Merge pull request #315 from mitodl/rebuild

Bumping version and repackaging scripts

Author: jolyonb

README.md

@@ -4,9 +4,9 @@
 
 A library of graders for edX Custom Response problems.
 
-Version 2.3.2 ([changelog](docs/changelog.md))
+Version 2.3.3 ([changelog](docs/changelog.md))
 
-Copyright 2017-2021 Jolyon Bloomfield and Chris Chudzicki
+Copyright 2017-2022 Jolyon Bloomfield and Chris Chudzicki
 
 Licensed under the [BSD-3 License](LICENSE).
 

docs/changelog.md

@@ -6,6 +6,10 @@ This is a major new version with many new features added. We have been very care
 
 ## Version 2.3
 
+### Version 2.3.3
+
+* Refactored sampling to allow sibling variables to be used in `DependentSampler`s.
+
 ### Version 2.3.2
 
 * Fixed a bug where long integers passed in as a variable (via `DiscreteSet`) or user-defined constant would cause an error in evaluation (such numbers are now cast as floats).

docs/grading_math/sampling.md

@@ -242,7 +242,7 @@ Sample from any discrete set of values that are specified in a tuple. A single v
 
 ### DependentSampler
 
-Compute a value for a variable based on the values of other constants/variables. The sampler is simply initialized with the desired formula, which can use any base or user-defined functions (except for randomly sampled functions, see below). `DependentSampler`s can depend on other dependent variables. If you construct a self-referential chain, an error will occur. Note that `DependentSampler` can depend on vector/matrix quantities as well as scalars.
+Compute a value for a variable based on the values of other constants/variables. The sampler is simply initialized with the desired formula, which can use any base or user-defined functions (except for randomly sampled functions, see below). `DependentSampler`s can depend on other dependent variables. If you construct a self-referential chain, an error will occur. Note that `DependentSampler` can depend on vector/matrix quantities as well as scalars. You can even use sibling variables in a `DependentSampler`.
 
 ```pycon
 >>> # Set radius based on the random values of x, y and z
@@ -310,3 +310,6 @@ Finally, if you want to generate a complex random function, set `complex=True`.
 >>> functionsampler = RandomFunction(complex=True)
 
 ```
+
+!!! Note
+    - You cannot invoke a random functions within a `DependentSampler`. This is a limitation of the way that we have implemented random functions.

mitxgraders/version.py

@@ -4,4 +4,4 @@
 """
 from __future__ import print_function, division, absolute_import, unicode_literals
 
-__version__ = "2.3.2"
+__version__ = "2.3.3"