SBML Discussion Lists RDF feed

Re: Comp support

myers — 2013-05-15T14:03:23-00:00

That's great news! I look forward to discussing this with you at Harmony. We can wait until the next Jar release likely. Chris On May 15, 2013, at 3:29 AM, Nicolas Rodriguez <<a href="mailto:rodrigue@ebi.ac.uk" target="_blank">rodrigue@ebi.ac.uk</a>> wrote: > Hi Chris, > > > On 14/05/13 16:21, Chris J. Myers wrote: >> Hi, >> >> Just curious when comp will be supported by jSBML. We really want to use jSBML, but we cannot until this is supported. > > This is already there. The support is implemented in the repository but we did not provide any updated jar files > that contain it yet. > > To get it running, you just have to go in the extension/comp folder > and run 'ant jar'. Then if you place the generated jar in your classpath, comp should > be available to read and write. > > I am trying to get few things done before or during Harmony2013 so I can make a new build available then with > comp, fbc, qual, layout, render and spatial included. > > cheers, > Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Comp support

rodrigue — 2013-05-15T09:29:45-00:00

Hi Chris, On 14/05/13 16:21, Chris J. Myers wrote: > Hi, > > Just curious when comp will be supported by jSBML. We really want to use jSBML, but we cannot until this is supported. This is already there. The support is implemented in the repository but we did not provide any updated jar files that contain it yet. To get it running, you just have to go in the extension/comp folder and run 'ant jar'. Then if you place the generated jar in your classpath, comp should be available to read and write. I am trying to get few things done before or during Harmony2013 so I can make a new build available then with comp, fbc, qual, layout, render and spatial included. cheers, Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Comp support

myers — 2013-05-14T15:21:41-00:00

Hi, Just curious when comp will be supported by jSBML. We really want to use jSBML, but we cannot until this is supported. Thanks, Chris ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: JSBML writer question

Sarah Keating — 2013-05-03T11:00:05-00:00

> > Yes, indeed. LibSBML does not yet include validation for packages. Just to clarify that - libSBML includes validation for packages that have been officially approved and released. > > You can validate using the relaxNG schema, although I am not sure it > does reflect the latest changes/validation rules but it > should be pretty accurate. The RNG schema captures syntactic rules; it does not capture more involved rules such as qual-20608 : The QualitativeSpecies referred to by the attribute qual:qualitativeSpecies in an Output object must have the value of its qual:constant attribute set to false. You can use the RNG schemas for SBML L3 packages online at <a href="http://sbml.org/Valtest" target="_blank">http://sbml.org/Valtest</a> Sarah ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: JSBML writer question

rodrigue — 2013-05-02T15:35:33-00:00

On 04/30/2013 04:47 PM, AurÃ©lien Naldi wrote: > Hi Nicolas, > > thanks for the reply! Short version: "you nailed it", see below for the > long version. > >> There is no default values for attributes in SBML L3, so if you don't set >> it yourself, we don't write anything to the file. >> When you create an SBML file with jsbml, you can use >> SBMLDocument.checkConsistency(**) to do the same thing as when you use >> the SBML online validator. >> You can then loop over the SBMLDocument ErrorLog to list all the problems >> found. > > It doesn't seem to handle extensions yet: I only get this error (after > doing the change you suggest): > > This version of the SBML validator supports levels and versions of SBML up > to and including Level 3 Version 1 Core. It does not yet support the > validation of L3 packages. The submitted model appears to contain > information from an L3 package. For the purposes of validation this > information will be ignored and errors and warnings issued for Level 3 > Version 1 Core only. > Yes, indeed. LibSBML does not yet include validation for packages. You can validate using the relaxNG schema, although I am not sure it does reflect the latest changes/validation rules but it should be pretty accurate. I think you know already about that but I put again the instructions in case somebody else is interested : On linux, you can use the following commands: # download jing validator wget <a href="http://jing-trang.googlecode.com/files/jing-20091111.zip" target="_blank">http://jing-trang.googlecode.com/files/jing-20091111.zip</a> unzip jing-20091111.zip # checkout rng files svn co <a href="http://svn.code.sf.net/p/sbml/code/trunk/specifications/RelaxNG" target="_blank">http://svn.code.sf.net/p/sbml/code/trunk/specifications/RelaxNG</a> sbml-RelaxNG you have to apply the following patch to include sbml-qual in the validation process: Index: sbml.rng =================================================================== --- sbml.rng (revision 14634) +++ sbml.rng (working copy) @@ -15,6 +15,6 @@   -  + </grammar> ===== # do validation java -jar jing-20091111/bin/jing.jar -i sbml-RelaxNG/sbml.rng cheers, Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: JSBML writer question

aurelien.naldi — 2013-04-30T15:47:15-00:00

Hi Nicolas, thanks for the reply! Short version: "you nailed it", see below for the long version. > There is no default values for attributes in SBML L3, so if you don't set > it yourself, we don't write anything to the file. > When you create an SBML file with jsbml, you can use > SBMLDocument.checkConsistency(**) to do the same thing as when you use > the SBML online validator. > You can then loop over the SBMLDocument ErrorLog to list all the problems > found. It doesn't seem to handle extensions yet: I only get this error (after doing the change you suggest): This version of the SBML validator supports levels and versions of SBML up to and including Level 3 Version 1 Core. It does not yet support the validation of L3 packages. The submitted model appears to contain information from an L3 package. For the purposes of validation this information will be ignored and errors and warnings issued for Level 3 Version 1 Core only. Note that before doing the change, I did get other errors, but I would not have guessed that they were due to the lack of the "required" attribute! Unfortunately, I could not figure out how to add the required attribute on >> the extensions, am I missing something here? >> The code for the SBML export was based on a demo by Nicolas Rodriguez a >> while ago, and is visible here (extensions are added in SBMLQualHelper.java >> ): <a href="https://github.com/colomoto/**logicalmodel/tree/master/src/**" target="_blank">https://github.com/colomoto/**logicalmodel/tree/master/src/**</a> >> main/java/org/colomoto/**logicalmodel/io/sbml<<a href="https://github.com/colomoto/logicalmodel/tree/master/src/main/java/org/colomoto/logicalmodel/io/sbml" target="_blank">https://github.com/colomoto/logicalmodel/tree/master/src/main/java/org/colomoto/logicalmodel/io/sbml</a>> >> >> > The presentation I gave can be seen there <a href="http://co.mbine.org/colomoto/**" target="_blank">http://co.mbine.org/colomoto/**</a> > meetings/2012?q=system/files/**2012_03_COLOMOTO_JSBML_QUAL.**pdf<<a href="http://co.mbine.org/colomoto/meetings/2012?q=system/files/2012_03_COLOMOTO_JSBML_QUAL.pdf" target="_blank">http://co.mbine.org/colomoto/meetings/2012?q=system/files/2012_03_COLOMOTO_JSBML_QUAL.pdf</a>> > > We still did not put the required attribute automatically, so you are > missing what is on page 9 in the presentation : > > SBMLDocument sbmlDoc = new SBMLDocument(3, 1); > > sbmlDoc.**getSBMLDocumentAttributes().**put(QUAL_NS_PREFIX + > ":required", "true"); // will be done automatically in the future > > > Hope this solve all your problems, > Indeed it does, I guess I was just turned down by the fact that your example was missing the definition of QUAL_NS_PREFIX and seemed to work fine without this line. If I understand well, this QUAL_NS_PREFIX should be QualConstant.shortLabel, right? At least it gives a good-looking output like this. In our export, I will set required to true for the qual extension and to false for the layout extension. Thanks! -- AurÃ©lien Naldi ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: JSBML writer question

rodrigue — 2013-04-30T14:58:24-00:00

On 30/04/13 13:15, AurÃ©lien Naldi wrote: > Hi, > > As you may know, we have some SBML-qual support in GINsim, which > relies on JSBML. > After sending a file to Nicolas Le NovÃ¨re, he pointed out the > following problems: > > * the extension package declarations are missing a "required" attribute > * the compartment is missing a "constant" attribute > > I updated my JSBML snapshot, beside sending the compartment definition > to the end of the file, it didn't change anything. Thus I'm wondering > if I'm using JSBML in a wrong way or if I should report a bug to you ;) > > Going back to the code, I saw a setConstant() method on the > compartment, which fixes the second problem. I'm a bit confused here: > is the attribute left out on purpose as I did not set it up myself or > should it have a default value in JSBML or generate an error upon writing? > There is no default values for attributes in SBML L3, so if you don't set it yourself, we don't write anything to the file. When you create an SBML file with jsbml, you can use SBMLDocument.checkConsistency() to do the same thing as when you use the SBML online validator. You can then loop over the SBMLDocument ErrorLog to list all the problems found. > Unfortunately, I could not figure out how to add the required > attribute on the extensions, am I missing something here? > The code for the SBML export was based on a demo by Nicolas Rodriguez > a while ago, and is visible here (extensions are added in > SBMLQualHelper.java ): > <a href="https://github.com/colomoto/logicalmodel/tree/master/src/main/java/org/colomoto/logicalmodel/io/sbml" target="_blank">https://github.com/colomoto/logicalmodel/tree/master/src/main/java/org/colomoto/logicalmodel/io/sbml</a> > The presentation I gave can be seen there <a href="http://co.mbine.org/colomoto/meetings/2012?q=system/files/2012_03_COLOMOTO_JSBML_QUAL.pdf" target="_blank">http://co.mbine.org/colomoto/meetings/2012?q=system/files/2012_03_COLOMOTO_JSBML_QUAL.pdf</a> We still did not put the required attribute automatically, so you are missing what is on page 9 in the presentation : SBMLDocument sbmlDoc = new SBMLDocument(3, 1); sbmlDoc.getSBMLDocumentAttributes().put(QUAL_NS_PREFIX + ":required", "true"); // will be done automatically in the future Hope this solve all your problems, cheers, Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: JSBML and simple non rdf annotations.

rodrigue — 2013-04-08T15:19:11-00:00

Hi Pablo, I have to say that this part is a bit obscure at the moment and I am not sure that it is documented in the user guide as well. I attached an example code that do what you want. Basically at the moment, the rest of the annotation, meaning what is not described on the SBML specs is stored as a StringBuilder internally and you can access it and modify it through a String. And we call it the non RDF part of the annotation, so you have several methods to access it. We are planning to re-implement all of that for 1.0 to store and XMLNode instead of a String and the best for java developers would be if I manage to integrate it with DOM somehow as well. cheers, Nico On 04/08/2013 03:05 PM, Pablo Moreno wrote: > Hi Nicolas, > > I need to write an annotation block for an SBML model, but cannot > figure how to do it using the JSBML API. Basically I need to write > something like this in the block of the model: > > > > > … > </sidecomounds> > > So I have written something like this in the code: > > Annotation sideCompAnnot = new Annotation(); > Map namespace = new HashMap(); > namespace.put("sidecompounds", > "<a href="http://sbml.org/annotations/sidecompounds" target="_blank">http://sbml.org/annotations/sidecompounds</a>"); > sideCompAnnot.setAnnotationNamespaces(namespace); > > how should I get on with this? Do you add one annotation per each of > the lines of "compound" that I have. If that is the case, how do you > add them then to the model? There is a model.setAnnotation(annot), but > I would think that that would replace what you have added. > > Thanks! > Pablo > > > >

