| Resolve_all: bogus unresolved references [message #36492] |
Tue, 23 September 2003 07:40  |
William Gallery
Messages: 4
Registered: September 2003
|
Junior Member |
|
|
I have an idl program I wish to distribute as an idl save set. When I run
"resolve_all,/cont' on it, It lists three unresoved routines:
....
% Compiled module: POLY_FIT.
% Attempt to call undefined procedure/function: 'TRNLOG'.
% Execution halted at: RESOLVE_ALL_BODY 117
C:\RSI\IDL56\lib\resolve_all.pro
% RESOLVE_ALL 162
C:\RSI\IDL56\lib\resolve_all.pro
% $MAIN$
% Compiled module: COMBINE_SIGNALS.
% Attempt to call undefined procedure/function: 'DELLOG'.
% Execution halted at: RESOLVE_ALL_BODY 117
C:\RSI\IDL56\lib\resolve_all.pro
% RESOLVE_ALL 161
C:\RSI\IDL56\lib\resolve_all.pro
% $MAIN$
% Attempt to call undefined procedure/function: 'SETLOG'.
% Execution halted at: RESOLVE_ALL_BODY 117
C:\RSI\IDL56\lib\resolve_all.pro
% RESOLVE_ALL 161
C:\RSI\IDL56\lib\resolve_all.pro
% $MAIN$
% Compiled module: RESISTANT_MEAN.
...
However, I have searched everywhere for anything named trnlog, dellog or
setlog using:
grep -iR trnlog ./
I looked in the current directory and below, the main idl directory and
below, and all directories included in !Path and below and found nothing.
The program runs fine as is. I believe these messages are benign but I
would like to find their cause.
Is there a way to determine where trnlog is called from? Has anyone seen
similar messages?
IDL 5.6
Windows NT
Cheers
Bill Gallery
Frontier Technology, Inc.
|
|
|
|
| Re: Resolve_all: bogus unresolved references [message #37152 is a reply to message #36492] |
Mon, 17 November 2003 08:58  |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
Karsten Rodenacker <Karsten.Rodenacker@gsf.de> writes:
> Thank you for your additional information. I did not mention you as an
> offence, more as a documented trial. I am using your cmsave lib more
> frequent than the idl routines. So I did not consider your usage of
> forward_function as ill-minded.
> Karsten
Hi Karsten-- I hoped the smiley :-) would show that I was good humored
about the issue. A separate point of my message was to mention that
you could use the SKIP_ROUTINES keyword to RESOLVE_ALL to exclude the
offending routines, without resorting to editing library code.
Yours,
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@REMOVEcow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
| Re: Resolve_all: bogus unresolved references [message #37154 is a reply to message #36492] |
Mon, 17 November 2003 08:25 |
Karsten Rodenacker
Messages: 98
Registered: July 1997
|
Member |
|
|
Thank you for your additional information. I did not mention you as an
offence, more as a documented trial. I am using your cmsave lib more
frequent than the idl routines. So I did not consider your usage of
forward_function as ill-minded.
Karsten
Craig Markwardt schrieb:
> Karsten Rodenacker <Karsten.Rodenacker@gsf.de> writes:
>
>> In the meanwhile I found the source of these messages. All mentioned
>> routines are 'defined' by forward_function in the cmsavelib software of
>> Craig Markwardt. After commenting out these lines (without testing
>> further functionality) the messages disappear.
>
>
> Since my name was mentioned, I'll have to make a comment :-) My use of
> FORWARD_FUNCTION is a paranoic attempt to allow my library routines to
> compile on a wide range of IDL versions. By using FORWARD_FUNCTION,
> IDL understands that the mentioned routines are functions, even if
> compiled under a version of IDL where the (built-in) function doesn't
> exist. That being said, I think I was probably a bit *too* paranoic.
>
> But it is also unfortunate that RESOLVE_ALL is so narrow-minded.
