Hi, I would support this. To implement this, I would guess that the pipe_control.value module would need changing. Are you able to create a system test for this? Implementing a system test before changing the code would be the best as it will guarantee that the headers are not accidentally changed when code refactorisations occur (which is a relatively common event in relax, thanks to the test suite). I would also suggest changing the behaviour for dictionaries to match that for single values and lists by having the following header created: # Parameter description: The transversal relaxation rate for state A in the absence of exchange. # # mol_name res_num res_name spin_num spin_name value_SQ_CPMG_-_599.89086220_MHz error_SQ_CPMG_-_599.89086220_MHz This consistency would make parsing of different value files easier in the future. The header for single value parameters look like: # mol_name res_num res_name spin_num spin_name value error And for lists elements (here index 2): # mol_name res_num res_name spin_num spin_name value_2 error_2 For the implementation, you should really test this using Python 3 as well. This is another reason why the system test will be of benefit. String operations have changed significantly in Python 3, but it is easy enough to use a solution which works for both Python versions. Cheers, Edward On 19 February 2014 13:02, Troels Emtekær Linnet <tlinnet@xxxxxxxxxxxxx> wrote:
Hi Edward. If I use the function: value.write(param='r2a') The header file looks like this: # Parameter description: The transversal relaxation rate for state A in the absence of exchange. # # mol_name res_num res_name spin_num spin_name SQ CPMG - 599.89086220 MHz sd(SQ CPMG - 599.89086220 MHz) Ideally I would like it to have it similar to this: # Parameter description: The transversal relaxation rate for state A in the absence of exchange. # # mol_name res_num res_name spin_num spin_name SQ_CPMG_-_599.89086220_MHz sd(SQ_CPMG_-_599.89086220_MHz) so a string.split() will be of the same length as the data columns. Best Troels _______________________________________________ relax (http://www.nmr-relax.com) This is the relax-devel mailing list relax-devel@xxxxxxx To unsubscribe from this list, get a password reminder, or change your subscription options, visit the list information page at https://mail.gna.org/listinfo/relax-devel