Re: Exception with FunctionDefinition

Alastair Kilpatrick — 2013-03-15T15:40:13-00:00

No problem, thanks for the help! Alastair Quoting Andreas Dräger <<a href="mailto:andreas.draeger@uni-tuebingen.de" target="_blank">andreas.draeger@uni-tuebingen.de</a>> on Fri, 15 Mar 2013 07:41:23 +0100: > Dear Alastair, > > I just see that in the trunk version this problem does not occur because > a save toString conversion has been implemented in > SimpleTreeNodeChangeListener. This solution has now also been > incorporated into the 0.8 branch and will therefore be part of the next > release. Thank you for reporting this issue! > > Cheers > Andreas > > -- > Dr. Andreas Dräger > University of Tuebingen > Center for Bioinformatics Tuebingen (ZBIT) > Sand 1 > 72076 Tübingen > Germany > > Phone: +49-7071-29-78982 > Fax: +49-7071-29-5091 > > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> > > -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Exception with FunctionDefinition

Andreas Dräger — 2013-03-15T06:41:23-00:00

Dear Alastair, I just see that in the trunk version this problem does not occur because a save toString conversion has been implemented in SimpleTreeNodeChangeListener. This solution has now also been incorporated into the 0.8 branch and will therefore be part of the next release. Thank you for reporting this issue! Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Exception with FunctionDefinition

Andreas Dräger — 2013-03-15T06:34:05-00:00

Am 3/14/13 3:58 PM, schrieb Alastair Kilpatrick: > Yep, that's it, thanks a lot! good to know it can be ignored. Not > crucial, but just out of interest, is there a way of stopping this > sort of debug info appearing in the output? > Hi Alastair, The SimpleTreeNodeChangeListener is always added to a model when reading it from a file, so in this version of JSBML you can hardly avoid that this is going to happen. The easiest solution would be not to use the DEBUG mode. However, this can also have some negative consequences. We'll have to provide a better solution in the next version of JSBML. Thank you for this important comment! Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Exception with FunctionDefinition

Alastair Kilpatrick — 2013-03-14T14:58:53-00:00

Yep, that's it, thanks a lot! good to know it can be ignored. Not crucial, but just out of interest, is there a way of stopping this sort of debug info appearing in the output? Alastair Quoting Nicolas Rodriguez <<a href="mailto:rodrigue@ebi.ac.uk" target="_blank">rodrigue@ebi.ac.uk</a>> on Wed, 13 Mar 2013 15:34:22 +0000: > > > Just realise what this error was. > > Did the full output look like that : > > > DEBUG (SimpleTreeNodeChangeListener.java:93) - [CHG] > org.sbml.jsbml.util.TreeNodeChangeEvent[source=] > org.sbml.jsbml.SBMLException: Fractions must have one numerator and > one denominator, here 0 elements are given. > at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1620) > at > org.sbml.jsbml.util.compilers.FormulaCompiler.lambdaBody(FormulaCompiler.java:690) > at > org.sbml.jsbml.util.compilers.FormulaCompiler.lambda(FormulaCompiler.java:672) > at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1839) > at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3892) > at > org.sbml.jsbml.util.SimpleTreeNodeChangeListener.saveToString(SimpleTreeNodeChangeListener.java:110) > at > org.sbml.jsbml.util.SimpleTreeNodeChangeListener.nodeAdded(SimpleTreeNodeChangeListener.java:68) > at > org.sbml.jsbml.AbstractTreeNode.fireNodeAddedEvent(AbstractTreeNode.java:365) > at > org.sbml.jsbml.AbstractTreeNode.firePropertyChange(AbstractTreeNode.java:409) > at org.sbml.jsbml.AbstractTreeNode.setParent(AbstractTreeNode.java:652) > at org.sbml.jsbml.ASTNode.addChild(ASTNode.java:1480) > at > org.sbml.jsbml.xml.parsers.MathMLStaxParser.processStartElement(MathMLStaxParser.java:352) > at > org.sbml.jsbml.xml.stax.SBMLReader.processStartElement(SBMLReader.java:967) > at > org.sbml.jsbml.xml.stax.SBMLReader.readXMLFromXMLEventReader(SBMLReader.java:711) > at > org.sbml.jsbml.xml.stax.SBMLReader.readXMLFromStream(SBMLReader.java:537) > at org.sbml.jsbml.xml.stax.SBMLReader.readSBML(SBMLReader.java:319) > at org.sbml.jsbml.xml.stax.SBMLReader.readSBML(SBMLReader.java:302) > at org.sbml.jsbml.xml.stax.SBMLReader.readSBMLFile(SBMLReader.java:379) > at org.sbml.jsbml.xml.stax.SBMLWriter.main(SBMLWriter.java:145) > WARN (ASTNode.java:3901) - Could not compile ASTNode to formula: > Fractions must have one numerator and one denominator, here 0 > elements are given. > > Meaning that you have the debug output given by the > SimpleTreeNodeChangeListener. In this case, while reading the > mathml, it is trying to print the intermediary math formula parsed. > In this case, until it read the final " 2 </cn>", > it consider that the formula is incomplete. > > These can be ignore, as you noticed, the mathml is read properly at > the end and the ASTNode is fine. > > Cheers, > Nico > > > On 13/03/13 15:22, Nicolas Rodriguez wrote: >> On 13/03/13 11:48, Alastair Kilpatrick wrote: >>> It's version 0.8 - the build.xml file says 1367 2012-08-16 - worth >>> downloading a fresh copy from sourceforge? >>> >> I would have said that we did not change much on this part but I will >> try with 0.8. >> >> Nico >> >>> Quoting Nicolas Rodriguez <<a href="mailto:rodrigue@ebi.ac.uk" target="_blank">rodrigue@ebi.ac.uk</a>> on Wed, 13 Mar 2013 >>> 10:55:40 +0000: >>> >>>> Which version of JSBML are you using ? I don't seems to have any >>>> problems using trunk. >>>> >>>> Cheers, >>>> Nico >>>> >>>> >>>> On 13/03/13 10:38, Alastair Kilpatrick wrote: >>>>> Hi all, hoping you can help - I'm trying to read in an SBML >>>>> FunctionDefinition with JSBML but keep getting an exception with >>>>> 'divide'. My FunctionDefinition (implementing a min(x,y) function) is: >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> x </ci> >>>>> </bvar> >>>>> >>>>> y </ci> >>>>> </bvar> >>>>> >>>>>  >>>>> >>>>> >>>>> >>>>> >>>>> x </ci> >>>>> y </ci> >>>>> </apply> >>>>> >>>>> >>>>> >>>>> >>>>> x </ci> >>>>> y </ci> >>>>> </apply> >>>>> </apply> >>>>> </apply> >>>>> 2 </cn> >>>>> </apply> >>>>> </lambda> >>>>> </math> >>>>> </functionDefinition> >>>>> >>>>> but reading this gives 6 exceptions like: >>>>> org.sbml.jsbml.SBMLException: Fractions must have one numerator and >>>>> one denominator, here 0 elements are given. >>>>> at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1606) >>>>> at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3862)... >>>>> (3 have '0 elements' as above, the other 3 have '1 element'). >>>>> >>>>> Swapping 'divide' out for 'plus', 'minus' doesn't give any errors and >>>>> I *think* the MathML is ok - it all validates and doing >>>>> ASTNode.formulaToString(jsbmlModel.getListOfFunctionDefinitions().get(0).getMath()) >>>>> looks fine: lambda(x, y, >>>>> (x+y-abs(x-y))/2) >>>>> >>>>> Any ideas would be appreciated! Thanks, >>>>> >>>>> Alastair >>>>> >>>> ____________________________________________________________ >>>> To manage your jsbml-development list subscription, visit >>>> <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> >>>> >>>> For a web interface to the jsbml-development mailing list, visit >>>> <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> >>>> >>>> For questions or feedback about the jsbml-development list, >>>> contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> >>>> >>>> >>> >>> >> ____________________________________________________________ >> To manage your jsbml-development list subscription, visit >> <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> >> >> For a web interface to the jsbml-development mailing list, visit >> <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> >> >> For questions or feedback about the jsbml-development list, >> contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> > > > -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Exception with FunctionDefinition

