aimmspy
Release Notes
This document provides a summary of the changes made in each release of the aimmspy
module. It includes bug fixes, improvements, and new features introduced in each version.
Version 25.2.1.2 (August 15, 2025)
Improvements
Added
multi_data()
method onProject
instance to retrieve data from multiple AIMMS identifiers in a single call.Added
show_dependencies()
method to quickly list the versions of all modules being used.Assignment to set identifiers initially required a Python list object. Now also Python set objects are accepted.
Resolved Issues
Fixed an issue where it was not possible to initialize a 1-dimensional AIMMS identifier declared over an AIMMS calendar using a Python dictionary in which the keys were
datetime
objects.
Version 25.1.1.11 (July 29, 2025)
Improvements
Initial release of the
aimmspy
module.The
get_model
method can now be called multiple times; it will update the model if it has changed since the last call. This is useful when the model is updated in the AIMMS IDE or changed dynamically in the AIMMS project.Improved error messages when attempting to extend a read-only domain set.
Added
AimmsException
andAimmsPyException
classes to handle AIMMS-specific exceptions more gracefully.Changed the way to refer to identifiers that are declared inside an AIMMS namespace. We now use the library (or module) prefix instead of the library (or module) name (similar to how it is done in the AIMMS language itself).
Introduced a
Model
class that can be retrieved using theget_model()
method on theProject
instance. This change allows to automatically create a stub file that is needed/used to access exposed AIMMS identifier in your Python script.Added a
mapping
option to deal with differences between column names om a date frame and AIMMS identifiers.Added a
return_type
option to specify the return type on a perdata()
call basis.Added a
extendType
option to control the behavior when adding new elements to a domain set.
Resolved Issues
Fixed an issue where it was not possible to make an assignment to a scalar element.
Fixed an issue where consecutive calls to
data()
for the same identifier with different data return types did not work correctly. For example, calling first withreturn_type=DataReturnTypes.ARROW
and then withreturn_type=DataReturnTypes.PANDAS
would always return the same data return type as the first call. This has been resolved.Due to an issue with the Polars Python module in combination with Python 3.11 on Windows (see https://github.com/pola-rs/polars/issues/23142), the Polars library was downgraded from version 1.31.0 to 1.30.0. This is a temporary measure until the issue is resolved in a future release of Polars.
Fixed an issue when reading a integer column from a data frame into an AIMMS identifier.
Version 1.0.1post23 (May 21, 2025)
Improvements
Initial beta release of the
aimmspy
module.