~ Latest available version: 1.3.2 ~
What is ccwatcher?
ccwatcher monitors the progress of computational chemistry calculations during their runtime. It parses important output and plots SCF energies. ccwatcher is platform-independent thanks to Python and the Qt Toolkit.
Features:
Supports most 'big' computational chemistry programs
Qt GUI and CLI interface
Platform independent
Interactive parsing, plotting and updating of logfile info
Energies plotting via Qwt (free zooming) or Gnuplot (command-line plotting, PNG export)
Supports multi-file load and comparison
ccwatcher currently supports Gaussian, ORCA, Turbomole, Molpro, GAMESS(US), NWChem, Jaguar, Firefly (Ex-PCGAMESS) and ADF. Check out the screenshots gallery for some visual impressions.
Any feedback is appreciated!
Why ccwatcher instead of ...?
There are of course other programs for inspecting computational output; However, all have different usability goals. ccwatcher's unique feature is first of all the ability to monitor the progress of the calculations and update energy plots and parse text messages. To the best of my knowledge, that is so far otherwise only partly supported by the commercial ADF package's 'movie' tool. For linux cluster environments, gnuplot's ASCII plotting is a very uncomplicated way to inspect your jobs (I believe this is also unique). ccwatcher also gives you the ability to compare energy traces of several files. The application is intended to be lightweight and small so it can provide you with an overview of the progress of a 50 MB calculation within seconds.
If you are interested in what other programs have to offer, please look at Gausssum and QMForge. An interesting point is that while QMForge and GaussSum use cclib as a backend, ccwatcher meanwhile uses its own routines to give you access to even more information and text messages. For molecule inspections I would recommend the great projects Avogadro and Gabedit. And no, I don't consider them competition but partners in the free software universe.
License
ccwatcher is placed under the GPL. That means you can download, use, modify and redistribute it for free, amongst other freedoms. Licensing of material other than code (icons etc.) can be found in the license.txt file of the package, but be assured everything follows an open-content philosophy.
Documentation
As I hope ccwatcher is rather self-explanatory, there's not too much. Please look at the files "README.txt" and "doc.txt" in the program package (you could use the online access to my mercurial trunk to look at them, too).
Installation:
I have made some effort to make installation for most users rather easy, so that not many of you should have to install the dependencies manually, except maybe Python. Detailed installation instructions can be found in the README.txt file.
- Windows: The Py2Exe-Installer (ccwatcher_setup_windows.exe) should work for most Windows versions (only XP+7 tested) - you do not need to install Python anymore! If you get a MSVCP90.DLL error, you need to install the Microsoft Visual C++ 2008 (VC2008) Redist Pack. As a fallback, please use the 'win32-dynamic-python27' version and install a 32bit(!) Python2.7 before. Please report problems.You should hopefully have no problems with the new exe installer. If it doesn't work, please use the fallback as long as you install Python 2.x before. However, please make sure you have a 32bit Python installed - even if your Windows is 64bit. EMail me if something does not work.
- Linux: For quite a few distro's, I have built binary packages using the openSUSE build service. So if you run a popular distro, there should be a RPM, DEB or TAR.XZ file that works for you. Mageia users use the Mandriva2011 RPM. CentOS users: I do not think there are python-qwt bindings for you. One thing to mind when using the binaries is that I simply declared PyQt, PyQwt AND Gnuplot as dependencies. If there is no binary package for your distro, I think installing the latter via your package manager and using ccwatcher without formal installation is the simplest option.
- Others: Any system with Python 2.x, PyQt and either Gnuplot or Pyqwt installed should be able to run ccwatcher (download the 'src' package). Please see README.txt for more Information.
Development and Bug reporting
I'd be happy if you'd like to help improve ccwatcher. Please report bugs and whishes to the sourceforge tracker. If you think a translation is necessary, contact me.
If you own an Apple Mac Operating System, It would be great if you could give me information on ccwatcher running on that OS, as I haven't had any chance to test this. Thanks!
The latest (unstable!) ccwatcher version can always be acquired via my mercurial trunk - just "hg clone http://hg.code.sf.net/p/ccwatcher/code ccwatcher-code" if you have Mercurial installed.
If you'd like to contribute code, you are of course very welcome. ccwatcher is pure python and small, so it's easy to start - and there are "junior jobs" if you're a newbie. If you're a C(++) programmer or a good coder, or if you are an artist or chemist please also consider contributing to more important projects like:
Avogadro |
cclib |
openbabel |
kalzium |
Contact
Please use the means of communication given at the sourceforge project page. If that is for some reason not possible, try xaverxn AT users.sourceforge.net .
