Skip to content

Releases: sbmlteam/libsbml

Release 5.20.4

07 Aug 11:39
Compare
Choose a tag to compare

This is mostly a bug fix release enhancing stability and improving
performance when validating models. Main points:

  • support compilation when using libxml 2.12.0 and later
  • improved performance when testing for assignment cycles
  • cmake build system honors the BUILD_SHARED_LIB flag if defined

What's Changed

New Contributors

Full Changelog: v5.20.2...v5.20.4

v5.20.2

21 Sep 11:21
Compare
Choose a tag to compare

This is mostly a bug fix release enhancing stability when validating
and converting models.

  • New features / improvements

    • Access to error logs from the C API

    • Attributes of the ModelHistory can now be accessed from
      const objects.

    • The FBC V3 package has been updated to reflect the latest
      specification. There are now optional attributes fbc:reaction2
      on the FluxObjective and fbc:variable2 on the
      UserDefinedConstraintComponent.

  • Bug fixes

    • A bug converting / validating models with the 'comp' package
      has been fixed.

    • A crash when inferring units has been resolved.

    • A fix to the CMake Build system ensures that check for dependencies
      is more robust. Thanks to first time contributor @adelhpour

    • The type of the FBC V3 coefficient attribute on the
      UserDefinedConstraintComponent has been corrected, it was mistakenly
      a double before, now it has been corrected to be a SIdRef as
      described in the specification.


FILES

Source

libSBML src archive zip
libSBML src archive tar.gz

Documentation

C++ API Documentation

MATLAB bindings for all OS

libSBML-5.20.2-matlab-binaries.tar.gz
libSBML-5.20.2-matlab-binaries.zip

Python bindings

Available as python-libsbml from PyPi.

R bindings

libSBML-5.20.2 src archive

v5.20.1

25 May 09:45
Compare
Choose a tag to compare
v5.20.1 Pre-release
Pre-release
bump version

v5.20.0

03 Apr 11:07
fdfda61
Compare
Choose a tag to compare

The SBML Team with the addition of new members from ARC at UCL is pleased to announce the 5.20.0 release of libSBML.


FILES

Source

libSBML src archive zip
libSBML src archive tar.gz

Documentation

C++ API Documentation

MATLAB bindings for all OS

libSBML-5.20.0-matlab-binaries.tar.gz
libSBML-5.20.0-matlab-binaries.zip

Python bindings

Available as python-libsbml from PyPi.

R bindings

libSBML-5.20.0 src archive


NEWS

Given the length of time since our last release the list of changes
and acknowledgements of contributions is incomplete. We apologise and
offer thanks to anyone who has been involved and those people who have
contributed their spare time to continuing maintenance and development.

  • Addition of support for the SBML Spatial package

    • Now that the first official version of the specification for the
      SBML Level 3 Spatial package is available, the libSBML API for
      'spatial' is now included in the stable release of libSBML.
      Thus a CMake build using '-WITH_STABLE_PACKAGES' will now include
      the spatial package code.
  • New features

    • There is a new converter SBMLRateRuleConverter which will take
      a model with RateRules and infer the corresponding reaction network.
      This is still considered under development and we would really
      appreciate any feedback.
  • Bug fixes

    • A large number of code 'smells' have been cleaned up. Thanks to
      Jamie Quinn for undertaking this laborious task.

    • The python examples have been updated for Python 3.8 changes.
      Thanks to Akira Funahashi for committing these fixes.

    • The python code used for creating documentation has been
      updated for Python 3.10 changes. Thanks to
      Zbigniew Jędrzejewski-Szmek for submitting the changes.

  • Configuration/build system changes

    • Extensive continuous integration testing has been added to
      the GitHub repository.
  • Miscellaneous


What's Changed

New Contributors

Full Changelog: v5.19.7...v5.20.0

Release 5.19.7

15 Sep 11:37
6687b73
Compare
Choose a tag to compare
Release 5.19.7 Pre-release
Pre-release

What's Changed

Full Changelog: v5.19.6...v5.19.7

Release 5.19.6

15 Jul 11:55
cb42f00
Compare
Choose a tag to compare
Release 5.19.6 Pre-release
Pre-release

This release solves a number of issues:

What's Changed

New Contributors

Full Changelog: v5.19.5...v5.19.6

libSBML-5.19.5

02 May 09:50
ca5afe1
Compare
Choose a tag to compare
libSBML-5.19.5 Pre-release
Pre-release

This prerelease, updates the build process to use cmake import libraries throughout. It ensures compatibility with latest version of expat, libxml and xerces-c to mitigate denial of service attacks.

