2008/9/10 Peter W. Draper <[log in to unmask]>:
> On Wed, 10 Sep 2008, David Berry wrote:
>
>> 2008/9/10 Peter W. Draper <[log in to unmask]>:
>>>
>>> On Wed, 10 Sep 2008, David Berry wrote:
>>>
>>>> I think there are a few uses of "%" in kappa. E.g. listmake.f:
>>>>
>>>>
>>>> * Report an error if we are using pixel centres.
>>>> IF( MODE .EQ. 'PIXEL' .OR. MODE .EQ. 'GOOD' ) THEN
>>>> STATUS = SAI__ERROR
>>>> CALL ERR_REP( 'LISTMAKE_ERR1', 'Parameter %MODE has been '//
>>>> : 'set to a value which requires an NDF, '//
>>>> : 'but a null value was given for parameter '//
>>>> : '%NDF.', STATUS )
>>>> END IF
>>>>
>>>> There are MSG_OUT calls in elprof.f that use "%".
>>>
>>> CCDPACK uses the more testing:
>>>
>>> CALL MSG_SETC( 'NAME', NAME )
>>> CALL ERR_REP( 'CCD1_NDFAC',
>>> : ' Unable to obtain valid list of NDF names using'//
>>> : ' parameter %^NAME' , STATUS )
>>
>> So what happens if you include both the "%" character AND the "^"
>> character? The "Parameter References" section of SUN104 implies you
>> replace "^" by "%".
>
> Seems you get the name of the parameter, 'IN' in this case, so clearly the
> ^NAME expansion trumps in the current MERs and then the parameter keyword
> lookup happens (which is redundant of course).
Obviously really - if I'd thought about it for a second.
David
|