> After all, it certainly *knows* the names of all the built-in
> functions (by calling the ROUTINE_NAMES() function). Thankfully,
> rather than editing my code, you can use the SKIP_ROUTINES documented
> keyword to avoid the offending functions. [ Strangely, it's only
> documented in the header of the code. ]
>
> Craig
>
|
|
|
|
| Re: Resolve_all: bogus unresolved references [message #37155 is a reply to message #36492] |
Mon, 17 November 2003 07:44 |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
Karsten Rodenacker <Karsten.Rodenacker@gsf.de> writes:
> In the meanwhile I found the source of these messages. All mentioned
> routines are 'defined' by forward_function in the cmsavelib software of
> Craig Markwardt. After commenting out these lines (without testing
> further functionality) the messages disappear.
Since my name was mentioned, I'll have to make a comment :-) My use of
FORWARD_FUNCTION is a paranoic attempt to allow my library routines to
compile on a wide range of IDL versions. By using FORWARD_FUNCTION,
IDL understands that the mentioned routines are functions, even if
compiled under a version of IDL where the (built-in) function doesn't
exist. That being said, I think I was probably a bit *too* paranoic.
But it is also unfortunate that RESOLVE_ALL is so narrow-minded.
After all, it certainly *knows* the names of all the built-in
functions (by calling the ROUTINE_NAMES() function). Thankfully,
rather than editing my code, you can use the SKIP_ROUTINES documented
keyword to avoid the offending functions. [ Strangely, it's only
documented in the header of the code. ]
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@REMOVEcow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
| Re: Resolve_all: bogus unresolved references [message #37157 is a reply to message #36492] |
Mon, 17 November 2003 06:32 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Karsten Rodenacker writes:
> It was a bit cumbersome to 'look up' this FW but however, here is a list
> of commands and results showing where the fw_ comes from:
>
> IDL> .full_reset_session
> IDL> .reset_session
> IDL> resolve_routine,'XROI',/either,/comp
> IDL> print,routine_info(/unreso)
> IPLOT LIVE_DESTROY XMANAGER
> IDL> resolve_routine,'LIVE_DESTROY',/either,/comp
> IDL> print,routine_info(/unreso)
> FW_LOAD_FILE FW_LOAD_GENERIC FW_UPDATEDATAWINDOW INSPUT IPLOT
Oh, dear. I hope you gave him a kick in the ass
when you found him! :-)
Cheers,
David
--
David W. Fanning, Ph.D.
Fanning Software Consulting, Inc.
Phone: 970-221-0438, E-mail: david@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Resolve_all: bogus unresolved references [message #37158 is a reply to message #36492] |
Mon, 17 November 2003 06:27 |
Karsten Rodenacker
Messages: 98
Registered: July 1997
|
Member |
|
|
It was a bit cumbersome to 'look up' this FW but however, here is a list
of commands and results showing where the fw_ comes from:
IDL> .full_reset_session
IDL> .reset_session
IDL> resolve_routine,'XROI',/either,/comp
IDL> print,routine_info(/unreso)
IPLOT LIVE_DESTROY XMANAGER
IDL> resolve_routine,'LIVE_DESTROY',/either,/comp
IDL> print,routine_info(/unreso)
FW_LOAD_FILE FW_LOAD_GENERIC FW_UPDATEDATAWINDOW INSPUT IPLOT
Regards
Karsten
David Fanning schrieb:
> Karsten Rodenacker writes:
>
>
>> However after compiling an object I am messing now around with other
>> missing routines I never heard about. Perhaps there is another
>> historical OS still hidden?
>
>
> I would guess so. These FW_*** routines are not IDL routines. :-)
>
> It looks to me like you need to look up this "FW" and
> add his library to your path.
>
> Cheers,
>
> David
|
|
|
|
| Re: Resolve_all: bogus unresolved references [message #37161 is a reply to message #36492] |
Mon, 17 November 2003 05:26 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Karsten Rodenacker writes:
> However after compiling an object I am messing now around with other
> missing routines I never heard about. Perhaps there is another
> historical OS still hidden?
I would guess so. These FW_*** routines are not IDL routines. :-)
It looks to me like you need to look up this "FW" and
add his library to your path.
Cheers,
David
--
David W. Fanning, Ph.D.
Fanning Software Consulting, Inc.
