| what am I doing wrong? [message #3602] |
Thu, 16 February 1995 16:09  |
babaka
Messages: 3
Registered: February 1995
|
Junior Member |
|
|
Dear Netters,
I'm trying to plot the funtion:
f(x)=x*log(x/w)-x (x>0,w>0)
using PV~WAVE.
I know that this function has a minimum at x=w.
To see this, note that
f`(x)=log(x/w)
Setting f`(x)=0 gives x=w. On the other hand:
f``(x)=1/x > 0 for all x
Therefore x=w is the minimum.
THE PROBLEM IS that when I plot this function in PV~WAVE
the minimum appears at another point.
I'm using the following set of commands:
WAVE> w=5.0
WAVE> x=INDGEN(400)
WAVE> x=x+0.1 + 0.01 ; x goes from 0.01 to 40.01
WAVE> fx=x*ALOG10(x/w)-x
WAVE> PLOT,x,fx
I would appreciate if you would tell me what I'm doing
wrong?
Regards,
Babak
--
Babak A. Ardekani babaka@atom.ansto.gov.au
Dept. of Nuclear Medicine phone: +61-2-516-8011
Royal Prince Alfred Hospital fax: +61-2-550-5172
Missinden Rd., Camperdown, NSW 2050 AUSTRALIA
|
|
|
|
| Re: what am I doing wrong? [message #13028 is a reply to message #3602] |
Tue, 06 October 1998 00:00  |
thompson
Messages: 584
Registered: August 1991
|
Senior Member |
|
|
Bruce Bowler <bbowler@bigelow.org> writes:
> William Thompson wrote:
>>
>> Bruce Bowler <bbowler@bigelow.org> writes:
>>
> (stuff deleted)
>>> if I run it as follows...
>>
>>> # idl
>>> idl> .run 6953
>>
>>> it works fine...
>>
>>> if I run it like
>>
>>> # idl 6953
>>
>>> the results are
>>
>> (stuff deleted)
>>
>>> unfortunately, because of the way I need to interact with some other
>>> stuff, I really need to run it the second way. What can I do to make it
>>> work like the first way? I don't think precompiling will work either as
>>> I generate the file "on the fly".
>>
>> If you're using Unix, then what you need to do is to create a file like the
>> following:
>>
>> #!/bin/csh
>> setenv MY_IDL_SCRIPT $argv
>> idl << done
>> .run $MY_IDL_SCRIPT
>> done
>>
>> Use "chmod +x" to make the file executable, and then you can use this to
>> execute whatever IDL script you want.
> Unfortunately, as I said before, that won't work in my case. I am
> passing this to a third party package that invokes it as
> idl xxx
> I have no control over that.
There are a couple of ways you can accomplish this. One way, as already
pointed out by Stein Vidar, is to end each line within a loop with the
characters "& $", so that IDL treats it all as one line. As I recall, this
works unless the loop is very large.
Another way is to use two procedure files. One file will contain the
main-level program that you wish to run. For example, let's say you save it in
a file called myprogram.idl. The second file will then contain the single line
.run myprogram
You can then pass the name of this second file to IDL. For example, if the
second file is named run_myprogram.pro, then the statement
idl run_myprogram
will execute the main-level procedure myprogram. It would be just the same as
if you had typed ".run myprogram" at the command prompt.
William Thompson
|
|
|
|
| Re: what am I doing wrong? [message #13029 is a reply to message #3602] |
Tue, 06 October 1998 00:00 |
steinhh
Messages: 260
Registered: June 1994
|
Senior Member |
|
|
In article <361A08FE.D4EEB7A1@bigelow.org> Bruce Bowler
<bbowler@bigelow.org> writes:
> Unfortunately, as I said before, that won't work in my case. I am
> passing this to a third party package that invokes it as
>
> idl xxx
>
> I have no control over that.
Then you have to put "&$" on the end of each line in the
IF .. THEN BEGIN ... END block. This causes the lines
to be interpreted as if they were entered as a single
line, with "&" inbetween each statement.
Regards,
Stein Vidar
|
|
|
|
| Re: what am I doing wrong? [message #13031 is a reply to message #3602] |
Tue, 06 October 1998 00:00 |
Bruce Bowler
Messages: 128
Registered: September 1998
|
Senior Member |
|
|
William Thompson wrote:
>
> Bruce Bowler <bbowler@bigelow.org> writes:
>
(stuff deleted)
>> if I run it as follows...
>
>> # idl
>> idl> .run 6953
>
>> it works fine...
