRMID: Expression must be a scalar in this context: <LONG Array[3]>.
% Execution halted at: $MAIN$
IDL> print, !version
{ alpha OSF unix 5.2 Oct 30 1998}
Is this something new in IDL 5.3?
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@cow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
