URL: <http://gna.org/bugs/?10022> Summary: string representation test of previous vs. current models in full_analysis Project: relax Submitted by: dkojetin Submitted on: Monday 09/17/2007 at 23:07 Category: relax's source code Severity: 4 - Important Priority: 5 - Normal Status: None Privacy: Public Assigned to: None Originator Name: Originator Email: Open/Closed: Open Discussion Lock: Any Release: Repository: 1.2 line Operating System: None _______________________________________________________ Details: (relax 1.2 svn revision 3357) Comparing string representation of the previous and current runs within the full_analysis.py convergence definition fails. When unassigned/deselected residues in the previous run are loaded in the load_tensor definition, information for all residues is read, including residues that have 'None' as a model in the previous run. However, information from the current run does not contain the 'None' model for unassigned/deselected residues. Therefore, the literal string test between lines 373-378 fails when the string representation of the previous run (the 'prev_models' variable, created between lines 363-365) is directly compared to the string representation of current run (the 'curr_models' variable, created between lines 368-371) within the convergence definition (lines 373-378). The attached patch corrects this error by changing the literal string test between the prev_models and curr_models variables, created by concatenating the model information, to a direct residue-by-residue comparison of models such that residues with a model of 'None' are not included in the comparison. The code could likely be modified to contain fewer lines, but it seems to work as is, although it should be tested. _______________________________________________________ File Attachments: ------------------------------------------------------- Date: Monday 09/17/2007 at 23:07 Name: patch Size: 3kB By: dkojetin <http://gna.org/bugs/download.php?file_id=2962> _______________________________________________________ Reply to this item at: <http://gna.org/bugs/?10022> _______________________________________________ Message sent via/by Gna! http://gna.org/