Storm(py) Information¶
Stormpy links against a particular installation of Storm.
The module stormpy.info provides information about this installation.
This is especially useful for debugging and for writing bug reports.
First, we import the relevant modules:
import stormpy
import stormpy.info
Storm Version¶
The Storm version that stormpy is linked against is available via storm_version():
stormpy.info.storm_version()
'1.14.0'
The flag storm_development_version() tells whether the linked Storm is a development version, i.e. modified since the last release:
stormpy.info.storm_development_version()
True
Further information about the Storm build is available through Version.long and Version.build_info.
These are particularly useful when reporting issues, as they also list the compiler and the build date:
stormpy.info.Version.long
'Version 1.14.0 (dev) build from revision GITDIR-NOTFOUND (potentially dirty)'
stormpy.info.Version.build_info
"Compiled on Linux 6.17.0-1022-azure using gcc (ccache) 15.2.0 with flags 'compiled with NDEBUG'"
Build Configuration¶
The build type of Storm states how the library was originally compiled.
Debug builds are often slower, but easier to debug, while Release builds are the standard choice:
stormpy.info.storm_build_type()
'Release'
Origin of the Storm Installation¶
The function storm_from_system() reports whether stormpy relies on a pre-existing system installation of Storm, or whether Storm was fetched and installed during the build of stormpy:
stormpy.info.storm_from_system()
True
The function storm_origin_info() provides the origin of the Storm installation: the repository URL, the repository tag and the commit hash.
If a system installation was used, the repository and the tag are None.
If Storm was installed as part of the build of stormpy, repository and tag are given; for a local source directory, the tag reads __local-source-dir__:
repo, tag, hash = stormpy.info.storm_origin_info()
print("Repository:", repo)
print("Tag:", tag)
print("Commit hash:", hash)
Repository: None
Tag: None
Commit hash: GITDIR-NOTFOUND
If a system installation of Storm was used, storm_directory() returns the path to this installation. Otherwise, None is returned:
stormpy.info.storm_directory()
'/usr/local/lib/cmake/storm'
Number representations¶
Storm can use different libraries for exact arithmetic.
This is relevant if you compute with exact numbers or rational functions, as the performance may be influenced by the library.
The functions storm_exact_use_cln() and storm_ratfunc_use_cln() report whether CLN (instead of GMP) is used for exact numbers and rational functions, respectively:
print("Represent exact numbers using " + ("CLN" if stormpy.info.storm_exact_use_cln() else "GMP"))
print("Represent rational functions using " + ("CLN" if stormpy.info.storm_ratfunc_use_cln() else "GMP"))
Represent exact numbers using GMP
Represent rational functions using CLN