Re: [ccp4bb] further inconsistencies (was: tcltkblt and python distributions)

Protein crystallography Programs for crystallography X-ray detectors
Basic tutorials:
Protein Peptide

CCP4bb navigation

CCP4bb <-- 2007 <-- April 2007 <-- 20 April 2007

Previous message:
Subject: further inconsistencies (was: tcltkblt and python distributions)
From: o {- dot -} h {- dot -} weiergraeber {- at -} FZ-JUELICH {- dot -} DE o {- dot -} h {- dot -} weiergraeber {- at -} FZ-JUELICH {- dot -} DE
Date: 2007-04-20

Next message:
Subject: Re: further inconsistencies (was: tcltkblt and python distributions)
From: "Remacle, F (Francois)" f {- dot -} remacle {- at -} DL {- dot -} AC {- dot -} UK
Date: 2007-04-20

Subject: Re: further inconsistencies (was: tcltkblt and python distributions)
From: Tim Grune tim {- dot -} grune {- at -} SYNCHROTRON {- dot -} VIC {- dot -} GOV {- dot -} AU
Date: 2007-04-20

Hello Oliver,

if by 'conventional' you mean
- download the source code
- ./configure OS && make && make install

I did not share your experience a couple of days ago when I did that. I used
both the ftp and the download interface and nothing was missing.
However from from your mentioning of 'BINARY.setup' you may have meant
installation of the binary version. I never payed attention to that file,
though, even though I also installed a binary set on a PC running CentOS
4.4.
All I did was change a few variable in $CCP4/include/ccp4.setup-bash:
CCP4_MASTER to the correct root of the installation
CCP4I_TCLTK to /usr/bin to use the system wish
and remove the MANPATH entries since with an unset MANPATH bash (or man
itself, not sure) figures out the correct path from PATH - which makes
maintenance a lot easier.

For CentOS and Debian I also have to change bltwish to wish in the
tcl-scripts under $CCP4/ccp4i/bin since in both systems blt comes as a
library.

Finally I used good old g77 instead of gfortran since the latter made mosflm
crash with a segmentation fault as soon as it tried to display an image.

Maybe you can just try to re-download it? Unless this was all for a
different platform in which case my advise may not be applicable - sorry.

Cheers, Tim

-----Original Message-----

From: o.h.weiergraeber@FZ-JUELICH.DE

To: CCP4BB@JISCMAIL.AC.UK

Date: Fri, 20 Apr 2007 10:44:37 +0200

Subject: [ccp4bb] further inconsistencies (was: tcltkblt and python
distributions)

Dear ccp4 developers,

when trying to do a conventional install of ccp4 (i.e. avoiding the
install.sh script), I ran across additional problems/inconsistencies with
the package:

1. In the tarball I freshly downloaded this morning, the cctbx tree under
ccp4-6.0.2/lib was completely missing!!!

2. The BINARY.setup script checks for src/phaser, which does not exist
(instead there are phaser-ucs and phaser-utf directories). Simply creating
the phaser directory may resolve this, but of course the phaser installation
is corrupted due to the missing cctbx files.

It would be nice if the responsible persons could comment on this. Is the
conventional install no longer supported or not even checked?

Greetings

Oliver

-------------------------------

Dr. Oliver H. Weiergraeber

Institute of Neurosciences and Biophysics

Molecular Biophysics

Research Centre Juelich

D-52425 Juelich

Germany

Phone: +49-2461-612028

Fax: +49-2461-612020

-------------------------------

CCP4bb navigation

CCP4bb <-- 2007 <-- April 2007 <-- 20 April 2007