Sorry for the spam. I just wanted to say many thanks again for 
everyones' very rapid responses on this issue. It's good to know there 
are people out there with the knowledge...
John


On 01/10/10 15:20, Peter W. Draper wrote:
> On Fri, 1 Oct 2010, John Barnes wrote:
>
>    
>> Many thanks for such a swift response. I do know about the STATUS issue,
>> but had in fact made a mistake with previous mapping of the axis array,
>> so thanks for this comment. I'm still having problems however after
>> correcting this.
>>
>> Here again is the input NDF structure (from hdstrace):
>>
>> DYNTIMES<NDF>
>>
>>   DATA_ARRAY<ARRAY>            {structure}
>>      DATA(12)<_REAL>            300,300,300,300,300,300,300,300,300,
>>                                     300,300,300
>>      BAD_PIXEL<_LOGICAL>         FALSE
>>      ORIGIN(1)<_INTEGER>         1
>>
>>   AXIS(1)<AXIS>             {structure}
>>      DATA_ARRAY<ARRAY>            {structure}
>>         ORIGIN(1)<_INTEGER>         1
>>         DATA(12)<_REAL>            0,91536.9,183074,274611,366147,
>>                                        ... 732295,823832,915369,1006910
>>
>>   MORE<EXT>              {structure}
>>      FIGARO<EXT>              {structure}
>>         HJD0<_REAL>            42512.5
>>
>> End of Trace.
>>
>>
>>
>> Here is the relevant fragment of my code that deals with this:
>>
>>      CALL NDF_ASSOC('TIMECAL','READ',TIMECAL,STATUS)
>>      CALL NDF_MAP(TIMECAL,'Data','_REAL','READ',DW_PTR,NELM2,STATUS)
>>      CALL NDF_DIM(TIMECAL, 2, DIMS, NDIMS2, STATUS )
>>      CALL NDF_AMAP(TIMECAL,'CENTRE',1,'_REAL','READ',
>>      :                                        TS_PTR,WELM,STATUS)
>>
>>         write(*,*) 'status=',status
>>         CALL NDF_XLOC(TIMECAL, 'MORE', 'READ', LOC, STATUS )
>>      
> Hi John,
>
> that should be:
>
>           CALL NDF_XLOC(TIMECAL, 'FIGARO', 'READ', LOC, STATUS )
>
> all extensions are implicitly in "MORE". See Accessing Existing Extensions
> in SUN/33.
>
> If you just want HJD0 that should be a one liner:
>
>         CALL NDF_XGT0I( TIMECAL, 'FIGARO', 'HJD0', HJD0, STATUS )
>
>
>    
>>         print *,'LOC=',loc
>>         WRITE(*,*) 'status=',status
>>
>>
>>
>> And here is the output I get from the above fragment (note that status is 0
>> before the call to NDF_XLOC):
>>
>> status= 0
>> LOC=<NOT A LOCATOR>
>> status= 232950290
>>
>>
>> So perhaps my understanding of what a locator is is incorrect. Additionally,
>>
>> CALL NDF_XSTAT( TIMECAL, 'MORE', THERE, STATUS )
>>
>> Also returns THERE=.FALSE.
>>
>> The info for all this is at
>> http://www.starlink.ac.uk/docs/sun33.htx/node70.html
>>
>> Am I using the wrong subroutine to access/map the extension - I want to read
>> the value in MORE.FIGARO.HJD0  (=42512.5)
>>
>> Cheers,
>> John
>>
>>
>>
>>
>>
>>
>> On 01/10/10 14:18, David Berry wrote:
>>      
>>> On 1 October 2010 14:09, John Barnes<[log in to unmask]>   wrote:
>>>
>>>        
>>>> Hi,
>>>>
>>>> I'm using 64 bit star-hawaiki and trying to access data in NDF extensions
>>>> using my own routines. I have tried following the instructions at
>>>> http://www.starlink.ac.uk/docs/sun33.htx/node66.html
>>>>
>>>> For example enquiring the presence of a structure with:
>>>>
>>>>        CALL NDF_XSTAT( INDF, 'MORE', THERE, STATUS )
>>>>
>>>> where INDF is an already opened NDF for which I've successfully mapped the
>>>> data array and MORE is the name of the extension,
>>>> always returns THERE = FALSE even when the MORE structure is clearly
>>>> present.
>>>>
>>>> Similarly using
>>>>
>>>>        INCLUDE 'DAT_PAR'
>>>>        CHARACTER * ( DAT__SZLOC ) LOC
