| Re: IDL Save Files: A Journey into Craig's World [message #38913] |
Mon, 05 April 2004 09:43  |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
Michael Wallace <mwallace.removethis@swri.edu.invalid> writes:
> No. But I was curious if there were any differences. I'm working on a
> project where one of my requirements is to support IDL save files (much
> to my chagrin) as an output. However, I have no control over what
> version of IDL folks are using. I will always be using the latest
> version of IDL, but who knows what kind of incompatibilities may arise
> since the format of save files could change at any time. RSI has been
> known to change things without warning before...
If you control the writing of output files, then I might suggest that
you use my (mid- or low-level) library routines to make IDL save
files. Since the library is a totally independent implementation of
save-file functionality, you don't have to worry about it changing
with new versions of IDL. Unfortunately it does not do compressed
files, or files with objects inside, if either of those are important
to you.
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@REMOVEcow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
|
|
|
|
| Re: IDL Save Files: A Journey into Craig's World [message #38916 is a reply to message #38915] |
Mon, 05 April 2004 09:02  |
Michael Wallace
Messages: 409
Registered: December 2003
|
Senior Member |
|
|
> The variable-type descriptors are not full fledged records. What you
> are really asking is how the descriptors appear within a single
> variable's record. What you will find is that a structure descriptor
> starts with a value of "9". Then, if the structure contains any
> arrays or structures within itself, then descriptors for those follow,
> first arrays, then structures. For your problem, you would see the
> following byte stream:
> STRUCT_DESCR (N tags, including M arrays)
> ARRAY_DESCR (1)
> ...
> ARRAY_DESCR (M)
Ah. Okay. Yes, that's what I'm really asking.
> Do you have any reason to believe that IDL 6 save file formats have
> changed?
No. But I was curious if there were any differences. I'm working on a
project where one of my requirements is to support IDL save files (much
to my chagrin) as an output. However, I have no control over what
version of IDL folks are using. I will always be using the latest
version of IDL, but who knows what kind of incompatibilities may arise
since the format of save files could change at any time. RSI has been
known to change things without warning before...
|
|
|
|
| Re: IDL Save Files: A Journey into Craig's World [message #38940 is a reply to message #38916] |
Fri, 02 April 2004 21:58 |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
Greetings!
Michael Wallace <mwallace.removethis@swri.edu.invalid> writes:
>
> What I'm interested in is an example of a save file which consists of a
> structure with multiple one-dimensional arrays. The arrays are the same
> size, but may be of different primitive data types (usually integers,
> sometimes floats, sometimes strings). I guess I'm trying to figure out
> how all the records line up and where everything falls for a complicated
> structure like this. I want to determine how easy it would be to create
> a save file such as this outside of IDL.
If you are looking for an example, then probably you could make an
example pretty easily with some sample data, right?
You can also easily use the mid- or low-level routines in my library
to create sample save files of any type you desire.
The variable-type descriptors are not full fledged records. What you
are really asking is how the descriptors appear within a single
variable's record. What you will find is that a structure descriptor
starts with a value of "9". Then, if the structure contains any
arrays or structures within itself, then descriptors for those follow,
first arrays, then structures. For your problem, you would see the
following byte stream:
STRUCT_DESCR (N tags, including M arrays)
ARRAY_DESCR (1)
...
ARRAY_DESCR (M)
After the first definition of a structure, any following structures of
the same data type refer back to the original instead of redefining it
again.
> By the way, do you (Craig) have plans to update the document for IDL 6.0?
Let's say that the implied legal threats against me by RSI and Kodak
have more or less cooled my interest in IDL save files.
Do you have any reason to believe that IDL 6 save file formats have
changed?
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@REMOVEcow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
| Re: IDL Save Files: A Journey into Craig's World [message #39164 is a reply to message #38913] |
Sun, 18 April 2004 19:26 |
p.sommer
Messages: 20
Registered: April 2004
|
Junior Member |
|
|
Craig Markwardt <craigmnet@REMOVEcow.physics.wisc.edu> wrote in message news:<ony8paml3e.fsf@cow.physics.wisc.edu>...
> Michael Wallace <mwallace.removethis@swri.edu.invalid> writes:
>> No. But I was curious if there were any differences. I'm working on a
>> project where one of my requirements is to support IDL save files (much
>> to my chagrin) as an output. However, I have no control over what
>> version of IDL folks are using. I will always be using the latest
>> version of IDL, but who knows what kind of incompatibilities may arise
>> since the format of save files could change at any time. RSI has been
>> known to change things without warning before...
>
> If you control the writing of output files, then I might suggest that
> you use my (mid- or low-level) library routines to make IDL save
> files. Since the library is a totally independent implementation of
> save-file functionality, you don't have to worry about it changing
> with new versions of IDL. Unfortunately it does not do compressed
> files, or files with objects inside, if either of those are important
> to you.
>
> Craig
For what it's worth, 6.1 will support a sav file query -- new routines
to handle this. I've seen alpha versions of the functionality and it
looks promising going forward.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
