Dear Miguel, Thanks for this helpful response. I now managed to get the svn to work and have reproduced the problem with the latest Jmol.
At the moment, all the Molpro interface does is to pull cartesian coordinates and vibrational modes. This can be tested and demonstrated adequately with just one example file, which I have put at http://theory.chem.cf.ac.uk/knowles/phoh.xml . This is essentially the same example as the previous molpro/vib.xml which was in Jmol's source tree but doesn't seem to be now; however the schema changed a little since last year, so the old file should be discarded. In order to validate, this file should load correctly, and it should be possible to display animated vibrational modes. I do not know what you have set up to check essentially interactive features like this, but I guess we have learned that we need to keep testing Jmol here so that problems are trapped at the right time. Should I do something further to help resolve the bug, or is it best to leave it to someone else? Best regards, Peter -- Prof. Peter J. Knowles School of Chemistry, Cardiff University, Main Building, Park Place, Cardiff CF10 3AT, UK Telephone: +44 29208 79182 Fax: +44 2920874030 Email: [EMAIL PROTECTED] Web: http://www.cf.ac.uk/chemy/staff/knowles.html >>> "Miguel" <[EMAIL PROTECTED]> 02/06/2006 14:12:22 >>> > Dear Miguel, > > You might remember that about a year ago we discussed > the inclusion into the jmol source of some changes I > made in order to parse Molpro output files. Yes > At the time, the code was at version 10.0. We are just > about to release a new version of Molpro that includes > the facility to produce facility that jmol > should recognize. Good > In the final stages of checking, we have pulled down > the latest stable version of jmol, ie 10.2, and > unfortunately the Molpro interface no longer works > - it crashes somehow when parsing the file. I apologize ... we must have introduced some type of bug. > Unfortunately, I can get nowhere in trying to debug: > it is impossible to use the full 10.2 > tarball on sourceforge, since it does not build (ant > says missing build-i18n.xml); Hmmm ... none of the jmol developers ever build from the tarball, so I am sure that this was never tested. Nico, can you look into this? > svn.sourceforge.net seems to be dead so no possibility to > pull the latest code. sourceforge seems to have had some problems over the past few weeks, but svn.sourceforge.net is working OK for me right now. > Can you please suggest a way forward? Please retry svn.sourceforge.net > I am ready to do all the work, but at the > moment am completely stuck. Please send all messages to [email protected] so that other Jmol developers can offer assistance and so that they are aware of your issues. Note that Nico did some good work with using JUnit to build some test verification for the file readers. These tests were put in place to help ensure that we do not break things with the file readers. We have a molpro directory, but it only has one file in it ... vib.xml ** 2 minutes later ** Hmmm ... molpro/vib.xml is not being tested. We should get this bug resolved. Then we should work with you to identify some good test files to put in the molpro directory ... and we should enable the tests. Egon and Nico are more familiar with the XML file formats than I am, so they can probably offer better assistance than I can. Again, please join [email protected] and send messages there until we get this resolved. > It would be nice to be able to tell Molpro users > that there is a version of jmol that they can use. Agreed. > Thanks very much. Thank you. Miguel _______________________________________________ Jmol-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jmol-developers
