Merge pull request #1434 from RosettaCommons/rhiju/rna_denovo_setup
Rhiju/rna denovo setup
Some small updates to input & help in rna_denovo
• iccbuild is failing due to license issue on test server
• in integration tests, new test `rna_puzzle5_P15P3P8_noP4P6_4RB_G208phosphate_ligation` is having a problem -- will fix in a separate pull request.
• `hybridization` test was broken previously.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #1432 from RosettaCommons/rhiju/rna_denovo_setup
Rhiju/rna denovo setup
+ integration tests for `rna_denovo` give cosmetic changes -- all confirmed/checked locally.
+ some new integration tests `rna_puzzle_*`.
+ other integration tests that change only do so in linux.release_debug or mac.clang but not both, suggesting uninitialized variables or other stochastic choices depending on system state.
+ icc build broken on Hojo3, but error message is about license problem, so i don't think that's this pull request.
+ profile may have changed in 1-2 tests, but getting timeout on test server. 1-2 profile test changes in previous build appear within bounds of variability of these tests.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #1428 from RosettaCommons/roccomoretti/integrationpy_fixes
Several integration.py fixes
* absolute directories were making it into the integration test diffs
* diffs were broken for the regular integration tests run locally. (There was an extra "test" messing up the diff command.)
* re-running single tests resulted in an error message. (Again, ultimately due to extra "test" parameters.)
* there were error messages during the SHA1 generation, due to not having a tools or demos checkout at the appropriate locations.
* Intermittent encoding issues on integration.demos
This pull request does not fix the missing valgrind_results.yaml issue, but I felt the current fixes need to go in as soon as possible, as we're getting spurious diffs on the test server.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #1427 from RosettaCommons/roccomoretti/job_completed_debug
More Debugging output for job completed status
In JD2 JobOutputters, add debug-level output for diagnosing
that jobs are skipped, and also why jobs are skipped.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #1398 from RosettaCommons/tlinsky/tomponents
Updates to denovo classes, move some more things to master
Integration test changes expected for code_templates_src -- this is due to some cleanups to the Mover template. I added virtual keywords in the class declaration where appropriate for readability, and cleaned up spacing.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #1390 from RosettaCommons/jadolfbr/fix_nbr_selector
Jadolfbr/fix nbr selector
1) Re-enable the ability to embed selectors as the focus for NeighborhoodResidueSelector. Thanks to @basantab for showing me that I broke it in the first place! To make something useful come out of this error, the code for grabbing embedded selector(s) has been generalized and moved to ``` core/select/residue_selector/util.hh ``` You can use this to grab embeded selector(s) from anything (movers, taskops, etc.)
2) Change the name of glycan and antibody selector RS names to fit better within the short-name abbreviations used for parsing selectors in RS.
- Finish documentation for selectors:
- https://www.rosettacommons.org/docs/wiki/scripting_documentation/RosettaScripts/TaskOperations/taskoperations_pages/ResidueSelectors#residueselectors_conformation-independent-residue-selectors_carbohydrate-residue-selectors
- https://www.rosettacommons.org/docs/wiki/scripting_documentation/RosettaScripts/TaskOperations/taskoperations_pages/ResidueSelectors#residueselectors_conformation-independent-residue-selectors_antibody-residue-selectors
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #1426 from RosettaCommons/vmullig/fix_pepbond_atoms
Fixing minor issue in simple_cycpep_predict application
Peptide bond-dependent atoms at cyclization points were sometimes not being updated properly. Now they should be okay.
Small trajectory changes expected in simple_cycpep_predict integration tests.
Tasks:
- Add a correction step for the positions of those atoms following FastRelax or FastDesign steps.
- Beauty.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #1410 from RosettaCommons/jadolfbr/doc_xrw2016
Overview
=======
This PR is part of the Documentation XRW 2016 effort and enables demo [integration] testing through ```integration.py```
This is done through an additional ```--demos``` and ```--tutorials``` cmd-line option, which indicate to the integration script to either run the demo or tutorial tests. Additionally, single scripts can be passed at the end of the command line (Either names or relative paths or full paths or what have you.).
Tests are run for everything in the tutorials directory and public directory (which will change before the final merge.)
How it works
=========
If a ```command``` file is in the demo directory, we use that. Alternatively, we parse all .md files looking for ```$>``` characters in the beginning of a (properly stripped) line, which indicates to test this command.
If there are flags files, there can be additional flags.short files that the integration test is looking for in the command. If these exist, it will add them to the command. This is to enable short-testing runs, without having to maintain two separate versions of flags files. While some options cannot be overridden in this way, this works for most cases and is simple to maintain. For example, this command in an md file (with a .short file in the directory):
```
$>../../../main/source/bin/score_jd2.default.linuxclangrelease -s 1ubq.pdb -out:suffix crystal @crystal_score_flags
```
Will be written to the output command file as:
```
%(bin)s/score_jd2.%(binext)s -s 1ubq.pdb -out:suffix crystal @crystal_score_flags @crystal_score_flags.short -database %(database)s -run:constant_seed -nodelay 2>&1 | egrep -vf path/to/ignore_list > log1
```
Individual tests can also be given, but the full relative path or the name of the demo will need to be given, for example,``` tutorials/Tutorial_4_Relax``` in addition to the ```--tutorials``` option (or ```--demos``` if this was a demo).
This should allow automated demo testing using our current integration test system, without forking integration.py.
Normal integration test changes
========================
Integration test changes are cosmetic and seem to be doing what that code block was originally intended to do.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #1420 from RosettaCommons/rfalford12/fix-using-namespace-membrane-offenders
Rfalford12/fix using namespace membrane offenders
This pull request removes illegal using namespace statements from membrane-related classes in core.
The following files are updated from the list of shame emailed out to everyone:
src/core/scoring/membrane/FaMPEnvEnergyCreator.hh
src/core/scoring/membrane/FaMPEnvSmoothEnergy.cc
src/core/scoring/membrane/FaMPEnvSmoothEnergy.hh
src/core/scoring/membrane/FaMPEnvSmoothEnergyCreator.hh
src/core/scoring/membrane/FaMPSolvEnergyCreator.hh
src/core/scoring/membrane/MPCBetaEnergy.hh
src/core/scoring/membrane/MPCBetaEnergyCreator.hh
src/core/scoring/membrane/MPEnvEnergy.hh
src/core/scoring/membrane/MPEnvEnergyCreator.hh
src/core/scoring/membrane/MPLipoEnergy.hh
src/core/scoring/membrane/MPLipoEnergyCreator.hh
src/core/scoring/membrane/MPNonHelixPenalty.cc
src/core/scoring/membrane/MPNonHelixPenalty.hh
src/core/scoring/membrane/MPNonHelixPenaltyCreator.hh
src/core/scoring/membrane/MPPairEnergy.hh
src/core/scoring/membrane/MPPairEnergyCreator.hh
src/core/scoring/membrane/MPTMProjPenalty.cc
src/core/scoring/membrane/MPTMProjPenalty.hh
src/core/scoring/membrane/MPTMProjPenaltyCreator.hh
src/core/scoring/membrane/MPTerminiPenalty.cc
src/core/scoring/membrane/MPTerminiPenalty.hh
src/core/scoring/membrane/MPTerminiPenaltyCreator.hh
src/core/scoring/membrane/MembraneData.cc
notify author
notify list [rosetta-logs@googlegroups.com]