Activity for Frank Bergmann
-
Frank Bergmann posted a comment on ticket #3
First of all let me recommend that you dont use them (as they have not been updated since 2004, you'd be much better of with newer optimization tools, such as COPASI: http://copasi.org). That being said, If you really wanted to use them, you'd require: - a windows machine (or a virtual machine running windows(preferably 32bit)) - first install the SBW framework (as available here: https://sourceforge.net/projects/sbw/files/sbw/) note i do not know which version will work with the SBW optimizers....
-
- properly update version
-
Installation in MATLAB
-
thanks for letting us know, i've applied the fix to the git repo, so it will be part of the next release.
-
Hello Joel, to clarify, that we know exactly which steps you took: you downloaded this archive: https://sourceforge.net/projects/sbml/files/libsbml/5.19.0/stable/MATLAB%20interface/libSBML-5.19.0-matlab-binaries.zip/download extracted it, and added the folder into your matlab path? then changing into the folder and calling TranslateSBML('test.xml') gave you which error message? Could we see the complete log? the archive contains the binaries for windows (64bit / 32bit), linux (64bit / 32bit) and...
-
Sorry for the inconvenience, we have not yet made a new version of the validator available, so the forwarding problem from before remains. Your registration did go through and i will send you the API key to your registered email address.
-
I can change the 1.2 name to 'Systems Biology Simulation Core Library (SBSCL)' (you or whoever created the entry should also be possible to edit / remove it). If you give me the go ahead i will.
-
Hello Andreas, I'm not sure how likely it is to change the display at this point. The results should have been grouped, if the names of the simulators are the same (in that case only the latest version would appear in the list). Since here the names differ, that would be an issue. you could just let me know which entries you'd like to have removed. Also please drop me a mail, when you want the new results to go public. cheers Frank
-
I've just encountered the same issue. For the time you can find all documents also in the SVN repository here: https://sourceforge.net/p/sbml/code/HEAD/tree/trunk/specifications/sbml-level-3/version-1/ Sorry for the inconvenience. Frank
-
- added missing addxml / remove xml
-
- fix an issue with retrieving elements from https
-
That would explain it. Here is what is happening. The sybilSBML package, does not use the libSBML R bindings at all. Instead it includes a precompiled libSBML (version 5.8.0 on windows) and exposes a subset of the functions there. So there is not much we can help you out there, and you would have to contact them to figure out how to get a newer libSBML version that works with sybil.
-
I think we have to first find out, what is going on here. As the binaries ive created do support the SBML Level 3 FBC and Groups Extensions. Is there a way i can reproduce what happened on your machine locally? if i just recompiled, i don't expect any changes for R 4.0.2 and 4.0.3, that would explain what is going on here.
-
can you please provide an error message, as to what went wrong?
-
Here a version for R 4.0.2 (win x64): https://share.getcloudapp.com/mXu6DkoX
-
Here a version of libSBML for R 4.0.2 (win x64): https://share.getcloudapp.com/mXu6DkoX
-
If you tell me what operating system you use and whether 32 or 64 bit R interpreter, i can create some bindings for you. Sorry, we have not gotten around to making a new release for the newer R versions.
-
Indeed, the binaries we've created are a bit to old for newer versions of R. I've built binaries for R 3.6.3 (win x64) before that you could use for now: https://share.getcloudapp.com/v1u21NNm I can create bindings for 4.x later in the week if needed. The source archive unfortunately works for now only on non-windows systems.
-
I've experimented with a libsbml vcpkg, unfortunately i had problems with the check library. Its been a while though, so something might have improved there, here the experiment: https://github.com/fbergmann/vcpkg In the end i decided to use conan rather than vcpkg, since vcpkg did not work as nicely on other platforms. (and i found vcpkg to work nice on windows with visual studio, but not as nice for other things). Maybe things have improved: https://github.com/fbergmann/conan-libsbml cheers Fr...
-
I've just tried on my machine and it seems to be working for me ... so I'm afraid i will have to ask for more information. If you could try again in Chrome, by navigating to the site, pressing right click and selecting "inspect element" anywhere on the page, and then switch over to the console tab. Are there any error messages? (you might have to refresh the page after selecting the console) thanks Frank
-
I'd be fine to lift the limitations on the format. Many times i would like to save the model history, without having author information for example (timestamps are easy to generate at the timepoint, but figuring out who is using the software is harder). So I would support lifting limitations to sub elements. We should discuss this with the rest of the team though.
-
I agree with this suggestion.
-
Thanks for letting us know, the actual validation service runs on a separate machine, that is experiencing issues at this time. We are looking into it.
-
- support overriding based on objectRole
-
- update the Broker.rc.cmake file with the needed changes for 2019
-
- update resource file for 2019
-
- update rc file
-
- set pointers to 0 after delete
-
- fix compilation with msvc 2019
-
- update install directories
-
- remove visual studio 2008 files
-
- update visual studio files
-
Further things to try: is there a MATLABPATH environment variable defined? since you mention octave, there could be octaverc files lying around with it there are some magic directories like ~/Documents/MATLAB that automatically get added (try calling userpath to figure out whether that could be an issue for you) also try and locate pathdef.m files maybe in /usr/share or therabouts maybe also run matlabroot and look for a file there. sorry for the trouble, i have not encountered any of this before....