rodrigue — 2013-03-13T15:34:22-00:00

Just realise what this error was. Did the full output look like that : DEBUG (SimpleTreeNodeChangeListener.java:93) - [CHG] org.sbml.jsbml.util.TreeNodeChangeEvent[source=] org.sbml.jsbml.SBMLException: Fractions must have one numerator and one denominator, here 0 elements are given. at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1620) at org.sbml.jsbml.util.compilers.FormulaCompiler.lambdaBody(FormulaCompiler.java:690) at org.sbml.jsbml.util.compilers.FormulaCompiler.lambda(FormulaCompiler.java:672) at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1839) at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3892) at org.sbml.jsbml.util.SimpleTreeNodeChangeListener.saveToString(SimpleTreeNodeChangeListener.java:110) at org.sbml.jsbml.util.SimpleTreeNodeChangeListener.nodeAdded(SimpleTreeNodeChangeListener.java:68) at org.sbml.jsbml.AbstractTreeNode.fireNodeAddedEvent(AbstractTreeNode.java:365) at org.sbml.jsbml.AbstractTreeNode.firePropertyChange(AbstractTreeNode.java:409) at org.sbml.jsbml.AbstractTreeNode.setParent(AbstractTreeNode.java:652) at org.sbml.jsbml.ASTNode.addChild(ASTNode.java:1480) at org.sbml.jsbml.xml.parsers.MathMLStaxParser.processStartElement(MathMLStaxParser.java:352) at org.sbml.jsbml.xml.stax.SBMLReader.processStartElement(SBMLReader.java:967) at org.sbml.jsbml.xml.stax.SBMLReader.readXMLFromXMLEventReader(SBMLReader.java:711) at org.sbml.jsbml.xml.stax.SBMLReader.readXMLFromStream(SBMLReader.java:537) at org.sbml.jsbml.xml.stax.SBMLReader.readSBML(SBMLReader.java:319) at org.sbml.jsbml.xml.stax.SBMLReader.readSBML(SBMLReader.java:302) at org.sbml.jsbml.xml.stax.SBMLReader.readSBMLFile(SBMLReader.java:379) at org.sbml.jsbml.xml.stax.SBMLWriter.main(SBMLWriter.java:145) WARN (ASTNode.java:3901) - Could not compile ASTNode to formula: Fractions must have one numerator and one denominator, here 0 elements are given. Meaning that you have the debug output given by the SimpleTreeNodeChangeListener. In this case, while reading the mathml, it is trying to print the intermediary math formula parsed. In this case, until it read the final " 2 </cn>", it consider that the formula is incomplete. These can be ignore, as you noticed, the mathml is read properly at the end and the ASTNode is fine. Cheers, Nico On 13/03/13 15:22, Nicolas Rodriguez wrote: > On 13/03/13 11:48, Alastair Kilpatrick wrote: >> It's version 0.8 - the build.xml file says 1367 2012-08-16 - worth >> downloading a fresh copy from sourceforge? >> > I would have said that we did not change much on this part but I will > try with 0.8. > > Nico > >> Quoting Nicolas Rodriguez <<a href="mailto:rodrigue@ebi.ac.uk" target="_blank">rodrigue@ebi.ac.uk</a>> on Wed, 13 Mar 2013 >> 10:55:40 +0000: >> >>> Which version of JSBML are you using ? I don't seems to have any >>> problems using trunk. >>> >>> Cheers, >>> Nico >>> >>> >>> On 13/03/13 10:38, Alastair Kilpatrick wrote: >>>> Hi all, hoping you can help - I'm trying to read in an SBML >>>> FunctionDefinition with JSBML but keep getting an exception with >>>> 'divide'. My FunctionDefinition (implementing a min(x,y) function) is: >>>> >>>> >>>> >>>> >>>> >>>> x </ci> >>>> </bvar> >>>> >>>> y </ci> >>>> </bvar> >>>> >>>>  >>>> >>>> >>>> >>>> >>>> x </ci> >>>> y </ci> >>>> </apply> >>>> >>>> >>>> >>>> >>>> x </ci> >>>> y </ci> >>>> </apply> >>>> </apply> >>>> </apply> >>>> 2 </cn> >>>> </apply> >>>> </lambda> >>>> </math> >>>> </functionDefinition> >>>> >>>> but reading this gives 6 exceptions like: >>>> org.sbml.jsbml.SBMLException: Fractions must have one numerator and >>>> one denominator, here 0 elements are given. >>>> at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1606) >>>> at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3862)... >>>> (3 have '0 elements' as above, the other 3 have '1 element'). >>>> >>>> Swapping 'divide' out for 'plus', 'minus' doesn't give any errors and >>>> I *think* the MathML is ok - it all validates and doing >>>> ASTNode.formulaToString(jsbmlModel.getListOfFunctionDefinitions().get(0).getMath()) >>>> looks fine: lambda(x, y, >>>> (x+y-abs(x-y))/2) >>>> >>>> Any ideas would be appreciated! Thanks, >>>> >>>> Alastair >>>> >>> ____________________________________________________________ >>> To manage your jsbml-development list subscription, visit >>> <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> >>> >>> For a web interface to the jsbml-development mailing list, visit >>> <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> >>> >>> For questions or feedback about the jsbml-development list, >>> contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> >>> >>> >> >> > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Exception with FunctionDefinition

rodrigue — 2013-03-13T15:22:19-00:00

On 13/03/13 11:48, Alastair Kilpatrick wrote: > It's version 0.8 - the build.xml file says 1367 2012-08-16 - worth > downloading a fresh copy from sourceforge? > I would have said that we did not change much on this part but I will try with 0.8. Nico > > Quoting Nicolas Rodriguez <<a href="mailto:rodrigue@ebi.ac.uk" target="_blank">rodrigue@ebi.ac.uk</a>> on Wed, 13 Mar 2013 > 10:55:40 +0000: > >> >> Which version of JSBML are you using ? I don't seems to have any >> problems using trunk. >> >> Cheers, >> Nico >> >> >> On 13/03/13 10:38, Alastair Kilpatrick wrote: >>> Hi all, hoping you can help - I'm trying to read in an SBML >>> FunctionDefinition with JSBML but keep getting an exception with >>> 'divide'. My FunctionDefinition (implementing a min(x,y) function) is: >>> >>> >>> >>> >>> >>> x </ci> >>> </bvar> >>> >>> y </ci> >>> </bvar> >>> >>>  >>> >>> >>> >>> >>> x </ci> >>> y </ci> >>> </apply> >>> >>> >>> >>> >>> x </ci> >>> y </ci> >>> </apply> >>> </apply> >>> </apply> >>> 2 </cn> >>> </apply> >>> </lambda> >>> </math> >>> </functionDefinition> >>> >>> but reading this gives 6 exceptions like: >>> org.sbml.jsbml.SBMLException: Fractions must have one numerator and >>> one denominator, here 0 elements are given. >>> at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1606) >>> at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3862)... >>> (3 have '0 elements' as above, the other 3 have '1 element'). >>> >>> Swapping 'divide' out for 'plus', 'minus' doesn't give any errors and >>> I *think* the MathML is ok - it all validates and doing >>> ASTNode.formulaToString(jsbmlModel.getListOfFunctionDefinitions().get(0).getMath()) >>> looks fine: lambda(x, y, >>> (x+y-abs(x-y))/2) >>> >>> Any ideas would be appreciated! Thanks, >>> >>> Alastair >>> >> >> ____________________________________________________________ >> To manage your jsbml-development list subscription, visit >> <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> >> >> For a web interface to the jsbml-development mailing list, visit >> <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> >> >> For questions or feedback about the jsbml-development list, >> contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> >> >> > > > ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Exception with FunctionDefinition

Alastair Kilpatrick — 2013-03-13T11:48:20-00:00

It's version 0.8 - the build.xml file says 1367 2012-08-16 - worth downloading a fresh copy from sourceforge? Thanks, Alastair Quoting Nicolas Rodriguez <<a href="mailto:rodrigue@ebi.ac.uk" target="_blank">rodrigue@ebi.ac.uk</a>> on Wed, 13 Mar 2013 10:55:40 +0000: > > Which version of JSBML are you using ? I don't seems to have any > problems using trunk. > > Cheers, > Nico > > > On 13/03/13 10:38, Alastair Kilpatrick wrote: >> Hi all, hoping you can help - I'm trying to read in an SBML >> FunctionDefinition with JSBML but keep getting an exception with >> 'divide'. My FunctionDefinition (implementing a min(x,y) function) is: >> >> >> >> >> >> x </ci> >> </bvar> >> >> y </ci> >> </bvar> >> >>  >> >> >> >> >> x </ci> >> y </ci> >> </apply> >> >> >> >> >> x </ci> >> y </ci> >> </apply> >> </apply> >> </apply> >> 2 </cn> >> </apply> >> </lambda> >> </math> >> </functionDefinition> >> >> but reading this gives 6 exceptions like: >> org.sbml.jsbml.SBMLException: Fractions must have one numerator and >> one denominator, here 0 elements are given. >> at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1606) >> at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3862)... >> (3 have '0 elements' as above, the other 3 have '1 element'). >> >> Swapping 'divide' out for 'plus', 'minus' doesn't give any errors and >> I *think* the MathML is ok - it all validates and doing >> ASTNode.formulaToString(jsbmlModel.getListOfFunctionDefinitions().get(0).getMath()) looks fine: lambda(x, >> y, >> (x+y-abs(x-y))/2) >> >> Any ideas would be appreciated! Thanks, >> >> Alastair >> > > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> > > -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Exception with FunctionDefinition

rodrigue — 2013-03-13T10:55:40-00:00

Which version of JSBML are you using ? I don't seems to have any problems using trunk. Cheers, Nico On 13/03/13 10:38, Alastair Kilpatrick wrote: > Hi all, hoping you can help - I'm trying to read in an SBML > FunctionDefinition with JSBML but keep getting an exception with > 'divide'. My FunctionDefinition (implementing a min(x,y) function) is: > > > > > > x </ci> > </bvar> > > y </ci> > </bvar> > >  > > > > > x </ci> > y </ci> > </apply> > > > > > x </ci> > y </ci> > </apply> > </apply> > </apply> > 2 </cn> > </apply> > </lambda> > </math> > </functionDefinition> > > but reading this gives 6 exceptions like: > org.sbml.jsbml.SBMLException: Fractions must have one numerator and > one denominator, here 0 elements are given. > at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1606) > at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3862)... > (3 have '0 elements' as above, the other 3 have '1 element'). > > Swapping 'divide' out for 'plus', 'minus' doesn't give any errors and > I *think* the MathML is ok - it all validates and doing > ASTNode.formulaToString(jsbmlModel.getListOfFunctionDefinitions().get(0).getMath()) looks fine: lambda(x, y, > (x+y-abs(x-y))/2) > > Any ideas would be appreciated! Thanks, > > Alastair > ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Exception with FunctionDefinition

Alastair Kilpatrick — 2013-03-13T10:38:06-00:00

Hi all, hoping you can help - I'm trying to read in an SBML FunctionDefinition with JSBML but keep getting an exception with 'divide'. My FunctionDefinition (implementing a min(x,y) function) is: x </ci> </bvar> y </ci> </bvar>  x </ci> y </ci> </apply> x </ci> y </ci> </apply> </apply> </apply> 2 </cn> </apply> </lambda> </math> </functionDefinition> but reading this gives 6 exceptions like: org.sbml.jsbml.SBMLException: Fractions must have one numerator and one denominator, here 0 elements are given. at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1606) at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3862)... (3 have '0 elements' as above, the other 3 have '1 element'). Swapping 'divide' out for 'plus', 'minus' doesn't give any errors and I *think* the MathML is ok - it all validates and doing ASTNode.formulaToString(jsbmlModel.getListOfFunctionDefinitions().get(0).getMath()) looks fine: lambda(x, y, (x+y-abs(x-y))/2) Any ideas would be appreciated! Thanks, Alastair -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Layout extension

Andreas Dräger — 2013-03-01T14:31:58-00:00

Am 3/1/13 3:11 PM, schrieb Nicolas Rodriguez: > Le 01/03/2013 12:46, Andreas Dräger a écrit : >> As it the case with the actual SBML packages, also from the perspective >> of its implementation in JSBML the Layout package was our first package. >> At this time we named the "plugin" class for layout extended models >> "ExtendedLayoutModel", but now our naming convention follows libSBML in >> all other package implementation, i.e., it should better be named >> LayoutModelPlugin. I therefore request to rename this class. What do >> people think? > I am fine with the renaming but I would let the old ExtendedLayoutModel, > extending > > LayoutModelPlugin with just the constructor so that it does not break any current code > as people are using the trunk. > Hi, Shouldn't we mark the old class as deprecated then? Maybe there are also some other packages where we didn't follow the naming conventions. I think we should go through all types of SBasePlugin and check the names of the classes accordingly, maybe making some more classes deprecated? Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Layout extension

rodrigue — 2013-03-01T14:11:38-00:00

Le 01/03/2013 12:46, Andreas Dräger a écrit : > Dear all, > > As it the case with the actual SBML packages, also from the perspective > of its implementation in JSBML the Layout package was our first package. > At this time we named the "plugin" class for layout extended models > "ExtendedLayoutModel", but now our naming convention follows libSBML in > all other package implementation, i.e., it should better be named > LayoutModelPlugin. I therefore request to rename this class. What do > people think? I am fine with the renaming but I would let the old ExtendedLayoutModel, extending LayoutModelPlugin with just the constructor so that it does not break any current code as people are using the trunk. cheers, Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Layout extension

Andreas Dräger — 2013-03-01T12:46:30-00:00

Dear all, As it the case with the actual SBML packages, also from the perspective of its implementation in JSBML the Layout package was our first package. At this time we named the "plugin" class for layout extended models "ExtendedLayoutModel", but now our naming convention follows libSBML in all other package implementation, i.e., it should better be named LayoutModelPlugin. I therefore request to rename this class. What do people think? Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: How to deal with invalid elements that are added to ListOf* objects?