Spatial is updated to solve an issue with validation.

libSBML-5.19.2

23 Dec 13:09
78df833
Compare
Choose a tag to compare
libSBML-5.19.2 Pre-release
Pre-release

LibSBML 5.19.2 fixes an issue, where libSBML could not be built using Python 3.10.

libSBML-5.19.0

25 Nov 09:21
118ffbf
Compare
Choose a tag to compare

This is our first release on GitHub.

The usual collection of source and binary files can be found on SourceForge.

Python installation is available via PyPI.

We also now have a package on CondaForge.
Many thanks to Moritz Beber for his persistence and help in making this happen.

    The libSBML source code has moved to GitHub.
        (https://github.com/sbmlteam/libsbml). 

The stable branch reflects the latest released code. The development
branch is the main branch and we encourage users to create and
submit pull requests against this branch. We are no longer
maintaining a separate repository for experimental code for as yet
unfinished packages.

Please use the GitHub issue tracker to file any reports or submit a
fix and do a Pull Request.

We will gradually move away from SourceForge, but for this release
code will be available from both SourceForge and GitHub.

  • Addition of support for the SBML Distributions package

    • Now that the first official version of the specification for the
      SBML Level 3 Distributions package is available, the libSBML API for
      'distrib' is now included in the stable release of libSBML. This
      means that all prebuilt binaries for the stable release will
      include the 'distrib' package code. The src archive containing
      libSBML core code will continue to be available, with additional
      archives of the individual package code.

      NOTE: The libSBML GNU make-based build system has not been updated
      for packages. Thus, to build from src, it is necessary to
      use the CMake build system.

  • New features

    • Functions get/set/isSetMath have been added to the SBase class.
      These do not allow interacting with a non-existant element
      but may facilitate using more generic functions.
  • Bug fixes

    • There was a memory leak introduced in the SWIG wrapper. Thanks to
      Fabian Fröhlich for spotting and fixing this.

    • The SBMLNamespaces class was being leaked in a number of places,
      these have now been plugged.

    • The MathML to infix parser was also leaking memory. Thanks to
      Liam Keegan for spotting and fixing these.

    • The validation code detecting assignment cycles was incorrectly
      including local parameters within its search. These have now
      been removed from the relevant code.

    • Conversion code was failing to detect when a local parameter used
      the 'id' of a speciesReference which is allowed in SBML Level 3.
      These can be converted to SBML Level 2 and the code has now been
      corrected so this conversion will take place.

    • The L3 infix parser has been improved to recognize the use of a
      name without parentheses as a misused function name or a csymbol,
      if appropriate.

    • There was an anomaly when using the addCVTerm() and
      appendAnnotation() functions which meant that if used in a certain
      order information could be lost. This has been corrected. Thanks
      to Colin Halupczok for the detailed report.

    • Converting a to a global object was
      failing if the metaid attribute was set. This has been changed to
      produce the correct behavior.

  • Configuration/build system changes

    • Import of the python library has been sorted in line with recent Linux
      changes to the location of the shared file.

    • Code has been fixed to be MingGW compliant.

    • The build configuration was updated to work with the latest version of R.

  • Miscellaneous

    • A number of memory leaks and possible access violations have been solved.

STABLE PACKAGES

  • 'distrib' package-specific updates:

    • A new converter, AnnotationToDistribConverter, has been added
      to allow conversion from the existing annotation format used
      with to the new 'distrib' format. This
      complements the existing DistribToAnnotationConverter that converts
      'distrib' to annotations.
  • Bug fixes

    • 'comp' package-specific bug fixes:

      • Comp flattening dropped conversion factors for particular
        instances of assignment rules. This has been corrected.
        Thanks to Matthias König for reporting this.
    • 'distrib' package-specific bug fixes:

      • There was unfortunately an infinite loop in the code using
        the element. This has been removed.
    • 'render' package-specific bug fixes:

      • Some render classes were not handled correctly by language
        bindings. This has been fixed.

EXPERIMENTAL PACKAGES

  • New features

    • 'spatial' package-specific updates:

      • Compression in the 'spatial' package is now based on strings.
        The code has been changed to reflect this.

      • Support for int and uint as DataKinds has been added to the
        spatial code in line with changes in the specification.

      • A range of validation tests have been added to the spatial
        code. As this is an experimental passage package still in
        development, we cannot guarantee that all validation has been
        implemented.

  • Bug fixes

    • 'spatial' package-specific bug fixes:

      • There were some memory leaks in the spatial code. Thanks to
        Liam Keegan for spotting these and congratulations
        for being our first GitHub user to address problems using a
        Pull Request.