-
it might be easier to just call getGeneProductAssociation()->toSBML() that will give you the element as xml string.
-
I've tried for some time to make it work, but i really just don't understand why it is not. So the best i can offer is a workaround, that extends the geneProductAssociation by methods that oroduce the infix, or get the specific element. If this does work for you, i'll be happy to commit that. I'll attach an example, as well as a the updated local-fbc.i, that you will have to place into src/bindings/perl before compilation, so the example will work for you.
-
I'm sorry for the inconvenience, the SBMLToolbox information is stored within the matlab path, which you can for example edit by executing pathtool. If you don't want to use the graphical user interface, you can use rmpathwhich will remove the path given as argument from MATLABs searchpath. Once you've made your changes run savepath for the changes to persist just in case.
-
- while all this compression code is not working as it is, it currently gives falls positives for ASANs, so this commit takes away these issues
-
- need to check first rule .. than initial assignment
-
- add an example of how one would get to the initial values of a species
-
- fix compilation when compiled with namespaces enabled
-
Sorry for the inconvenience, we can certainly improve the documentation: http://sbml.org/Software/libSBML/5.18.0/docs/cpp-api/operation_return_values_8h.html The issue is, that there is no metaid set on the element in question, and so the annotation cannot be attached to it.
-
- fix memory issues
-
- fix when using namespaces
-
Merged revision(s) 26200-26201 from trunk/libsbml:
-
- fix compilation on mingw (all other platforms treat long as 'long long' only mingw decided to use 'long int' and then fail)
-
- issue #169923908: rename duplicated symbol
-
- allow compilation with namespaces
-
- fix include statements
-
- add an example for creating and using a function definition
-
spatial spec example problematic
-
- issue #168098033: swig is generating 'object' rather than '_object', thus search and replace would not attach python3 metaclass
-
- issue #168098033: swig is generating 'object' rather than '_object', thus search and replace would not attach python3 metaclass
-
- add the option of not building static library
-
- remove declarations of finite and isnan that we never declared (and used) as it now causes warnings
-
- use libsbml defined isnan to silence mingw warning
-
- use the correct length
-
- issue #167255390: fix compilation with gcc
-
- issue #167255390: fix compilation with gcc (remove const iterators)
-
- issue #167255390: fix compilation with gcc
-
- issue #167255390: sample field needs to store different datatypes, additionally we allow to store / retrieve them as strings. compression / uncompression is now based on strings rather than on the integer representation of the array.
-
- allow to change level /version of elements from language bindings
-
update version number for release
-
Merged revision(s) 26075-26076 from trunk/libsbml:
-
- fix C# bindings
-
Merged revision(s) 26072-26074 from trunk/libsbml:
-
- update gnumake version number to be consistent
-
Merged revision(s) 26069-26072 from trunk/libsbml:
-
- allow access to SBase::read / SBase::write
-
- catch crashes when no namespaces or document is defined
-
- fix examples to use polygons rather than curves as they can be filled
-
Merged revision(s) 26066-26068 from trunk/libsbml:
-
- hide variables fix build for python 2.5
-
- fix compilation
-
Hello Carlos, sorry for the trouble, it seems indeed that some of the settings on the forwarding machine have changed, and processing of the registration / authentication through the website is currently not working. As workaround (in case you only use the online validator via REST calls) i could offer that you register using the following link: http://constraint.caltech.edu:8888/validator_servlet/Registration.action and while you will not get an automated mail with the API key sent, i will be able...
-
Merged revision(s) 26060-26065 from trunk/libsbml:
-
- fix python import on linux (location of _libsbml.so was no longer working)
-
- issue #166837864: when converting to local parameters delete old parameters so that meta id validator wont' complain (also fix use after delete crash in lv converter, requiring document to become a friend)
-
Merged revision(s) 26044-26047 from trunk/libsbml:
-
- explicitly type all variables so it compiles with ancient compilers
-
- ensure that vs / vscode folders don't get added to source archives
-
Merged revision(s) 26020-26044 from trunk/libsbml:
-
- update version number for gnumake
-
- bump version number to 5.18.1
-
- issue #165670058: render information not correctly downcast, and so wasn't accessible. also corrected issue for java bindings on qual and comp when downcasting lists. added c# render examples for testing.
-
- fix misc prefix on windows
-
- fix windows build
-
- allow to roundtrip a model containing both a bounding box and a curve
-
- only strip packages if packages need to be stripped
-
- fix error where listeners couldnt' be removed
-
- make gcc 8 happy
-
- simplified method
-
- sort compile issue on linux
-
- once more without warnings and c++11
-
- swig definition url registry add return url by index
-
getting make dist-zip to work
-
not a local include
-
update version number
-
- add newer vs files to ignore list
-
- add callback support to bindings and added python example on how to use it
-
- implemented processing callbacks, so users can interrupt long running operations (for now read), adjusted validate example for windows
1 >