McStas 1.5: if you encouter compilation problems

[Per-Olof Åstrand]

Dear McStas user,

To explain this it requires a basic understanding of the philosophy of
McStas. McStas has a layered structure which consists of one level with
a "kernel" and another level consisting of  components and instruments,
where the components and instruments are written in a McStas
meta-language. The meta-language is defined in the McStas manual.

To extend the useability of the components, it is of course tempting to
look into and use internal information in the kernel. It can, however,
not be expected that the internals of the kernel remain unchanged. In
contrast, it can be required that the kernel calls documented in the
manual always will not be changed. So even if it is tempting to include
undocumented features, it should be avoided. It is the only way to
guarantee backward compability.

This is today not a serious problem. I would assume it is only a few
components using undocumented features.

Using the same arguments, it is unacceptable and very unfortunate with
different McStas kernels available on the web. If the direction is that
we will have many various McStas kernels available on the web, it will
soon lead to a large degree of incompability in the components. I hope
this will not be a future problem, and we will do our best to include
the "McStas-ill" features in the official release as soon as possible.

But to conclude, an extensive use of undocumented kernel features and
various McStas kernels on the web, and McStas will very soon be a mess.

Best regards,

Per-Olof Åstrand


Emmanuel Farhi wrote:

> Hello McStas users,
>
> I realised that there was an undocumented modification in the Mcstas
> version 1.5, that makes it a few percent uncompatible with some
> previous components (and perhaps the components that you wrote...)
> Per-Olof Astrand changed the nature of one of the symbols in the
> kernel ('mccompcurname' is not a char string anymore).
> Thus if you encounter any compilation problem with McStas 1.5, just
> look for the 'mccompcurname' symbol and change it to
> 'NAME_CURRENT_COMP'.
>
> Invitation:
> Use version 1.6-ill <http://www.ill.fr/tas/mcstas/> that one from
> which Per-Olof says "The parallel development of McStas at ILL is very
> unfortunate". Wow ! you're not very kind Per-Olof, why ?
>
> Cheers. EF.
>
> --
> What's up Doc ?
> --------------------------------------------
> Emmanuel FARHI, http://www.ill.fr/tas/people/Farhi.html   \|/ ____ \|/
> CS-Group ILL4/156, Institut Laue-Langevin (ILL) Grenoble  ~@-/ oO \-@~
> 6 rue J. Horowitz, BP 156, 38042 Grenoble Cedex 9,France  /_( \__/ )_\
> Work :Tel (33/0) 4 76 20 71 35. Fax (33/0) 4 76 48 39 06     \__U_/
>
>

--
Per-Olof Åstrand
Dept. of Chemistry, University of Copenhagen
and Materials Research Department, Risø National Laboratory
per-olof.aastrand at theory.ki.ku.dk or per-olof.aastrand at risoe.dk
http://theochem.ki.ku.dk/~peo/