>
>> if I run it like
>
>> # idl 6953
>
>> the results are
>
> (stuff deleted)
>
>> unfortunately, because of the way I need to interact with some other
>> stuff, I really need to run it the second way. What can I do to make it
>> work like the first way? I don't think precompiling will work either as
>> I generate the file "on the fly".
>
> If you're using Unix, then what you need to do is to create a file like the
> following:
>
> #!/bin/csh
> setenv MY_IDL_SCRIPT $argv
> idl << done
> .run $MY_IDL_SCRIPT
> done
>
> Use "chmod +x" to make the file executable, and then you can use this to
> execute whatever IDL script you want.
Unfortunately, as I said before, that won't work in my case. I am
passing this to a third party package that invokes it as
idl xxx
I have no control over that.
|
|
|
|
| Re: what am I doing wrong? [message #13035 is a reply to message #3602] |
Mon, 05 October 1998 00:00 |
thompson
Messages: 584
Registered: August 1991
|
Senior Member |
|
|
Bruce Bowler <bbowler@bigelow.org> writes:
> a couple of things... It is a top level script but there are no
> continuation lines that I can see... (true, my parser is *quite*
> different than IDL's :-)
> second, what happens depends on how I run it, which I find quite strange
> but then again, I'm a relative newbie at IDL...
> I've made some changes, while trying to figure things out. Here is the
> new script in it's entirety... (note I realize some of this [especially
> the case statement] aren't the best way to do things but like I said,
> I'm a newbie and I'm trying some things out and this is stripped down
> procedure anyway...)
(stuff deleted)
> if I run it as follows...
> # idl
> idl> .run 6953
> it works fine...
> if I run it like
> # idl 6953
> the results are
(stuff deleted)
> unfortunately, because of the way I need to interact with some other
> stuff, I really need to run it the second way. What can I do to make it
> work like the first way? I don't think precompiling will work either as
> I generate the file "on the fly".
If you're using Unix, then what you need to do is to create a file like the
following:
#!/bin/csh
setenv MY_IDL_SCRIPT $argv
idl << done
.run $MY_IDL_SCRIPT
done
Use "chmod +x" to make the file executable, and then you can use this to
execute whatever IDL script you want.
William Thompson
|
|
|
|
| Re: what am I doing wrong? [message #13036 is a reply to message #3602] |
Mon, 05 October 1998 00:00 |
Bruce Bowler
Messages: 128
Registered: September 1998
|
Senior Member |
|
|
a couple of things... It is a top level script but there are no
continuation lines that I can see... (true, my parser is *quite*
different than IDL's :-)
second, what happens depends on how I run it, which I find quite strange
but then again, I'm a relative newbie at IDL...
I've made some changes, while trying to figure things out. Here is the
new script in it's entirety... (note I realize some of this [especially
the case statement] aren't the best way to do things but like I said,
I'm a newbie and I'm trying some things out and this is stripped down
procedure anyway...)
!path = !path + ":" + getenv("SEADAS") + "/idl_lib" +
":/disk2/seadas_data"
inputFile = "S1998228192646.L1A_HNSG"
baseDir = strmid(inputFile,0,rstrpos(inputFile,"/")+1)
baseFile = strmid(inputFile,rstrpos(inputFile,"/")+1,999)
passId = strmid(baseFile,0,rstrpos(baseFile,"."))
datePart = strmid(passId,0,8)
ozoneFile = datePart + '12_EPTOMS.OZONE'
temp = size(findfile(ozonefile))
help, temp
print, temp
defOzone = 0
CASE 1 OF
(temp[0] EQ 1): BEGIN
ozoneFile = '$SDSDATA/CLIMATOLOGY.OZONE'
defOzone = 1
print, 'using default ozone data'
END
(temp[0] NE 1): BEGIN
defOzone = 0
print, 'ozone data found'
END
ELSE: print ,'things are really wacked'
ENDCASE
print, "after case"
end
if I run it as follows...
# idl
idl> .run 6953
it works fine...
if I run it like
# idl 6953
the results are
> For basic information, enter "IDLInfo" at the IDL> prompt.
>
> % Compiled module: RSTRPOS.
> % Compiled module: REVERSE.
> TEMP LONG = Array[4]
> 1 1 7 1
> % Case statement found no matches.
> % Execution halted at: $MAIN$
>
> (temp[0] EQ 1): BEGIN
> ^
> % Syntax error.
> At: 6953, Line 15
> using default ozone data
>
> END
> ^
> % Syntax error.
> At: 6953, Line 19
>
> (temp[0] NE 1): BEGIN
> ^
> % Syntax error.
> At: 6953, Line 20
> ozone data found
>
> END
> ^
> % Syntax error.