>>>>        INTEGER OFFSET
>>>>        ...
>>>>       CALL NDF_XLOC( INDF, 'MORE', 'READ', LOC, STATUS )
>>>>
>>>> doesn't work and returns the value
>>>>
>>>> <NOT A LOCATOR>
>>>>
>>>> in the variable LOC. For both NDF_XSTAT and NDF_XLOC, STATUS is returned
>>>> with a non-zero value (a large integer value).
>>>>
>>>>          
>>> Hi John,
>>>
>>> A non-zero status means that something has gone wrong, either in the
>>> subroutine you just called, or in an earlier subroutine. Once an error
>>> has occurred, all subsequent starlink subroutine calls will return
>>> immediately without doing anything other than setting their returned
>>> arguments to "null" values. In particularly, they will return the
>>> supplied non-zero status value unchanged. So you need to work out
>>> where the original error is occurring (i.e. where and why STATUS first
>>> gets set non-zero), and fix it. If you allow your application to run
>>> to completion you should see an error message indicating the nature of
>>> the problem (so long as you don't explicitly cancel the error by
>>> calling ERR_ANNUL).
>>>
>>> David
>>>
>>>
>>>
>>>
>>>
>>>
>>>        
>>>> Is there perhaps a problem here in 64 bit mode or am I using the wrong
>>>> subroutine calls? Are there any alternative calls
>>>> I can use - I'm ultimately wanting to read the value in MORE.FIGARO.HJD0
>>>> (see below).
>>>>
>>>> Here are the contents of the NDF for which I want to access the MORE
>>>> structure:
>>>>
>>>>
>>>> % hdstrace dyntimes
>>>>
>>>> DYNTIMES<NDF>
>>>>
>>>>     DATA_ARRAY<ARRAY>            {structure}
>>>>        DATA(12)<_REAL>            300,300,300,300,300,300,300,300,300,
>>>>                                       300,300,300
>>>>        BAD_PIXEL<_LOGICAL>         FALSE
>>>>        ORIGIN(1)<_INTEGER>         1
>>>>
>>>>     AXIS(1)<AXIS>             {structure}
>>>>        DATA_ARRAY<ARRAY>            {structure}
>>>>           ORIGIN(1)<_INTEGER>         1
>>>>           DATA(12)<_REAL>            0,91536.9,183074,274611,366147,
>>>>                                          ... 732295,823832,915369,1006910
>>>>
>>>>     MORE<EXT>              {structure}
>>>>        FIGARO<EXT>              {structure}
>>>>           HJD0<_REAL>            42512.5
>>>>
>>>> End of Trace.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Many thanks,
>>>> John Barnes
>>>>
>>>>
>>>> --
>>>>
>>>>
>>>>
>>>> **********************************************************
>>>> John Barnes
>>>> Centre for Astrophysics Research,
>>>> Science&    Technology Research Institute
>>>> University of Hertfordshire,
>>>> College Lane, Hatfield.
>>>> Herts AL10 9AB. UK
>>>> Tel: 01707 285162
>>>> http://star-www.herts.ac.uk/~jbarnes
>>>>
>>>> **********************************************************
>>>>
>>>>
>>>>          
>>> .
>>>
>>>
>>>        
>>
>> --
>> **********************************************************
>> John Barnes
>> Centre for Astrophysics Research,
>> Science&   Technology Research Institute
>> University of Hertfordshire,
>> College Lane, Hatfield.
>> Herts AL10 9AB. UK
>> Tel: 01707 285162
>> http://star-www.herts.ac.uk/~jbarnes
>>
>> **********************************************************
>>
>>      
> --
> Peter W. Draper, http://astro.dur.ac.uk/~pdraper
> .
>
>    


-- 
**********************************************************
John Barnes
Centre for Astrophysics Research,
Science&  Technology Research Institute
University of Hertfordshire,
College Lane, Hatfield.
Herts AL10 9AB. UK
Tel: 01707 285162
http://star-www.herts.ac.uk/~jbarnes

**********************************************************