Alan W. Irwin wrote:
On 2008-03-10 08:54+0100 Arjen Markus wrote:

  
I do not mind a "7" instead of a "C" - it simply seemed a good idea at
the time,
as they figure in the FORTRAN 77 and the Fortran 95 bindings and work around
a calling convention difference with C.
    

Hi Arjen:

Following up on your above comment, I have just changed (revision 8268) the
internal plmap and plmeridians suffix from "c" to "7" to be consistent with
the rest of the internal suffixes.  As a test of these changes I have
confirmed that the f77 and f95 results for example 19 are still identical to
the corresponding C results.

  
Come to think of it, the intermediate function plsetmapform is
superfluous - it can
be put in plmap and plmeridians ...
    

Arjen, eliminating this one-line function sounds like a good idea, but I
don't completely understand what that one line does so could you implement
this change, please?
  
Sure, no problem.
Here is the current status for f77 and f95:

software@raven> scripts/check_api_completeness.sh f77 |grep '[+-]' |grep -v
'@@' |grep -v '\-\-\-' |grep -v '+++'
-plarrows
-plenv0
-plgcolbga
-plgfci
-plimage
-plimagefr
-plot3dcl
+plsetmapformc
-plsfci
-plsurf3dl

software@raven> scripts/check_api_completeness.sh f95 |grep '[+-]' |grep -v
'@@' |grep -v '\-\-\-' |grep -v '+++'
-plarrows
-plenv0
-plgfci
-plimage
-plimagefr
-plot3dcl
+plsetmapformc
-plsfci
-plsurf3dl

So ignoring plarrows (discussed previously in this thread but not resolved
yet) and plsetmapformc (discussed above), there is obviously still some work
to do to make the f77 and f95 interfaces have the same degree of
completeness as our core C library.
  
I will have a look at these too. Nice to have a tool like that :). Checking
such things manually is very tedious.

Regards,

Arjen