Andreas Dräger — 2013-02-12T06:58:00-00:00

Am 2/7/13 3:10 PM, schrieb Andreas Dräger: > 1) In the add method of ListOf insert the following code in the catch > clause: > > if (exc instanceof RuntimeException) { > throw (RuntimeException) exc; > } Dear all, in agreement with Nicolas Rodriguez, I am going to commit this solution because there was no complain about it. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

How to deal with invalid elements that are added to ListOf* objects?

Andreas Dräger — 2013-02-07T14:10:52-00:00

Dear all, Nico and I just detected a problem in the current trunk in JSBML, which leads to a failure of some test cases of the project.The error goes back to a very old change from September last year: <a href="http://jsbml.svn.sourceforge.net/viewvc/jsbml/trunk/core/src/org/sbml/jsbml/ListOf.java?r1=1416&r2=1415&pathrev=1416." target="_blank">http://jsbml.svn.sourceforge.net/viewvc/jsbml/trunk/core/src/org/sbml/jsbml/ListOf.java?r1=1416&r2=1415&pathrev=1416.</a> To better understand the following explanation, please note that in JSBML the ListOf<?> class is actually a wrapper around some List implementation from the standard Java library. The problem before this change was that registerChild was called in ListOf before adding the element to the actual internal list, i.e., listeners were notified about a newly added element, but this new element could not be found anywhere in the model because it was not (really) in the list! So, what had to be done was to first add the element to the internal list (within the ListOf), irrespective of any restrictions, then calling registerChild, and reverting the change in the model if some error occurs. In case of an error, false would be returned. Only a LevelVersionError was thrown, all other exceptions were caught and passed to the logger as a debug message. The method then returned false. The test case, however, expects that an IllegalArgumentException is thrown, which would come from the registerChild method (that was called before adding the element to the internal list, i.e., interrupting the add method in ListOf). The test cases don't accept the return value of "false", they expect the method being interrupted by the IllegalArgumentException. I have now to possible solutions what we can do: 1) In the add method of ListOf insert the following code in the catch clause: if (exc instanceof RuntimeException) { throw (RuntimeException) exc; } I tested this solution, all test cases would work with it. However, this would cause that the add method won't return false in case of a problem, it would simply be interrupted. Therefore, I have a second solution: 2) Change all create* methods in all classes to something like public Species createSpecies(String id) { Species species = new Species(id, getLevel(), getVersion()); if (addSpecies(species)) { return species; } return null; } Then the test cases would also have to be changed and would have to check if the create method returns null, i.e., failure (expected), or not (pass). Solution 2 causes more effort and will affect more portions of the code. It can also lead to NullPointerExceptions elsewhere in the code if the return value of some create* method is not checked immeadiately by users. What do people think? Which solution should be applied? Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: jsbml 0.8-b1 writing related questions

rodrigue — 2012-12-05T16:34:39-00:00

On 01/12/12 21:28, smaor wrote: > Hi guys, > > I'll admit I'm a bit new to jsbml. So any help would be extremely appreciated as I've been trying to search through google to find some ideas for way too long now. > > I'm currently working on a project that's implementing the jsbml 0.8-b1 library and one of my tasks is to write out a sbml after manipulation within the application. I would strongly encourage you to not use 0.8-b1 but the final 0.8 release (located here <a href="https://sourceforge.net/projects/jsbml/files/jsbml/0.8/" target="_blank">https://sourceforge.net/projects/jsbml/files/jsbml/0.8/</a>) There is a bit more than one year of bug fixes and improvement between them. The full version in sourceforge will contain the manual and API as well. > > I've been wondering if anyone still has a copy of the manual or tutorial for jsbml-0.8-b1? Any literature on it would be appreciated. > > Also, just a couple questions: > > The list of parameters, is there a way to change the node attributes from name to id s.t: > > becomes > > > > Regarding namespaces: > how would I set up the following tag? > a workspace such as... > > xmlns:html="<a href="http://www.w3.org/1999/xhtml" target="_blank">http://www.w3.org/1999/xhtml</a>"> > > for > I am not sure if you find out how to do that but sbmlDocument.addNamespace("html", "xmlns", "<a href="http://www.w3.org/1999/xhtml" target="_blank">http://www.w3.org/1999/xhtml</a>"); should work. > > Regarding Notes, what's the best way to implement them? > I would need to create something of this format: > GENE_ASSOCIATION: ______</html:p> > PROTEIN_ASSOCIATION: ______</html:p> > SUBSYSTEM: ______</html:p> > PROTEIN_CLASS: ______</html:p> You can just create your html String like that and use SBase.setNotes(String) Let us know if we solved all your problems, Thanks, Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: jsbml 0.8-b1 writing related questions

Andreas Dräger — 2012-12-02T06:23:52-00:00

Am 12/1/12 10:28 PM, schrieb smaor: > The list of parameters, is there a way to change the node attributes from name to id s.t: > > becomes > Hi, try this: myParameter.setId(myParameter.getName()); myParameter.unsetName(); Here, myParameter is a Java object representing the instance of Parameter that you like to manipulate. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: jsbml 0.8-b1 writing related questions

Mike Hucka — 2012-12-02T01:25:26-00:00

On Sat, 1 Dec 2012 13:28:56 -0800 (PST), smaor wrote: > I've been wondering if anyone still has a copy of the manual or > tutorial for jsbml-0.8-b1? Any literature on it would be appreciated. Hi, Have you seen the "Documentation" link at <a href="http://sbml.org/Software/JSBML" target="_blank">http://sbml.org/Software/JSBML</a> ? Best regards, MH -- Mike Hucka, Ph.D. -- <a href="mailto:mhucka@caltech.edu" target="_blank">mhucka@caltech.edu</a> -- <a href="http://www.cds.caltech.edu/~mhucka" target="_blank">http://www.cds.caltech.edu/~mhucka</a> Computing and Mathematical Sciences, California Institute of Technology, Pasadena, California, USA -- Twitter: @mhucka -- Skype: michaelhucka -- You can give me anonymous feedback via <a href="http://tinyurl.com/mhuckafeedback" target="_blank">http://tinyurl.com/mhuckafeedback</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

jsbml 0.8-b1 writing related questions

smaor — 2012-11-30T09:29:15-00:00

Hi guys, I'll admit I'm a bit new to jsbml. So any help would be extremely appreciated as I've been trying to search through google to find some ideas for way too long now. I'm currently working on a project that's implementing the jsbml 0.8-b1 library and one of my tasks is to write out a sbml after manipulation within the application. I've been wondering if anyone still has a copy of the manual or tutorial for jsbml-0.8-b1? Any literature on it would be appreciated. Also, just a couple questions: The list of parameters, is there a way to change the node attributes from name to id s.t: becomes Regarding namespaces: how would I set up the following tag? a workspace such as... xmlns:html="<a href="http://www.w3.org/1999/xhtml" target="_blank">http://www.w3.org/1999/xhtml</a>"> for Regarding Notes, what's the best way to implement them? I would need to create something of this format: GENE_ASSOCIATION: ______</html:p> PROTEIN_ASSOCIATION: ______</html:p> SUBSYSTEM: ______</html:p> PROTEIN_CLASS: ______</html:p> Thanks in advance.

Re: [jsbml-team] Extensions of extensions

Andreas Dräger — 2012-08-30T14:41:29-00:00

Am 5/10/12 5:00 PM, schrieb Sarah Keating: >> Again, I want to stress to check what decisions Franck and the libsbml >> team have taken already to see if we store it as a child of the model >> element or some layout elements. >> > The libsbml render extension code has two plugins > > RenderLayoutPlugin > > RenderListOfLayoutsPlugin > > because render extends the listOfLayouts element and the individual > layout elements NOT the model. > > Our naming convention is > > ExtensionObjectExtendedObjectPlugin > > so for example layout has only one plugin class > > LayoutModelPlugin > > as it extends the Model component of core. > > We envision that other packages will want to extend packages not just > core and so we went with the design decision that Plugins can have child > Plugins. > > Sarah > Dear all, According to this convention, I think the ExtendedLayoutModel class in JSBML in the layout package should better be renamed to LayoutModelPlugin. Otherwise it will not be consistent with other packages in SBML and probably also not with libSBML. Fortunately, the SBML extension packages are not part of the stable release 0.8 of JSBML and can still be renamed. What do you think? Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Release of JSBML-0.8

rodrigue — 2012-08-24T22:56:00-00:00

------------------------- JSBML-0.8 ------------------------- We are pleased to announce the first stable release of JSBML-0.8, which is available for downloading from <a href="https://sourceforge.net/projects/jsbml/files/jsbml/0.8" target="_blank">https://sourceforge.net/projects/jsbml/files/jsbml/0.8</a> JSBML is a community-driven project to create a free, open-source, pure Java library for reading, writing, and manipulating SBML files and data streams. It is an alternative to the mixed Java/native code-based interface provided in libSBML. For more details, please visit <a href="http://sbml.org/Software/JSBML" target="_blank">http://sbml.org/Software/JSBML</a> ====================================================================== Version 0.8 (24-08-2012) ====================================================================== * New Features: - The new specialized EventObject TreeNodeRemovedEvent has been implemented in order to make sure that deleted nodes do no longer have a pointer to their previous parent within the tree. At the same time, it is still possible to identify their previous location using the field in the new EventObject. For people using already the change listeners it means one API change that could break existing code. * Bug Fixes: - The ASTNode methods insertChild, replaceChild, prependChild have been corrected to setup correctly the parentSBMLObject in the added ASTNode tree. Thanks to Jason Stevens who reported the problem. - When creating an ASTNode with a CallableSBase or using the setVariable method, the reference to the id of the CallableSBase was not stored. - There is now a rounding correction for multipliers when merging units that tries to shift multipliers to the scale of a unit. This can only be done if the logarithm to the base 10 of the multiplier is nearly an integer number. Nearly means that the distance to the next integer is not larger than the ensured double precision. - [Tracker item #3511439] : The cloning of a Species was not copying the speciesType attribute. Thanks to Thomas Bernard who noticed and reported the problem. - The units attribute on ASTNode was sometimes not written out and the SBML namespace was sometimes missing from the MathML. It should be fine now. - Improved the registration/un-registration of elements in particular for local parameters. - Modified the parser to be able to read invalid notes, anything inside the notes element will be read even if not in the html namespace. - Corrected several other clone methods. The Reaction clone constructor was not cloning the compartment attribute. The Model clone was not cloning the new L3 attributes. The Constraint.clone() was not properly cloning the Message. When cloning ASTNodes, the direct pointer to referenced variables (instances of CallableSBase) are set to null. - An other small API change: the species.setBoundaryCondition method was expecting a Boolean instead of a boolean, this has been modified. It should not break any existing code as the JVM will autocast the Boolean. - Added the biological qualifier hasTaxon. - Submodel generation corrections related to event and namespace * Known issues : - Reading of more than one RDF Description elements in the RDF annotation block (the official SBML annotation scheme) is not supported at the moment. Planned for the next major release. - The RDF block at the top level of the annotation is considered to be the official SBML annotation. - SBML L3 packages are not supported and not kept. Thanks The JSBML team. ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: creating an astnode with a function call

