GalSim-developers / GalSim / 710
100%
releases/1.4: 100%

LAST BUILD BRANCH: config_truth_fix
DEFAULT BRANCH: releases/1.4
Ran
Jobs 1
Files 84
Run time 5s
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
710

push

travis-ci

rmjarvis 
#811 - InclinedSersic profile (#843)

* Declaring the inclined Sersic profile as a friend of the regular
profile, so methods can be reused.

* Setting up headers for inclined sersic

* Filling in source for inclined sersic

* Adding pysrc files for inclined sersic

* Adding python class for InclinedSersic

* Adding import of InclinedSersic

* Adding InclinedSersic.cpp

* Fixing indentation

* Misc bugfixes - compiles and passes initial sanity checks

* Fixing bug with fillKImage

* Setting up for debug logging

* Some fixes based on tests

* Bugfixes in calculating stepK (should use _re instead) and maxk (should
use value we solved for)

* Setting _ksq_min before solving for maxk and ksq_max

* Fixing n in repr method

* Reordering get methods

* Correcting comment

* correcting getInclination and getScaleHeight

* Correcting __getinitargs__

* got __getinitargs__ working now

* Forcing truncation of Sersic when testing pickling (needed to prevent a
corner-case bug with the tester)

* Making sure it matches Sersic (to limit of ringing effects) in face-on
case

* Updating documentation and changelog

* Trying to fix symlink errors
(#811)

* Updating code with new version that allows cline-args for sersic index
and truncation factor

* Adding inclined sersic images to test against, and code to test against
them

* Fixing bugs in comparing inclined Sersic profiles

* Bugfix in argument order

* Adding in regression test images.

* Removing debug cout statement

* Setting method so we don't get warned about it being changed

* Setting convolution method based on trunc

* Reverting previous commit - always has to be FFT, since there's no
real-space representation.

* Suppressing recommendation to do real-space convolution if any profiles
aren't analytic, since it's impossible in that case

* Adding missing test image to index
(#811)

* Adding missing "of"

* Docstring updates per rmjarvis's comments

* Using R instead of r on both sides of equation

* Adding checks that input parameters are valid

* Setting up to properly accept either scale_height or scale_h_over_r
(needs to be pushed through to C++ layer)

* Correcting to HeightType

* Correcting rType -> hType

* Adding missing "break"s to switch statement

* Correcting __getinitargs__ and __repr__

* Correcting parameters in docstring

* Auto-formatting

* Fixing double minus sign

* Don't need to check in convolution; just set this to say it doesn't have
hard edges

* Refactoring, and setting up sanity checks on InclinedSersic-only
parameters

* Trimming out portions of comment copied from the SBSersic class

* Removing unnecessary overrides

* Noting photon shooting is not yet implemented

* Removing unused _shootnorm

* Removing references to _xnorm

* Moving SBInclinedSersicKValueFunctor to .cpp file

* Setting up inclined sersic to share cache with regular sersic

* Dropping unnecessary serialize comment

* Turning off debugging statements

* Using inverse _re instead of equivalent exponential re

* Always use solver for _maxk and _clipk

* Fixed compilation error

* Missed committing this

* Using _inv_r0 for stepK calculation

* Updating tests to better check stepK

* Fixing calculation of ksq_min for InclinedSersic profile

* Fixing maxSB calculation for InclinedSersic

* Fixing some tests; regression tests aren't working as well now for some
reason, and flux for InclinedSersic isn't always right

* Adding test images to index
(#811)

* Removing unnecessary files
(#811)

* Updating Lance's code, fix to tests. Still doesn't pass flux check, but
I'm pretty sure that's due to me not using infinite truncation radius
with Lance's code - will regenerate it then update it

* Cleaning up how test profiles are assigned to tests, and other test
fixes

* Noting need to fill in Sersic regression tests

* Adding option to specify half_light_radius for exponential profile

* Adding getHalfLightRadius method

* Commenting on use of hlr factor from Exponential class

* Adding all optional parameters to initialization

* Adding new regression test images
(#811)

* Fixing a couple filename-related mistakes

* Fixing up regression tests to properly use rtol and atol

* Implementing Rachel's requested changes to comments/exception type

* Adding tests of InclinedExponential and InclinedSersic

* Removing realspace convolution test from InclinedSersic, since it isn't
applicable

* Uncommenting out other tests

* Fixing stepk test to properly test only pixels that are partly contained
within pi/stepk

* Adding more detailed output for when too much flux is lost to folding

* Removing unneeded exposure of kderiv6

* No longer depending on sersicinfo's kderiv6

* Cleanup of _kderiv6 -> kderiv6

* Missing comma

* Adding tests that expected exceptions are thrown (and making sure they
are)

* Adding tests of value retrieval and making sure they work

* Adding sanity tests for flux_untruncated

* Removing unused test image

* Undoing formatting

* Removing unnecessary test

* Correcting to hType

* No longer TODO

* Readability of indexing

* Adding test case, making sure we hit a couple special branches

* Removing maxSB tests

* Disabling maxSB method - will now throw exception saying it's NYI

* Loosening test requirements for edge-on, correcting inclination angle

* Adding new test image to index

* Fixing mistaken deletion, undoing reorganizing of imports

* Restoring maxSB tests

* Restoring maxSB calculations

* Fixing maxSB methods to work as well as possible and always overestimate

* Removing conservative factor from maxSB estimate

14134 of 14628 relevant lines covered (96.62%)

0.97 hits per line

Jobs
ID Job ID Ran Files Coverage
1 710.1 0
96.62
 Travis Job 710.1
Source Files on build 710
Detailed source file information is not available for this build.