> At: 6953, Line 23
>
> ELSE: print ,'things are really wacked'
> ^
> % Syntax error.
> At: 6953, Line 24
>
> ENDCASE
> ^
> % Syntax error.
> At: 6953, Line 25
> after case
>
> end
> ^
> % Syntax error.
> At: 6953, Line 27
unfortunately, because of the way I need to interact with some other
stuff, I really need to run it the second way. What can I do to make it
work like the first way? I don't think precompiling will work either as
I generate the file "on the fly".
Jeffrey Newmnark wrote:
>
> Hi Bruce,
>
> Is this a section of a procedure or a top level script? If it is top
> level then you need
> $ to continue lines, otherwise, IDL compiles each line by itself.
>
> cheers,
> Jeff Newmark
>
> Bruce Bowler wrote:
>
>> I have a script that does...
>>
>>> anydata, subtemp, 45.0, 40.0, -71.5, -65.0, inbounds
>>>
>>> print, subtemp
>>> print, inbounds
>>>
>>> if (inbounds eq 1) then begin
>>> openw, tarunit, 'tarfile.include', /GET_LUN
>>> printf, tarunit, l1a
>>> printf, tarunit, ozonefile
>>> printf, tarunit, metFile1
>>> printf, tarunit, metFile2
>>> printf, tarunit, 'SEAWIFS_SENSOR_CAL.TBL'
>>> close, tarunit
>>> endif
>>
>> When I run it, I get
>>
>>> S1998228192646.SST
>>> 0
>>
>> and tarfile.include is created, which since inbounds is 0, shouldn't
>> happen... can anyone explain why it's not working or some additional
>> diagnostics that might point to the cause?
>>
>> TIA,
>> Bruce
>
> ------------------------------------------------------------ ------------
> ------------------------------------------------------------ ------
>
> Dr. Jeffrey Newmark
> SOHO - EIT Team Scientist
>
> voice: (301) 286-3163 Code 682.3
> email: newmark@eitv2.nascom.nasa.gov GSFC/NASA
> newmark@gsfc.nasa.gov Greenbelt, MD 20771
|
|
|
|
| Re: what am I doing wrong? [message #13037 is a reply to message #3602] |
Mon, 05 October 1998 00:00 |
Jeffrey Newmark
Messages: 8
Registered: April 1996
|
Junior Member |
|
|
Hi Bruce,
Is this a section of a procedure or a top level script? If it is top
level then you need
$ to continue lines, otherwise, IDL compiles each line by itself.
cheers,
Jeff Newmark
Bruce Bowler wrote:
> I have a script that does...
>
>> anydata, subtemp, 45.0, 40.0, -71.5, -65.0, inbounds
>>
>> print, subtemp
>> print, inbounds
>>
>> if (inbounds eq 1) then begin
>> openw, tarunit, 'tarfile.include', /GET_LUN
>> printf, tarunit, l1a
>> printf, tarunit, ozonefile
>> printf, tarunit, metFile1
>> printf, tarunit, metFile2
>> printf, tarunit, 'SEAWIFS_SENSOR_CAL.TBL'
>> close, tarunit
>> endif
>
> When I run it, I get
>
>> S1998228192646.SST
>> 0
>
> and tarfile.include is created, which since inbounds is 0, shouldn't
> happen... can anyone explain why it's not working or some additional
> diagnostics that might point to the cause?
>
> TIA,
> Bruce
------------------------------------------------------------ ------
Dr. Jeffrey Newmark
SOHO - EIT Team Scientist
voice: (301) 286-3163 Code 682.3
email: newmark@eitv2.nascom.nasa.gov GSFC/NASA
newmark@gsfc.nasa.gov Greenbelt, MD 20771
-
Attachment: mail.inc
(Size: 0.29KB, Downloaded 71 times)
|
|
|
|
| Re: what am I doing wrong? [message #13127 is a reply to message #3602] |
Tue, 06 October 1998 00:00 |
Bruce Bowler
Messages: 128
Registered: September 1998
|
Senior Member |
|
|
William Thompson wrote:
> Another way is to use two procedure files. One file will contain the
> main-level program that you wish to run. For example, let's say you save it in
> a file called myprogram.idl. The second file will then contain the single line
>
> .run myprogram
>
> You can then pass the name of this second file to IDL. For example, if the
> second file is named run_myprogram.pro, then the statement
>
> idl run_myprogram
>
> will execute the main-level procedure myprogram. It would be just the same as
> if you had typed ".run myprogram" at the command prompt.
Hmmmmm... I like it...
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
