[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [ESPResSo-devel] blockfile
From: |
Axel Arnold |
Subject: |
Re: [ESPResSo-devel] blockfile |
Date: |
Tue, 28 Aug 2007 10:25:50 +0200 |
User-agent: |
KMail/1.8.2 |
On Monday 27 August 2007 13:57, Ulf Schiller wrote:
> I think it's a combination between the implementation of the
> blockfile_write_variable procedure and TLC's handling of lists.
> If the list of variables to be written is a list of length one, it is
> directly passed to the setmd command. The difference between box_l and
> { box_l } boils down to the difference of the strings "box_l" and
> " box_l ", which suggests that
> blockfile channel write variable {box_l}
> works, and it does.
> If the list is longer it is processed by foreach and the elements are
> successively passed to setmd. One may also consider this a bug in the
> implementation of blockfile_write_variable.
> Does anybody remember whether there was a reason that it is implemented
> this way (besides formatting output nicely)?
Hi all!
Well, for parsing with awk it is more convenient, and indeed looks nicer.
Plus, the format should be human writable.
I would suggest to make the read routine more robust, by trimming off leading
and trailing spaces from $which in blockfile_write_variable. In the course of
that, I would also like to add a whitelist feature, that is, if
blockfile_(tcl)variable_whitelist is set, only variable from this list are
read. This is sort of the invert of the blacklist feature and is quite
helpful if you use "blockfile write tclvariable" to avoid overwriting
filepointers etc. As long as you don't use a global variable
blockfile_(tcl)variable_whitelist, you won't notice the change.
Does anybody have objections to any of the changes? Otherwise I implement that
this afternoon.
Axel
--
Dr. Axel Arnold
FOM Institute for Atomic and Molecular Physics
Kruislaan 407 Phone: +31 20 6081 275
1098 SJ Amsterdam, The Netherlands E-mail: address@hidden