rodrigue — 2012-08-19T00:54:46-00:00

Le 19/08/2012 01:49, Jason Stevens a écrit : > Okay, thanks for the help. I'll look into the warnings a little more. > I also forgot to mention that I had tried to use replace child > originally, but that gave me an exception (I can get the details if > you'd like), so I removed and inserted instead. > Yes, the replaceChild had the same bug and it should be fixed as well. Nico > On Saturday, August 18, 2012, Nicolas Rodriguez wrote: > > Hi Jason, > > Thanks for the detailed email, I think I have fixed your problems. > I was about to do the 0.8 release but now we have to postpone to > next week at least to check that everything > is fine. > > The problem was linked to ASTNode.insertChild that was not setting > properly its parents. > > Le 18/08/2012 01:48, Jason Stevens a écrit : >> Hi again, >> >> I recently revisited this and I'm still having the same problems. >> The relevant code is spread over different parts of my program, >> but I've attached the bit where I think things are going wrong >> (though I'm really not sure where things are going wrong). What >> I'm trying to do is take an existing ASTNode (Node A) and replace >> it with another ASTNode (Node B). Node A has a function like >> this: neighborQuantity(speciesID), and I want to essentially add >> information by replacing Node A with another ASTNode of the form >> neighborQuantityFull(speciesID, >> compartmentLocationX(compartmentID), >> compartmentLocationY(compartmentID)). So what I attached is how >> I'm doing that. >> >> It seems to work (toFormula prints the correct thing). However, >> whenever I do a toFormula I get these warnings: >> >> 12/08/17 17:32:56 WARN jsbml.ASTNode: ASTNode of type FUNCTION >> but the variable is null: (neighborQuantityRightFull, null)! >> Check that your object is linked to a Model. >> 12/08/17 17:32:56 WARN jsbml.ASTNode: ASTNode of type FUNCTION >> but the variable is null: (getCompartmentLocationX, null)! Check >> that your object is linked to a Model. >> 12/08/17 17:32:56 WARN jsbml.ASTNode: ASTNode of type FUNCTION >> but the variable is null: (getCompartmentLocationY, null)! Check >> that your object is linked to a Model. > > These warnings should appear only if the ASTNode has not been > added to the model (so set as math to a MathContainer) or if the > FunctionDefinition have not been added to the model yet. If you > think you are getting these warnings when this two conditions are > fine, can you provide us with a small, complete example that > reproduce the problem ? (a cloned element or a removed is not > linked to the model anymore) > >> >> The IDs that I'm passing into parseFormula are valid, and it >> seems like parseFormula should be putting in the relevant >> variables, but that doesn't appear to be happening. >> >> Later on, I clone the list of events in the model (these formulas >> I'm creating are triggers for events) which leads to the >> exception below (the bottom part of which repeats). This seems >> to have something to do with the triggers I've created. >> >> Exception in thread "Thread-3" java.lang.StackOverflowError > ....... >> at org.sbml.jsbml.util.StringTools.concat(StringTools.java:105) >> at >> org.sbml.jsbml.util.compilers.FormulaCompiler.relation(FormulaCompiler.java:1038) >> at org.sbml.jsbml.util.compilers.FormulaCompiler.gt >> <<a href="http://org.sbml.jsbml.util.compilers.FormulaCompiler.gt" target="_blank">http://org.sbml.jsbml.util.compilers.FormulaCompiler.gt</a>>(FormulaCompiler.java:805) >> at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1823) >> at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3710) >> at >> org.sbml.jsbml.AbstractMathContainer.toString(AbstractMathContainer.java:326) >> > > So, this should be fixed in the latest trunk and I joined the test > class that I used to try to reproduce it. > > Cheers, > Nico > ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: creating an astnode with a function call

rodrigue — 2012-08-19T00:30:34-00:00

Hi Jason, Thanks for the detailed email, I think I have fixed your problems. I was about to do the 0.8 release but now we have to postpone to next week at least to check that everything is fine. The problem was linked to ASTNode.insertChild that was not setting properly its parents. Le 18/08/2012 01:48, Jason Stevens a écrit : > Hi again, > > I recently revisited this and I'm still having the same problems. The > relevant code is spread over different parts of my program, but I've > attached the bit where I think things are going wrong (though I'm > really not sure where things are going wrong). What I'm trying to do > is take an existing ASTNode (Node A) and replace it with another > ASTNode (Node B). Node A has a function like this: > neighborQuantity(speciesID), and I want to essentially add information > by replacing Node A with another ASTNode of the form > neighborQuantityFull(speciesID, compartmentLocationX(compartmentID), > compartmentLocationY(compartmentID)). So what I attached is how I'm > doing that. > > It seems to work (toFormula prints the correct thing). However, > whenever I do a toFormula I get these warnings: > > 12/08/17 17:32:56 WARN jsbml.ASTNode: ASTNode of type FUNCTION but the > variable is null: (neighborQuantityRightFull, null)! Check that your > object is linked to a Model. > 12/08/17 17:32:56 WARN jsbml.ASTNode: ASTNode of type FUNCTION but the > variable is null: (getCompartmentLocationX, null)! Check that your > object is linked to a Model. > 12/08/17 17:32:56 WARN jsbml.ASTNode: ASTNode of type FUNCTION but the > variable is null: (getCompartmentLocationY, null)! Check that your > object is linked to a Model. These warnings should appear only if the ASTNode has not been added to the model (so set as math to a MathContainer) or if the FunctionDefinition have not been added to the model yet. If you think you are getting these warnings when this two conditions are fine, can you provide us with a small, complete example that reproduce the problem ? (a cloned element or a removed is not linked to the model anymore) > > The IDs that I'm passing into parseFormula are valid, and it seems > like parseFormula should be putting in the relevant variables, but > that doesn't appear to be happening. > > Later on, I clone the list of events in the model (these formulas I'm > creating are triggers for events) which leads to the exception below > (the bottom part of which repeats). This seems to have something to > do with the triggers I've created. > > Exception in thread "Thread-3" java.lang.StackOverflowError ....... > at org.sbml.jsbml.util.StringTools.concat(StringTools.java:105) > at > org.sbml.jsbml.util.compilers.FormulaCompiler.relation(FormulaCompiler.java:1038) > at org.sbml.jsbml.util.compilers.FormulaCompiler.gt > <<a href="http://org.sbml.jsbml.util.compilers.FormulaCompiler.gt" target="_blank">http://org.sbml.jsbml.util.compilers.FormulaCompiler.gt</a>>(FormulaCompiler.java:805) > at org.sbml.jsbml.ASTNode.compile(ASTNode.java:1823) > at org.sbml.jsbml.ASTNode.toString(ASTNode.java:3710) > at > org.sbml.jsbml.AbstractMathContainer.toString(AbstractMathContainer.java:326) > So, this should be fixed in the latest trunk and I joined the test class that I used to try to reproduce it. Cheers, Nico

Re: Deprecated SBML-Qual attributes and elements

rodrigue — 2012-07-26T15:40:18-00:00

On 25/07/12 11:08, Martijn van Iersel wrote: > Hi all, > > I noticed that my patch hasn't been applied yet. Is there anything I > can do to help this move forward? Let me know if something is blocking > this. > The patch is applied now, everything seems to be fine. Nico > > On 03/07/12 12:48, Martijn van Iersel wrote: >> I have attached a patch that removes (instead of deprecates) all the >> relevant bits, including lisTofSymbolicValues and boundaryCondition. >> I've checked that it compiles and produces valid SBML-Qual output. >> >> I'll send notifications to the sbml-qual mailing list about the removal >> of these methods. >> >> -- >> Martijn >> >> On 29/06/12 10:50, Nicolas Rodriguez wrote: >>> On 26/06/12 15:33, Martijn van Iersel wrote: >>>> As you know, during the Colomoto and HARMONY meetings, we discussed >>>> removing some of the unused elements of SBML-Qual, in the hope that >>>> this will make the spec simpler and more mature. >>>> >>>> I've attached a patch that marks the concerned methods as deprecated >>>> in JSBML. This is a conservative approach: most of these are not >>>> actually used by any software at all. So it may be better to remove >>>> these functions instead of deprecating them. If you wish, I could make >>>> a patch for this as well. >>>> >>>> I think the only element that is somewhat used in software is >>>> boundaryCondition on QualitativeSpecies. So it may be a good idea to >>>> deprecate that one and remove all the rest. >>>> >>> >>> Yes, I would remove them. You should send a mail to the colomoto and/or >>> sbml-qual mailing list to see if anybody is using boundaryCondition and >>> remove it as well if nobody reply to complain in the next week. >>> >>>> One final note: the ListOfSymbolicValues on QualitativeSpecies may >>>> still come back in some form or other. However, it's certain that it's >>>> not going to stay in its current form, so it makes sense to remove it >>>> or deprecate it to prevent people from using it accidentally. >>>> >>> >>> Yes, we should remove it as well >>> >>> Let me know if you try to provide a patch for the removal of the unused >>> attributes and elements ? >>> >>> Thanks, >>> Nico >>> >>> >> >> > ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Deprecated SBML-Qual attributes and elements

rodrigue — 2012-07-25T10:18:45-00:00

Le 25/07/2012 11:08, Martijn van Iersel a écrit : > Hi all, > > I noticed that my patch hasn't been applied yet. Is there anything I > can do to help this move forward? Let me know if something is blocking > this. > Nope, I was just away or not much available since few weeks, I will try to apply the patch this week. Nico > > On 03/07/12 12:48, Martijn van Iersel wrote: >> I have attached a patch that removes (instead of deprecates) all the >> relevant bits, including lisTofSymbolicValues and boundaryCondition. >> I've checked that it compiles and produces valid SBML-Qual output. >> >> I'll send notifications to the sbml-qual mailing list about the removal >> of these methods. >> >> -- >> Martijn >> >> On 29/06/12 10:50, Nicolas Rodriguez wrote: >>> On 26/06/12 15:33, Martijn van Iersel wrote: >>>> As you know, during the Colomoto and HARMONY meetings, we discussed >>>> removing some of the unused elements of SBML-Qual, in the hope that >>>> this will make the spec simpler and more mature. >>>> >>>> I've attached a patch that marks the concerned methods as deprecated >>>> in JSBML. This is a conservative approach: most of these are not >>>> actually used by any software at all. So it may be better to remove >>>> these functions instead of deprecating them. If you wish, I could make >>>> a patch for this as well. >>>> >>>> I think the only element that is somewhat used in software is >>>> boundaryCondition on QualitativeSpecies. So it may be a good idea to >>>> deprecate that one and remove all the rest. >>>> >>> >>> Yes, I would remove them. You should send a mail to the colomoto and/or >>> sbml-qual mailing list to see if anybody is using boundaryCondition and >>> remove it as well if nobody reply to complain in the next week. >>> >>>> One final note: the ListOfSymbolicValues on QualitativeSpecies may >>>> still come back in some form or other. However, it's certain that it's >>>> not going to stay in its current form, so it makes sense to remove it >>>> or deprecate it to prevent people from using it accidentally. >>>> >>> >>> Yes, we should remove it as well >>> >>> Let me know if you try to provide a patch for the removal of the unused >>> attributes and elements ? >>> >>> Thanks, >>> Nico >>> >>> >> >> > ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Deprecated SBML-Qual attributes and elements

