Version 2.4.1 stable

This release implements an old feature request, concluding all unresolved issues with this package. At this stage, we do not anticipate upgrading this library further, unless edX performs a python/package version bump that breaks backwards compatibility.

Version 2.4.0

This version includes a major performance update through lazy loading of unnecessary libraries.

Version 2.3.1 stable

This incorporates all of the updates implemented since the version 2.0 release. At this stage, the library is stable, robust, and well-tested. We do not anticipate making any major changes to this library until edX considers the adoption of python 4.0.

Version 2.0.0

This is a major update to the library, and contains many new features. We recommend viewing the change log in the documentation site to see a list of all updates.

The primary cause of the new major version number is the update to support both Python 2.7 and Python 3.6/3.7. This required breaking internal backward compatibility in a number of locations, and may cause old custom code to break.

Version 1.2

This release includes a number of enhancements to the grading library.

• A new documentation website
• Math parser now supports multivariable functions and array input (vector, matrix, etc)
• Many improvements to our MathJax preprocessor
• Improvements to balanced bracket validator
• Added new class MatrixGrader along with supporting sampling classes RealMatrices and RealVectors
• When FormulaGrader (and its subclasses) are used inside an ordered ListGrader, authors can now grade multiple student inputs in comparison to each other by specifying answers in terms of sibling variables
• FormulaGrader (and its subclasses) now support comparer functions that can be used to grade student input more flexibly. For example, rather than checking checking that the student input and author input are equal, check that they are equal modulo a certain number. We include a number of builtin comparers.

Version 1.1

This release includes a number of enhancements to the grading library.

• Dependent sampling, tensor variable names, numbered variables and custom comparison functions in FormulaGrader
• Complex integrands in IntegralGrader
• Better error messages to students
• A number of minor bug fixes
• Case insensitive comparison in FormulaGrader has been removed (this breaks backwards compatibility for those who used case-insensitive input for formulas)

Version 1.0 - Initial Public Release

This is the initial public release of the MITx Grading Library!