CcpNmr version 2 has now undergone a degree of testing and is relatively stable, but as always there may still be occasional bugs.  We are also trying a few platform-specific binary builds for easy installation.
To report problems either use the ccpnmr mailing list (preferable) or email ccpn-dev [at] bioc.cam.ac.uk.

Upgrading v1 projects to v2

Please not that v1 projects will not automatically load into the v2 software.  The data files first need to be converted.  We have provided an upgrade script which accomplishes this.

Basic Requirements

For developers and those who wish to compile the software

For FormatConverter:

• Python 2.4+
• Tcl/Tk 8.3+

For Analysis:

• Python 2.4+
• Tcl/Tk 8.3+

• Optionally for faster contour graphics: OpenGL
• Optionally for automated assignment: Numpy
  

For the C API

• Python 2.4+

For the Java API

• Java 5+ (i.e. JDK 1.5+)

If you have Python 2.4 you will also need to install ElementTree, and set the environment variable ELEMENT_TREE_PATH to point to the ElementTree installation directory (so this directory should contain the directory elementTree in it) before you run the CCPN installation script.  If you have Python 2.5 and above you do not need to do this.

Downloads v2.1

1. Format Converter: format2.1.3.tar.gz
2. Analysis (source): analysis2.1.3.tar.gz  (this includes Format Converter)
3. Analysis (pre-compiled but includes source):
• 64-bit Linux: analysis2.1.3_1_linux64.tgz
• 32-bit Linux: analysis2.1.3_1_linux32.tgz
• Mac OSX Intel (>= OSX 10.4): analysis2.1.3_1_intel.tgz
• Mac OSX PPC (>= OSX 10.4): analysis2.1.3_1_ppc.tgz
• Windows: CcpnmrSetup-build-2.1.1-4546.exe
  
• Note: for a pre-compiled release you need to unpack it, then "cd ccpnmr" and then run "python configRelease.py" before you use it. 
4. C API: capi2.1.0.tar.gz  (this includes Analysis, as it happens)

Downloads v2.0

1. Format Converter: format2.0.6.1.tar.gz
2. Analysis (source): analysis2.0.7.1.tar.gz  (this includes Format Converter)
3. Analysis (pre-compiled but includes source):
• 64-bit Linux with Tcl/Tk 8.5: analysis2.0.7_linux64.tgz
• 64-bit Linux with Tcl/Tk 8.4: analysis2.0.7_linux64_tcltk84.tgz
• 32-bit Linux with Tcl/Tk 8.5: analysis2.0.7_linux32.tgz
• 32-bit Linux with Tcl/Tk 8.4: analysis2.0.7_linux32_tcltk84.tgz
• Mac OSX Intel (>= OSX 10.4): analysis2.0.7_intel.tgz
  
• Mac OSX PPC (>= OSX 10.4): analysis2.0.7_ppc.tgz
  
• Note: for a pre-compiled release you need to unpack it, then "cd ccpnmr" and then run "python configRelease.py" before you use it.  (This requires the default Python on your computer to be >= version 2.3.  For older versions instead try running "python2.5/bin/python configRelease.py".)  Please email us if the installation doesn't work for whatever reason (in the long run these pre-compiled releases will be the way to go).  The pre-compiled releases include Python 2.5 and Tcl/Tk 8.5 (except where mentioned) and Numpy 1.2.

• Microsoft Windows (32-bit): latest build

4. C API: capi2.0.1.tar.gz  (this includes Analysis, as it happens)
5. Java API jar file: ccpn-xml-2.0.2.jar
• Reference data for use with above: ccpn-data-2.0.2.tgz
6. Java API source distribution: java_xml2.0.2.tar.gz  (includes source code, documentation and API jar file)
   

Note that all of the above will unpack into the ccpnmr directory, and the actual CCPN code will unpack into ccpnmr/ccpnmr2.1 or ccpnmr2.0 (dependent on release), so will not trample existing v1 code.

From release 2.0.3 of the Format Converter and Analysis, the v1 scripts in the ccpnmr/bin directory are (hopefully) copied before the v2 scripts are added.

There is an Analysis v2 basic tutorial and an advanced tutorial.

Analysis specific notes

Currently v1 users will have to download the above gzipped tar file to do the installation.  So you cannot currently update to v2 of Analysis from inside v1 of Analysis.