Martijn van Iersel — 2012-07-25T10:08:13-00:00

Hi all, I noticed that my patch hasn't been applied yet. Is there anything I can do to help this move forward? Let me know if something is blocking this. -- Martijn On 03/07/12 12:48, Martijn van Iersel wrote: > I have attached a patch that removes (instead of deprecates) all the > relevant bits, including lisTofSymbolicValues and boundaryCondition. > I've checked that it compiles and produces valid SBML-Qual output. > > I'll send notifications to the sbml-qual mailing list about the removal > of these methods. > > -- > Martijn > > On 29/06/12 10:50, Nicolas Rodriguez wrote: >> On 26/06/12 15:33, Martijn van Iersel wrote: >>> As you know, during the Colomoto and HARMONY meetings, we discussed >>> removing some of the unused elements of SBML-Qual, in the hope that >>> this will make the spec simpler and more mature. >>> >>> I've attached a patch that marks the concerned methods as deprecated >>> in JSBML. This is a conservative approach: most of these are not >>> actually used by any software at all. So it may be better to remove >>> these functions instead of deprecating them. If you wish, I could make >>> a patch for this as well. >>> >>> I think the only element that is somewhat used in software is >>> boundaryCondition on QualitativeSpecies. So it may be a good idea to >>> deprecate that one and remove all the rest. >>> >> >> Yes, I would remove them. You should send a mail to the colomoto and/or >> sbml-qual mailing list to see if anybody is using boundaryCondition and >> remove it as well if nobody reply to complain in the next week. >> >>> One final note: the ListOfSymbolicValues on QualitativeSpecies may >>> still come back in some form or other. However, it's certain that it's >>> not going to stay in its current form, so it makes sense to remove it >>> or deprecate it to prevent people from using it accidentally. >>> >> >> Yes, we should remove it as well >> >> Let me know if you try to provide a patch for the removal of the unused >> attributes and elements ? >> >> Thanks, >> Nico >> >> > > ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Deprecated SBML-Qual attributes and elements

Martijn van Iersel — 2012-07-03T11:48:52-00:00

I have attached a patch that removes (instead of deprecates) all the relevant bits, including lisTofSymbolicValues and boundaryCondition. I've checked that it compiles and produces valid SBML-Qual output. I'll send notifications to the sbml-qual mailing list about the removal of these methods. -- Martijn On 29/06/12 10:50, Nicolas Rodriguez wrote: > On 26/06/12 15:33, Martijn van Iersel wrote: >> As you know, during the Colomoto and HARMONY meetings, we discussed >> removing some of the unused elements of SBML-Qual, in the hope that >> this will make the spec simpler and more mature. >> >> I've attached a patch that marks the concerned methods as deprecated >> in JSBML. This is a conservative approach: most of these are not >> actually used by any software at all. So it may be better to remove >> these functions instead of deprecating them. If you wish, I could make >> a patch for this as well. >> >> I think the only element that is somewhat used in software is >> boundaryCondition on QualitativeSpecies. So it may be a good idea to >> deprecate that one and remove all the rest. >> > > Yes, I would remove them. You should send a mail to the colomoto and/or > sbml-qual mailing list to see if anybody is using boundaryCondition and > remove it as well if nobody reply to complain in the next week. > >> One final note: the ListOfSymbolicValues on QualitativeSpecies may >> still come back in some form or other. However, it's certain that it's >> not going to stay in its current form, so it makes sense to remove it >> or deprecate it to prevent people from using it accidentally. >> > > Yes, we should remove it as well > > Let me know if you try to provide a patch for the removal of the unused > attributes and elements ? > > Thanks, > Nico > >

Re: Deprecated SBML-Qual attributes and elements

rodrigue — 2012-06-29T09:50:09-00:00

On 26/06/12 15:33, Martijn van Iersel wrote: > As you know, during the Colomoto and HARMONY meetings, we discussed > removing some of the unused elements of SBML-Qual, in the hope that > this will make the spec simpler and more mature. > > I've attached a patch that marks the concerned methods as deprecated > in JSBML. This is a conservative approach: most of these are not > actually used by any software at all. So it may be better to remove > these functions instead of deprecating them. If you wish, I could make > a patch for this as well. > > I think the only element that is somewhat used in software is > boundaryCondition on QualitativeSpecies. So it may be a good idea to > deprecate that one and remove all the rest. > Yes, I would remove them. You should send a mail to the colomoto and/or sbml-qual mailing list to see if anybody is using boundaryCondition and remove it as well if nobody reply to complain in the next week. > One final note: the ListOfSymbolicValues on QualitativeSpecies may > still come back in some form or other. However, it's certain that it's > not going to stay in its current form, so it makes sense to remove it > or deprecate it to prevent people from using it accidentally. > Yes, we should remove it as well Let me know if you try to provide a patch for the removal of the unused attributes and elements ? Thanks, Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Deprecated SBML-Qual attributes and elements

Martijn van Iersel — 2012-06-26T14:33:35-00:00

As you know, during the Colomoto and HARMONY meetings, we discussed removing some of the unused elements of SBML-Qual, in the hope that this will make the spec simpler and more mature. I've attached a patch that marks the concerned methods as deprecated in JSBML. This is a conservative approach: most of these are not actually used by any software at all. So it may be better to remove these functions instead of deprecating them. If you wish, I could make a patch for this as well. I think the only element that is somewhat used in software is boundaryCondition on QualitativeSpecies. So it may be a good idea to deprecate that one and remove all the rest. One final note: the ListOfSymbolicValues on QualitativeSpecies may still come back in some form or other. However, it's certain that it's not going to stay in its current form, so it makes sense to remove it or deprecate it to prevent people from using it accidentally. -- Martijn

Re: creating an astnode with a function call

rodrigue — 2012-06-25T12:14:58-00:00

Hi Jason, On 24/06/12 02:28, Jason Stevens wrote: > I'm trying to create an ASTNode that includes a function call. I have > a function definition in the model for a function called location, and > I'd like to be able to create an ASTNode that, in string form, would > be "location(compartmentID)". If I use > parseFormula("location(compartmentID)"), it looks fine, but then JSBML > gives me warnings about there not being a variable defined, which > causes lots of problems downstream. Can you be more specific about the warnings you are getting in the first place and what kind of problems you encounter afterwards ? Also, which version of jsbml are you using ? I just tried to create a small model, using the parseFormula method and the mathML generated looks fine and I see no warning. location </ci> cell </ci> </apply> </math> I attached the java source file I used to test. Cheers, Nico

creating an astnode with a function call

Jason Stevens — 2012-06-24T01:28:35-00:00

I'm trying to create an ASTNode that includes a function call. I have a function definition in the model for a function called location, and I'd like to be able to create an ASTNode that, in string form, would be "location(compartmentID)". If I use parseFormula("location(compartmentID)"), it looks fine, but then JSBML gives me warnings about there not being a variable defined, which causes lots of problems downstream. Thanks, Jason ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: [sbml-qual] SBML-Qual: defaultFunctionTerm

Martijn van Iersel — 2012-06-11T16:26:31-00:00

> Hi Marijn and all others, > > What I actually not understand in this definition is the following > question - I hope you can explain why it is as it is in the qual > specification: > > The DefaultTerm is stored in the ListOfFunctionTerms. The Transition has > at most one such list. Why is the DefaultTerm not directly stored within > the Transition? I think the designers thought it was logical this way. It certainly makes sense to group the defaultTerm and the functionalTerms together. But in the end, it doesn't matter at all, as long as there is a place to store the defaultTerm and we all agree to where that is :) -- Martijn ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: [sbml-qual] SBML-Qual: defaultFunctionTerm

Nicolas Le Novere — 2012-06-11T15:29:48-00:00

On 10/06/12 17:21, Andreas Dräger wrote: > By the way, just now I see that all the classes in the UML diagrams of > the qual specification are spelled in upper case, whereas most SBML > elements (i.e., the things that are written in XML) are spelled in lower > case, e.g., or. Should maybe the UML > diagrams also be labeled with lower case names? Hi Andreas, The classes are capitalised in the SBML specs as well. XML elements are instances of the classes. Hence the lowercase start. Best regards -- Nicolas LE NOVERE, Computational Systems Neurobiology, EMBL-EBI, WTGC, Hinxton CB101SD UK, Mob:+447833147074, Tel:+441223494521 Fax:468, <a href="mailto:lenov@ebi.ac.uk" target="_blank">lenov@ebi.ac.uk</a>, Skype:n.lenovere, twitter:@lenovere <a href="http://www.ebi.ac.uk/~lenov/," target="_blank">http://www.ebi.ac.uk/~lenov/,</a> <a href="http://www.ebi.ac.uk/compneur/" target="_blank">http://www.ebi.ac.uk/compneur/</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: SBML-Qual: defaultFunctionTerm

rodrigue — 2012-06-11T08:52:54-00:00

On 08/06/12 19:02, mvaniersel wrote: > SBML-Qual has a method for defining the default result of a transition called defaultTerm. See for example this snippet: > > > > >