Phone: 970-221-0438, E-mail: david@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Resolve_all: bogus unresolved references [message #37162 is a reply to message #36492] |
Mon, 17 November 2003 01:01 |
Karsten Rodenacker
Messages: 98
Registered: July 1997
|
Member |
|
|
In the meanwhile I found the source of these messages. All mentioned
routines are 'defined' by forward_function in the cmsavelib software of
Craig Markwardt. After commenting out these lines (without testing
further functionality) the messages disappear.
However after compiling an object I am messing now around with other
missing routines I never heard about. Perhaps there is another
historical OS still hidden?
Regards
Karsten
% Attempt to call undefined procedure/function: 'FW_QUERYDEFAULTSTYLE'.
% Execution halted at: RESOLVE_ALL_BODY 280
D:\RSI\IDL60\lib\resolve_all.pro
% RESOLVE_ALL 332
D:\RSI\IDL60\lib\resolve_all.pro
% $MAIN$
% Attempt to call undefined procedure/function: 'FW_QUERYNAMEDSTYLE'.
% Attempt to call undefined procedure/function: 'FW_UPDATEDATAWINDOW'.
% Attempt to call undefined procedure/function: 'FW_CWDIMSIZES'.
% Attempt to call undefined procedure/function: 'FW_DATABROWSER'.
% Attempt to call undefined procedure/function: 'FW_GETDATAFROMTEXTWIDGET'.
% Attempt to call undefined procedure/function: 'FW_QUERYOKTODELETEDATA'.
% Attempt to call undefined procedure/function: 'FW_QUERYSTYLELIST'.
Karsten Rodenacker schrieb:
> In addition to the undefined procedures TRNLOG,DELLOG,SETLOG mentioned
> by William Gallery I get messages with quite standard routines:
>
> % Attempt to call undefined procedure/function: 'COMPLEX'.
> % Execution halted at: RESOLVE_ALL_BODY 280
> D:\RSI\IDL60\lib\resolve_all.pro
> % RESOLVE_ALL 332
> D:\RSI\IDL60\lib\resolve_all.pro
> % $MAIN$
> ...
> % Attempt to call undefined procedure/function: 'DCOMPLEX'.
> % Attempt to call undefined procedure/function: 'FIX'.
> % Attempt to call undefined procedure/function: 'HANDLE_CREATE'.
> % Attempt to call undefined procedure/function: 'MAKE_ARRAY'.
> % Attempt to call undefined procedure/function: 'OBJ_CLASS'.
> % Attempt to call undefined procedure/function: 'PTRARR'.
> % Attempt to call undefined procedure/function: 'PTR_NEW'.
> % Attempt to call undefined procedure/function: 'ROUTINE_NAMES'.
> % Attempt to call undefined procedure/function: 'UINT'.
> ...
>
> Maybe there is an idea in the group to mend that behaviour?
> Regards
> Karsten Rodenacker
>
>
> William Gallery schrieb:
>
>> I have an idl program I wish to distribute as an idl save set. When I
>> run
>> "resolve_all,/cont' on it, It lists three unresoved routines:
>> ....
>> % Compiled module: POLY_FIT.
>> % Attempt to call undefined procedure/function: 'TRNLOG'.
>> % Execution halted at: RESOLVE_ALL_BODY 117
>> C:\RSI\IDL56\lib\resolve_all.pro
>> % RESOLVE_ALL 162
>> C:\RSI\IDL56\lib\resolve_all.pro
>> % $MAIN$
>> % Compiled module: COMBINE_SIGNALS.
>> % Attempt to call undefined procedure/function: 'DELLOG'.
>> % Execution halted at: RESOLVE_ALL_BODY 117
>> C:\RSI\IDL56\lib\resolve_all.pro
>> % RESOLVE_ALL 161
>> C:\RSI\IDL56\lib\resolve_all.pro
>> % $MAIN$
>> % Attempt to call undefined procedure/function: 'SETLOG'.
