Anonymous | Login | 2024-10-11 12:14 CDT |
My View | View Issues |
View Issue Details [ Jump to Notes ] | [ Issue History ] [ Print ] | ||||||||
ID | Project | Category | View Status | Date Submitted | Last Update | ||||
0000005 | Rosetta | [All Projects] Incorrect Results | public | 2012-06-19 11:18 | 2012-06-27 15:46 | ||||
Reporter | smlewis | ||||||||
Assigned To | smlewis | ||||||||
Priority | high | Severity | minor | Reproducibility | always | ||||
Status | resolved | Resolution | fixed | ||||||
Platform | x86 | OS | Linux | OS Version | Ubuntu | ||||
Product Version | Trunk | ||||||||
Fixed in Version | |||||||||
Summary | 0000005: something is lurking in ConstraintSet.show_definition() that alters ostream numeric precision | ||||||||
Description | running ConstraintSet.show_definition apparently leaves the input ostream object (Tracer) in an altered state w/r/t numeric precision | ||||||||
Steps To Reproduce | add a ConstraintSet.show_definition call to some code that outputs numbers to tracers and didn't already have one (UBQ_E2_thioester at the end of init_on_new_input is a good choice) | ||||||||
Additional Information | integration test diff from adding a ConstraintSet.show_defintion(TR, pose) call: < apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 1 1401.59 1401.59 --- > apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 1 1402 1402 73c143 < apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 2 1401.59 1401.59 --- > apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 2 1402 1402 75c145 < apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 3 955.906 955.906 --- > apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 3 955.9 955.9 77c147 < apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 4 921.04 921.04 --- > apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 4 921 921 79,80c149,150 < apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 5 909.923 909.923 < apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 6 909.923 909.923 --- > apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 5 909.9 909.9 > apps.public.scenarios.chemically_conjugated_docking.UBQ_E2_thioester: 6 909.9 909.9 Looks like a change in numeric precision to me. This should be fixed so that precision is reset before the function exists. (also, from cursory inspection, it's probably deeper than ConstraintSet....) | ||||||||
Tags | No tags attached. | ||||||||
Application(s) Affected | all | ||||||||
Command Line Used | any | ||||||||
Developer Options | |||||||||
Fixed in SVN Version | 49645 | ||||||||
Attached Files | |||||||||
Notes | |
(0000015) smlewis (Administrator) 2012-06-26 16:51 |
The bug is BoundConstraint.cc:124, where setprecision() is used without resetting to the input value. A fix is underway. |
(0000018) smlewis (Administrator) 2012-06-27 15:46 |
The bug was in BoundConstraint::show_definition; it used setprecision without ever resetting precision |
Issue History | |||
Date Modified | Username | Field | Change |
2012-06-19 11:18 | delucasl | New Issue | |
2012-06-19 11:18 | delucasl | Status | new => assigned |
2012-06-19 11:18 | delucasl | Assigned To | => smlewis |
2012-06-19 11:30 | smlewis | Reporter | delucasl => smlewis |
2012-06-26 16:51 | smlewis | Note Added: 0000015 | |
2012-06-27 15:46 | smlewis | Fixed in SVN Version | => 49645 |
2012-06-27 15:46 | smlewis | Note Added: 0000018 | |
2012-06-27 15:46 | smlewis | Status | assigned => resolved |
2012-06-27 15:46 | smlewis | Resolution | open => fixed |
Copyright © 2000 - 2012 MantisBT Group |