<  !âˆ’âˆ’ A>= 1âˆ’âˆ’><br />
> <br />
> <br />
> A</ c i><br />
> theta B A</ c i><br />
>      </apply><br />
>     </math><br />
>    </functionTerm><br />
> <br />
>   </listOfFunctionTerms><br />
><br />
><br />
> See also the current specs: <a href="https://sbml.svn.sourceforge.net/svnroot/sbml/trunk/specifications/sbml-level-3/version-1/qual/spec/qual-spec.pdf" target="_blank">https://sbml.svn.sourceforge.net/svnroot/sbml/trunk/specifications/sbml-level-3/version-1/qual/spec/qual-spec.pdf</a><br />
><br />
> defaultTerm is not currently implemented in JSBML. The problem is that this is a sub-element of a ListOf, which we can not add additional members to. Or can we?<br />
<br />
The defaultTerm is implemented, you need to create a FunctionTerm and <br />
call setDefaultTerm(true).<br />
<br />
Cheers,<br />
Nico<br />
<br />
____________________________________________________________<br />
To manage your jsbml-development list subscription, visit <br />
<a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a><br />
<br />
For a web interface to the jsbml-development mailing list, visit<br />
<a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a><br />
<br />
For questions or feedback about the jsbml-development list,<br />
contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a><br />
<br />

Re: SBML-Qual: defaultFunctionTerm

Andreas Dräger — 2012-06-10T16:21:34-00:00

Am 6/8/12 8:02 PM, schrieb mvaniersel: > SBML-Qual has a method for defining the default result of a transition called defaultTerm. See for example this snippet: > > > > >

<  !âˆ’âˆ’ A>= 1âˆ’âˆ’><br />
> <br />
> <br />
> A</ c i><br />
> theta B A</ c i><br />
>      </apply><br />
>     </math><br />
>    </functionTerm><br />
> <br />
>   </listOfFunctionTerms><br />
><br />
><br />
> See also the current specs: <a href="https://sbml.svn.sourceforge.net/svnroot/sbml/trunk/specifications/sbml-level-3/version-1/qual/spec/qual-spec.pdf" target="_blank">https://sbml.svn.sourceforge.net/svnroot/sbml/trunk/specifications/sbml-level-3/version-1/qual/spec/qual-spec.pdf</a><br />
><br />
> defaultTerm is not currently implemented in JSBML. The problem is that this is a sub-element of a ListOf, which we can not add additional members to. Or can we?<br />
><br />
><br />
<br />
<br />
Hi Marijn and all others,<br />
<br />
What I actually not understand in this definition is the following <br />
question - I hope you can explain why it is as it is in the qual <br />
specification:<br />
<br />
The DefaultTerm is stored in the ListOfFunctionTerms. The Transition has <br />
at most one such list. Why is the DefaultTerm not directly stored within <br />
the Transition? The list is directly associated with precisely one <br />
Transition. Hence, no other Transition has the identical list. Hence, <br />
the DefaultTerm is also uniquely defined for one particular Transition. <br />
 From my perspective it is not clear why the actual list contains this <br />
element besides its list members (i.e., instances of FunctionTerm). <br />
Wouldn't it make sense to define that the Transition has a DefaultTerm <br />
and optionally a ListOfFunctionTerms?<br />
<br />
By the way, just now I see that all the classes in the UML diagrams of <br />
the qual specification are spelled in upper case, whereas most SBML <br />
elements (i.e., the things that are written in XML) are spelled in lower <br />
case, e.g., or . Should maybe the UML <br />
diagrams also be labeled with lower case names?<br />
<br />
Best<br />
Andreas<br />
<br />
-- <br />
Dr. Andreas Dräger<br />
University of Tuebingen<br />
Center for Bioinformatics Tuebingen (ZBIT)<br />
Sand 1<br />
72076 Tübingen<br />
Germany<br />
<br />
Phone: +49-7071-29-78982<br />
Fax:   +49-7071-29-5091<br />
<br />
____________________________________________________________<br />
To manage your jsbml-development list subscription, visit <br />
<a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a><br />
<br />
For a web interface to the jsbml-development mailing list, visit<br />
<a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a><br />
<br />
For questions or feedback about the jsbml-development list,<br />
contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a><br />
<br />

SBML-Qual: defaultFunctionTerm

mvaniersel — 2012-06-08T15:05:15-00:00

SBML-Qual has a method for defining the default result of a transition called defaultTerm. See for example this snippet: <pre>

< !âˆ’âˆ’ A >= 1âˆ’âˆ’> A</ c i> theta B A</ c i>
    </apply>
   </math>
  </functionTerm> </listOfFunctionTerms>
</pre><br />
<br />
See also the current specs: <a href="https://sbml.svn.sourceforge.net/svnroot/sbml/trunk/specifications/sbml-level-3/version-1/qual/spec/qual-spec.pdf" target="_blank">https://sbml.svn.sourceforge.net/svnroot/sbml/trunk/specifications/sbml-level-3/version-1/qual/spec/qual-spec.pdf</a><br />
<br />
defaultTerm is not currently implemented in JSBML. The problem is that this is a sub-element of a ListOf, which we can not add additional members to. Or can we?<br />

Re: [sbml-render] Ids and names on elements

Andreas Dräger — 2012-05-15T06:29:31-00:00

Am 5/11/12 8:27 PM, schrieb Michael Hucka: > On Thu, 10 May 2012 07:31:26 +0200, Andreas Dräger wrote: >> My question is therefore, first, why it was actually decided not to have >> a name attribute in addition to the id for several elements in >> SBML-render, and second, if you have some proposal what we can do for >> our implementation? > I don't remember the precise reasons, but layout& render are the two oldest packages, predating even the final L3 specification. It probably has more than one oddity as a result of the fact that many principles were never fully worked out at the time they were first developed. > > These specifications are not 100% final, and small details like this could probably be adjusted. > > Hi all, In this case, I propose to let all classes from SBML-Render that have just an ID but no name in the specification extend AbstractNamedSBase, which gives the possibility to also set a name. The important point of this will be that then the model will be able to register links to these elements and take care to avoid duplicate identifiers. Otherwise, we would have to split the interface NamdeSBase (and AbstractNamedSBase as well) from JSBML's core into two distinct classes, one caring about the ID and a derived class that also has a name tag. This would make JSBML again a bit more complicated and would also cause additional effort. The easier way seems to be to introduce an optional name attribute on some Render elements. By making this optional, all existing render files would still be valid, no real change. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: [jsbml-team] Extensions of extensions

Sarah Keating — 2012-05-10T15:00:36-00:00

> Again, I want to stress to check what decisions Franck and the libsbml > team have taken already to see if we store it as a child of the model > element or some layout elements. > The libsbml render extension code has two plugins RenderLayoutPlugin RenderListOfLayoutsPlugin because render extends the listOfLayouts element and the individual layout elements NOT the model. Our naming convention is ExtensionObjectExtendedObjectPlugin so for example layout has only one plugin class LayoutModelPlugin as it extends the Model component of core. We envision that other packages will want to extend packages not just core and so we went with the design decision that Plugins can have child Plugins. Sarah ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: [jsbml-team] Extensions of extensions

rodrigue — 2012-05-10T14:10:22-00:00

On 10/05/12 06:20, Andreas Dräger wrote: > Dear all, > > In contrast to all other SBML extensions implemented so far, the > render extension does not extend the SBML core, but the layout > extension. This is why, I would like to start the discussion how we > can deal with this in the Java implementation. > > If we would strictly follow the specification of the render extension, > it would have to be embedded within the annotation of the > listOfLayouts inside of the layout extension. Currently, the > SBasePlugin cannot be extended with additional plugins. And the > annotation is very difficult to use based on XMLnode datastructures. You need to have a look at how it has been implemented in libSBML as the specification update are lacking. Interact with the render mailing list (<a href="https://lists.sourceforge.net/lists/listinfo/sbml-render," target="_blank">https://lists.sourceforge.net/lists/listinfo/sbml-render,</a> just saw that you posted a question there already) as well so that the decisions are at least posted and discussed there. Please don't speak about annotations, the whole point of doing an L3 package is to take it out of the annotations. > > I would therefore like to propose to implement the render extension > also as a regular extension of the model, i.e., there would be a class > such as ExtendedRenderModel that would be similar to the existing > ExtendedLayoutModel. Virtually, all data structures beloning to the > render extension would be linked as children to this element. Again, I want to stress to check what decisions Franck and the libsbml team have taken already to see if we store it as a child of the model element or some layout elements. > > On the other hand, however, it would also make sense to implement this > "plugin" as a plugin for the ExtendedLayoutModel, because the render > extension makes only sense in combination with the layout extension. > > Should we introduce a mechanism to extend (Abstract)SBasePlugin with > further plugins? > Yes, we just have to add the HashMap with the associated methods to SBasePlugin. Thanks, Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

Andreas Dräger — 2012-05-10T12:50:17-00:00

Am 5/10/12 2:38 PM, schrieb Nicolas Rodriguez: > >>>> Hi all, >>>> >>>> here is a minimal example that doesn't work properly with the current >>>> version of JSBML: >>>> >>>> public static void main(String[] args) throws Throwable { >>>> SBMLDocument doc = new SBMLDocument(3, 1); >>>> Model m = doc.createModel("m1"); >>>> Compartment c = m.createCompartment("c1"); >>>> Species s1 = m.createSpecies("s1", c); >>>> Species s2 = m.createSpecies("s2", c); >>>> Reaction r = m.createReaction("r1"); >>>> r.createReactant(s1).setStoichiometry(1d); >>>> r.createProduct(s2).setStoichiometry(1d); >>>> KineticLaw kl = r.createKineticLaw(); >>>> ASTNode one = new ASTNode(1, kl); >>>> ASTNode two = new ASTNode(2, kl); >>>> one.setUnits(Unit.Kind.MOLE.toString().toLowerCase()); >>>> two.setUnits(Unit.Kind.SECOND.toString().toLowerCase()); >>>> ASTNode ast = ASTNode.frac(one, two); >>>> kl.setMath(ast); >>>> SBMLWriter.write(doc, System.out, ' ', (short) 2); >>>> } >>>> >>>> You'll see that the first unit is not written to SBML at all. The second >>>> unit is written, but the namespace declaration is missing. I already >>>> found what is probably the reason, but can't fix it at the moment: The >>>> recursive check if the ASTNode contains unit declarations doesn't work. >>> Not sure of what is going on there !! >>> >>> The unit of the first astnode is reset somewhere. >>> >>> if I print the value of the units in >>> MathMLXMLStreamCompiler.compileInteger(ASTNode), the first astnode >>> return 'null'. >> The unit that get deleted is probably due to the code from >> org.sbml.jsbml.ASTNode.arithmeticOperation(Type, ASTNode) ?? >> >> The swap node does not copy the units ? > ok, both fixes are commited now and it should be working fine. > > I suppose this a good case to de-freeze the 0.8 branch. > > Should I commit there (in the 0.8 branch) the changes to > Model.registerIds as well ? > Hi Nico, Thank you a lot for your great job! Yes, I agree, please do a commit here! This is an absolutely necessary change. Best Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

