Use of $. problem

[GoogleCodeExporter]

Just was made aware of an overlooked bug in the restraints NMR-STAR files. The 
'$' character at the 
start of a string value is a reserved character to indicate that the value is a 
save frame framecode. 
The use of '$.' to indicate a null save frame framecode is not valid. The value 
should be just a 
period.

I hope this will be an easy fix.

Original issue reported on code.google.com by schulte....@gmail.com on 8 Oct 2009 at 6:44

[GoogleCodeExporter]

Took me a while to find, somewhere deep down in the code, but is fixed now. 
There is
still an issue with 'empty' software tables (e.g. _Conformer_family_software) - 
is
this a problem for you or not?

Original comment by wfvran...@gmail.com on 9 Oct 2009 at 12:34

[GoogleCodeExporter]

Also, update python/ccp/format/nmrStar directory for this fix to work

Original comment by wfvran...@gmail.com on 9 Oct 2009 at 12:35

[GoogleCodeExporter]

Every time I try to do an update this morning I get "Connection reset by peer."

Original comment by schulte....@gmail.com on 9 Oct 2009 at 3:01

[GoogleCodeExporter]

The sf site seems to be up for me.
Perhaps D. needs to open something up on grunt? Or are you on tang?

jd:Stella/ccpn/ cvs -z3 -d:ext:jurgenfd@ccpn.cvs.sourceforge.net:/cvsroot/ccpn 
update -r stable
jurgenfd@ccpn.cvs.sourceforge.net's password: 
? .DS_Store
? .project
? .pydevproject
? data/recoord
? model/doc/cambridge
? model/doc/ccp
? model/doc/ccpnmr
? model/doc/help.html
...

Original comment by jurge...@gmail.com on 9 Oct 2009 at 3:11

[GoogleCodeExporter]

Got it to work now. The fix is there on tang (no more $.). Maybe another update 
on grunt will get that working.

Original comment by schulte....@gmail.com on 9 Oct 2009 at 3:20

[GoogleCodeExporter]

Sorry, I did not spot the $. issue earlier.

The missing _Conformer_family_software loop is not a problem. It does not look 
easy
to distinguish the software used for calculating the structure from those used 
for
other purposes within the cif file.

Thanks,
Eldon

Original comment by webmas...@bmrb.wisc.edu on 12 Oct 2009 at 2:40

[GoogleCodeExporter]

Hi,

The $. problem still exists on grunt (at least the files being generate at the
moment, see 2kon). Please move the fix to grunt and regenerate the files.

Thanks,
Eldon

Original comment by Eldon.Ul...@gmail.com on 13 Oct 2009 at 9:53

[GoogleCodeExporter]

That was updated after the tang update that fixed the problem. Did the fix make 
it to the stable version of 
recoord?

Original comment by schulte....@gmail.com on 13 Oct 2009 at 10:40

[GoogleCodeExporter]

It's all in the repository, must be an update problem on your side... .

Original comment by wfvran...@gmail.com on 14 Oct 2009 at 7:56

[GoogleCodeExporter]

I'll redo it and see what happens.

Original comment by schulte....@gmail.com on 14 Oct 2009 at 1:37

[GoogleCodeExporter]

I've reinstalled this and made test runs that are on grunt, but not the public 
site.
see
http://grunt.bmrb.wisc.edu/NRG/MRGridServlet
and try if you have access.
2kbe, 2kbf, 2kok, and 2kon 

I've tested these with and without doSurplus, doViolAnal, etc. and found no 
differences in the output other than 
dates created and block ids. Therefore, I will run any large batches without 
these turned on. 

Original comment by schulte....@gmail.com on 14 Oct 2009 at 2:51

[GoogleCodeExporter]

http://grunt.bmrb.wisc.edu/NRG/MRGridServlet

is not available to outside use; nor should it be. My point is: no need to 
reference it here because we can't look 
at it all.

Original comment by jurge...@gmail.com on 10 Nov 2009 at 12:05

[GoogleCodeExporter]

Ok to close this issue?

Original comment by jurge...@gmail.com on 19 Nov 2009 at 9:08

• Changed state: Fixed