If the included installation script has problems with the C compilation, then copy ccpnmr/ccpnmr1.0/c/environment.txt to ccpnmr/ccpnmr2.0/c and run the installation script again, answering "n" (no) to the question about creating the environment file.

C API specfic notes

The C API uses the Python API, so is not quite as fast as a pure-C API would probably be.  Further, on some computers you will need extra linking flags than might be expected in order for executables to correctly pick up the Python shared library files.  For more details see:

  http://www.python.org/doc/2.5.2/ext/link-reqs.html

External Modules

For developers and those who wish to compile the software

Various software libraries external to CCPN are required for the releases that use Python. The graphical programs need Python and the Python Tcl/Tk libraries (Tkinter) and Analysis may be run with OpenGL libraries (or the equivalent Mesa package)

Note that on Mac OSX you should always use the Fink versions of Tcl/Tk and Python.

On many linux platforms it is often best to use external modules that come bundled with your linux distribution. These are usually obtained from the installation media or via network download. Examples of required linux packages are given below, but these vary with each distribution. If you cannot, or do not want to use bundled packages the following archives are available, which will install the required libraries with your CCPN istallation: 

• Python - python2.5.tgz
• Numpy - numpy1.2.tgz
  
• Tcl/Tk - tcltk8.5.tgz
• Mesa - mesa6.0.tar.gz
• ElementTree - elementTree.tar.gz (for Python 2.4)

Linux Packages

If you use certain linux distributions, by far the easiest way to install the libraries required to compile CcpNmr software (if you have a network connection) is to use the command line programs like apt-get or yum. This will automatically download native software packages and satisfy any inter-package dependencies. For example, if you're using Ubuntu 9.05, 8.10 or 8.05 you would issue the following command:

sudo apt-get install python-dev python-tk tk8.5-dev libx11-dev freeglut3-dev libglu1-mesa-dev

For Ubuntu 7.10 and earlier you would issue the following command:

sudo apt-get install python-dev python-tk tk8.4-dev libx11-dev freeglut3-dev libglu1-mesa-dev

Similarly, if you're using Fedora Core 6,7 or 8, all you need to do (usually as the root user) is to type at the command line:

yum install python-devel freeglut-devel tcl-devel tk-devel tkinter libX11-devel mesa-libGLU-devel

For Fedora Core 3 & 4 (5?) the equivalent would be:

yum install python-devel freeglut-devel tcl-devel tk-devel tkinter org-x11-Mesa-libGLU xorg-x11-devel

If you have the linux installation media you may also install the required packages with the RPM system, but you may have to resolve various dependencies manually. By way of example, to install the tcl development libraries you could type:

rpm -i tcl-devel-8.4.9-3.i386.rpm

Release History

12 August 2008: First (alpha) release of v2 of the Format Converter and Analysis software (2.0.2 to be precise). 

18 August 2008: Format Converter and Analysis 2.0.3 released.

2 September 2008: C API 2.0.1 released.

10 September 2008: Java API 2.0.1 released.

14 October 2008: Format Converter and Analysis 2.0.4 released.

15 October 2008: Java API 2.0.2 released.

17 October 2008: Analysis 2.0.5 released (very small incremental change over 2.0.4).

3 November 2008: Added 64-bit Linux pre-compiled release for Analysis.

11 November 2008: Added Tiger (OSX 10.4) and 32-bit Linux pre-compiled release for Analysis. (Updated slightly on 19 November to change environment variable settings in scripts in bin directory.)

23 December 2008: Added Numpy to all releases except OSX PowerPC one.

28 January 2009: Format Converter 2.0.5 and Analysis 2.0.6 released.

6 February 2009: Analysis 2.0.6 pre-compiled releases except for OSX PowerPC one.

10 February 2009: Analysis 2.0.6.1 made because had Tk XOR bug which meant crosshair not being drawn.

13 February 2009: OSX PPC 2.0.6.1 release added.

6 April 2009: FormatConverter 2.0.6 and Analysis 2.0.7 released (including pre-compiled releases)

17 April 2009: FormatConverter 2.0.6.1 and Analysis 2.0.7.1 released (pre-compiled releases not updated)

17 September 2009: FormatConverter 2.1.0 and Analysis 2.1.0 released

12 October 2009: FormatConverter 2.1.1 and Analysis 2.1.1 released