Legislative

[deanthedream]

Update of Survey Simulation resolving differences between master (mostly comments) and legislative. keepStarCatalog logic implemented. IPClusterEnsemble now properly handle imports on clusters runQueue upgraded with new path logic and more robust hanging process termination and restart adding plotKeepOutMap adding plotJointPDF Map SLSQPscheduler with incorporated fZmin, fZmax in calculation as well as integration of SLSQP A-J.

[coveralls]

Coverage decreased (-3.5%) to 54.234% when pulling 420ea9fc2c223376e8e6122ab9220b21c6966079 on legislative into 935ca517c756c53f235d0bcaa7702bf54817d95f on master.

[dsavransky]

A few things to fix before I merge:

1. For some reason, some of the commits add random amounts of trailing white space on blocks of lines (e.g., see lines 742-755 in SurveySimulation.py). This has the effect of making the code more fragile and also makes merges harder, so it'd be good to fix these (not really sure how that white space is getting added in the first place).
2. The new assert in geOutSpec (line 1739 of SurveySimulation.py) will fail when keepStarCatalog is True. As discussed, instead of this assert, you should clone the logic from TargetList that just uses the name of the module in the case where it remains a module. No assert should be necessary here.
3. The new logic in generateHashfname (lines 1794-1805 in SurveySimulation.py) is entirely based on use of .has_key(). As discussed, please change all instances of this to "key in dict" style statements.
4. The new logic for keepStarCatalog (lines 163-173) is now partially redundant with the leftover orig logic (lines 174-175). There is no reason for the else: clause to be there if its just a single pass statement. Finally "if not keepStarCatalog" is preferred to "if keepStarCatalog == False".
5. In SLSQPScheduler.py, why are all of the original self.vprint info messages commented with #DELETE ? I would prefer these be left there - vprint allows us to toggle off such outputs as needed. Conversely, there's a new vprint statement in self.inttimesfeps (lines 287) that I don't think needs to be there.
6. In general, there's a lot of debugging stuff left commented out in SLSQPScheduler.py, that I think should be removed (or just left on a different debug branch, but shouldn't get into master).
7. It looks like the updates to SLSQPScheduler fold in the functionality currently scattered in version A-H, but I'm not seeing a delete of those files in this commit. Is there still unique functionality left in those?

Thanks.

[deanthedream]

2-7 are done. Please check my fix on 4 to see if the final result is what you expect.

[deanthedream]

Fixes Issue #201 , #200 , #178 , #168