>> % Execution halted at: RESOLVE_ALL_BODY 117
>> C:\RSI\IDL56\lib\resolve_all.pro
>> % RESOLVE_ALL 161
>> C:\RSI\IDL56\lib\resolve_all.pro
>> % $MAIN$
>> % Compiled module: RESISTANT_MEAN.
>> ...
>>
>> However, I have searched everywhere for anything named trnlog, dellog or
>> setlog using:
>>
>> grep -iR trnlog ./
>>
>> I looked in the current directory and below, the main idl directory and
>> below, and all directories included in !Path and below and found nothing.
>> The program runs fine as is. I believe these messages are benign but I
>> would like to find their cause.
>>
>> Is there a way to determine where trnlog is called from? Has anyone seen
>> similar messages?
>>
>> IDL 5.6
>> Windows NT
>>
>>
>> Cheers
>>
>> Bill Gallery
>> Frontier Technology, Inc.
>>
>>
>
|
|
|
|
| Re: Resolve_all: bogus unresolved references [message #37163 is a reply to message #36492] |
Mon, 17 November 2003 00:25 |
Karsten Rodenacker
Messages: 98
Registered: July 1997
|
Member |
|
|
In addition to the undefined procedures TRNLOG,DELLOG,SETLOG mentioned
by William Gallery I get messages with quite standard routines:
% Attempt to call undefined procedure/function: 'COMPLEX'.
% Execution halted at: RESOLVE_ALL_BODY 280
D:\RSI\IDL60\lib\resolve_all.pro
% RESOLVE_ALL 332
D:\RSI\IDL60\lib\resolve_all.pro
% $MAIN$
...
% Attempt to call undefined procedure/function: 'DCOMPLEX'.
% Attempt to call undefined procedure/function: 'FIX'.
% Attempt to call undefined procedure/function: 'HANDLE_CREATE'.
% Attempt to call undefined procedure/function: 'MAKE_ARRAY'.
% Attempt to call undefined procedure/function: 'OBJ_CLASS'.
% Attempt to call undefined procedure/function: 'PTRARR'.
% Attempt to call undefined procedure/function: 'PTR_NEW'.
% Attempt to call undefined procedure/function: 'ROUTINE_NAMES'.
% Attempt to call undefined procedure/function: 'UINT'.
...
Maybe there is an idea in the group to mend that behaviour?
Regards
Karsten Rodenacker
William Gallery schrieb:
> I have an idl program I wish to distribute as an idl save set. When I run
> "resolve_all,/cont' on it, It lists three unresoved routines:
> ....
> % Compiled module: POLY_FIT.
> % Attempt to call undefined procedure/function: 'TRNLOG'.
> % Execution halted at: RESOLVE_ALL_BODY 117
> C:\RSI\IDL56\lib\resolve_all.pro
> % RESOLVE_ALL 162
> C:\RSI\IDL56\lib\resolve_all.pro
> % $MAIN$
> % Compiled module: COMBINE_SIGNALS.
> % Attempt to call undefined procedure/function: 'DELLOG'.
> % Execution halted at: RESOLVE_ALL_BODY 117
> C:\RSI\IDL56\lib\resolve_all.pro
> % RESOLVE_ALL 161
> C:\RSI\IDL56\lib\resolve_all.pro
> % $MAIN$
> % Attempt to call undefined procedure/function: 'SETLOG'.
> % Execution halted at: RESOLVE_ALL_BODY 117
> C:\RSI\IDL56\lib\resolve_all.pro
> % RESOLVE_ALL 161
> C:\RSI\IDL56\lib\resolve_all.pro
> % $MAIN$
> % Compiled module: RESISTANT_MEAN.
> ...
>
> However, I have searched everywhere for anything named trnlog, dellog or
> setlog using:
>
> grep -iR trnlog ./
>
> I looked in the current directory and below, the main idl directory and
> below, and all directories included in !Path and below and found nothing.
> The program runs fine as is. I believe these messages are benign but I
> would like to find their cause.
>
> Is there a way to determine where trnlog is called from? Has anyone seen
> similar messages?
>
> IDL 5.6
> Windows NT
>
>
> Cheers
>
> Bill Gallery
> Frontier Technology, Inc.
>
>
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