rodrigue — 2012-05-10T12:38:39-00:00

On 10/05/12 12:57, Nicolas Rodriguez wrote: > On 10/05/12 12:49, Nicolas Rodriguez wrote: >> On 09/05/12 19:42, Andreas Dräger wrote: >>> Am 5/8/12 3:43 PM, schrieb Nicolas Rodriguez: >>>> On 08/05/12 14:30, Nicolas Rodriguez wrote: >>>>> On 08/05/12 14:20, Andreas Dräger wrote: >>>>>> Am 5/8/12 3:17 PM, schrieb Nicolas Rodriguez: >>>>>>> On 08/05/12 13:57, Andreas Dräger wrote: >>>>>>>> |Hi all, >>>>>>>> >>>>>>>> It seems the SBML writer does currently write invalid SBML code when >>>>>>>> making use of the ability of SBML Level 3 to associate plain numbers >>>>>>>> with units: >>>>>>>> >>>>>>>> 1</cn> >>>>>>>> >>>>>>>> The above code is only valid if the namespace declaration of the math >>>>>>>> element also defines >>>>>>>> >>>>>>>> xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" >>>>>>>> >>>>>>>> Currently, it does only declare the regular MathML namesapce. As a >>>>>>>> result, JSBML is currently unable to parse models that itself wrote into >>>>>>>> a file before. Where can we fix this problem? >>>>>>> The check is in >>>>>>> org.sbml.jsbml.util.compilers.MathMLXMLStreamCompiler.isSBMLNamespaceNeeded(ASTNode) >>>>>>> >>>>>>> the namespace is added in >>>>>>> org.sbml.jsbml.xml.stax.SBMLWriter.writeMathML(MathContainer, >>>>>>> SMOutputElement, XMLStreamWriter, int) >>>>>>> >>>>>>> line 1084 to 1106. >>>>>>> >>>>>>> The compiler org.sbml.jsbml.util.compilers.FindUnitsCompiler is used to >>>>>>> say if there is a unit declaration or not in the mathML. >>>>>>> >>>>>>> This is probably there that the problem happen, and the value returned >>>>>>> is probably false. >>>>>>> It is probably due to the way you are declaring the units and then how >>>>>>> the mathML compiler is calling the methods. >>>>>>> >>>>>>> If I read a model with units, then the writing is fine so there is may >>>>>>> be something missing when we declare units from scratch. >>>>>>> >>>>>>> >>>>>> Hi Nico, >>>>>> >>>>>> Yes, it is probably because I simply say >>>>>> >>>>>> myASTNode.setUnit(Unit.Kind.DIMENSIONLESS.getName()); >>>>>> >>>>>> or something similar. I thought this would be enough and didn't care >>>>>> about namespaces at the moment. >>>>> It should be enough but it was probably not tested much until now. >>>>> >>>>> We should probably check how the reader is setting things up. >>>> So, the reader is not doing anything specific, just using the setUnits() >>>> method ( I suppose it was a typo error on your code as I don't see a >>>> method setUnit(). >>>> >>>> Strange, it was working fine before. >>>> >>>> >>> Hi all, >>> >>> here is a minimal example that doesn't work properly with the current >>> version of JSBML: >>> >>> public static void main(String[] args) throws Throwable { >>> SBMLDocument doc = new SBMLDocument(3, 1); >>> Model m = doc.createModel("m1"); >>> Compartment c = m.createCompartment("c1"); >>> Species s1 = m.createSpecies("s1", c); >>> Species s2 = m.createSpecies("s2", c); >>> Reaction r = m.createReaction("r1"); >>> r.createReactant(s1).setStoichiometry(1d); >>> r.createProduct(s2).setStoichiometry(1d); >>> KineticLaw kl = r.createKineticLaw(); >>> ASTNode one = new ASTNode(1, kl); >>> ASTNode two = new ASTNode(2, kl); >>> one.setUnits(Unit.Kind.MOLE.toString().toLowerCase()); >>> two.setUnits(Unit.Kind.SECOND.toString().toLowerCase()); >>> ASTNode ast = ASTNode.frac(one, two); >>> kl.setMath(ast); >>> SBMLWriter.write(doc, System.out, ' ', (short) 2); >>> } >>> >>> You'll see that the first unit is not written to SBML at all. The second >>> unit is written, but the namespace declaration is missing. I already >>> found what is probably the reason, but can't fix it at the moment: The >>> recursive check if the ASTNode contains unit declarations doesn't work. >> Not sure of what is going on there !! >> >> The unit of the first astnode is reset somewhere. >> >> if I print the value of the units in >> MathMLXMLStreamCompiler.compileInteger(ASTNode), the first astnode >> return 'null'. > The unit that get deleted is probably due to the code from > org.sbml.jsbml.ASTNode.arithmeticOperation(Type, ASTNode) ?? > > The swap node does not copy the units ? ok, both fixes are commited now and it should be working fine. I suppose this a good case to de-freeze the 0.8 branch. Should I commit there (in the 0.8 branch) the changes to Model.registerIds as well ? Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

Andreas Dräger — 2012-05-10T11:59:08-00:00

Hi Nicolas, >> Hi all, >> >> here is a minimal example that doesn't work properly with the current >> version of JSBML: >> >> public static void main(String[] args) throws Throwable { >> SBMLDocument doc = new SBMLDocument(3, 1); >> Model m = doc.createModel("m1"); >> Compartment c = m.createCompartment("c1"); >> Species s1 = m.createSpecies("s1", c); >> Species s2 = m.createSpecies("s2", c); >> Reaction r = m.createReaction("r1"); >> r.createReactant(s1).setStoichiometry(1d); >> r.createProduct(s2).setStoichiometry(1d); >> KineticLaw kl = r.createKineticLaw(); >> ASTNode one = new ASTNode(1, kl); >> ASTNode two = new ASTNode(2, kl); >> one.setUnits(Unit.Kind.MOLE.toString().toLowerCase()); >> two.setUnits(Unit.Kind.SECOND.toString().toLowerCase()); >> ASTNode ast = ASTNode.frac(one, two); >> kl.setMath(ast); >> SBMLWriter.write(doc, System.out, ' ', (short) 2); >> } >> >> You'll see that the first unit is not written to SBML at all. The second >> unit is written, but the namespace declaration is missing. I already >> found what is probably the reason, but can't fix it at the moment: The >> recursive check if the ASTNode contains unit declarations doesn't work. > Not sure of what is going on there !! > > The unit of the first astnode is reset somewhere. > > if I print the value of the units in > MathMLXMLStreamCompiler.compileInteger(ASTNode), the first astnode > return 'null'. > > The sbml namespace is not added because the method > FindUnitsCompiler.compile(int, String) is never called by the > ASTNodeCompiler. > Again, not sure why it would work sometimes and not others. > > Nico > > Please have a look just at the implementation of the frac method within the class org.sbml.jsbml.util.compilers.FindUnitsCompiler: public ASTNodeValue frac(ASTNode numerator, ASTNode denominator) throws SBMLException { return dummyValue; } As you can see, the method does not perform any recursion. It simply returns the dummyValue. If the numerator contains a number with a unit, the FindUnitsCompiler won't be able to find it because the recursion will already stop here. The same holds true for the denominator. No chance to identify any number deeper inside of the fraction. Similarly, all other methods in this class that are supposed to continue the recursion by at least calling something like numerator.compile(this) and denominator.compile(this) also do not perform any recursion. There are many, many methods in this class that will have to be changed. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

rodrigue — 2012-05-10T11:57:31-00:00

On 10/05/12 12:49, Nicolas Rodriguez wrote: > On 09/05/12 19:42, Andreas Dräger wrote: >> Am 5/8/12 3:43 PM, schrieb Nicolas Rodriguez: >>> On 08/05/12 14:30, Nicolas Rodriguez wrote: >>>> On 08/05/12 14:20, Andreas Dräger wrote: >>>>> Am 5/8/12 3:17 PM, schrieb Nicolas Rodriguez: >>>>>> On 08/05/12 13:57, Andreas Dräger wrote: >>>>>>> |Hi all, >>>>>>> >>>>>>> It seems the SBML writer does currently write invalid SBML code when >>>>>>> making use of the ability of SBML Level 3 to associate plain numbers >>>>>>> with units: >>>>>>> >>>>>>> 1</cn> >>>>>>> >>>>>>> The above code is only valid if the namespace declaration of the math >>>>>>> element also defines >>>>>>> >>>>>>> xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" >>>>>>> >>>>>>> Currently, it does only declare the regular MathML namesapce. As a >>>>>>> result, JSBML is currently unable to parse models that itself wrote into >>>>>>> a file before. Where can we fix this problem? >>>>>> The check is in >>>>>> org.sbml.jsbml.util.compilers.MathMLXMLStreamCompiler.isSBMLNamespaceNeeded(ASTNode) >>>>>> >>>>>> the namespace is added in >>>>>> org.sbml.jsbml.xml.stax.SBMLWriter.writeMathML(MathContainer, >>>>>> SMOutputElement, XMLStreamWriter, int) >>>>>> >>>>>> line 1084 to 1106. >>>>>> >>>>>> The compiler org.sbml.jsbml.util.compilers.FindUnitsCompiler is used to >>>>>> say if there is a unit declaration or not in the mathML. >>>>>> >>>>>> This is probably there that the problem happen, and the value returned >>>>>> is probably false. >>>>>> It is probably due to the way you are declaring the units and then how >>>>>> the mathML compiler is calling the methods. >>>>>> >>>>>> If I read a model with units, then the writing is fine so there is may >>>>>> be something missing when we declare units from scratch. >>>>>> >>>>>> >>>>> Hi Nico, >>>>> >>>>> Yes, it is probably because I simply say >>>>> >>>>> myASTNode.setUnit(Unit.Kind.DIMENSIONLESS.getName()); >>>>> >>>>> or something similar. I thought this would be enough and didn't care >>>>> about namespaces at the moment. >>>> It should be enough but it was probably not tested much until now. >>>> >>>> We should probably check how the reader is setting things up. >>> So, the reader is not doing anything specific, just using the setUnits() >>> method ( I suppose it was a typo error on your code as I don't see a >>> method setUnit(). >>> >>> Strange, it was working fine before. >>> >>> >> Hi all, >> >> here is a minimal example that doesn't work properly with the current >> version of JSBML: >> >> public static void main(String[] args) throws Throwable { >> SBMLDocument doc = new SBMLDocument(3, 1); >> Model m = doc.createModel("m1"); >> Compartment c = m.createCompartment("c1"); >> Species s1 = m.createSpecies("s1", c); >> Species s2 = m.createSpecies("s2", c); >> Reaction r = m.createReaction("r1"); >> r.createReactant(s1).setStoichiometry(1d); >> r.createProduct(s2).setStoichiometry(1d); >> KineticLaw kl = r.createKineticLaw(); >> ASTNode one = new ASTNode(1, kl); >> ASTNode two = new ASTNode(2, kl); >> one.setUnits(Unit.Kind.MOLE.toString().toLowerCase()); >> two.setUnits(Unit.Kind.SECOND.toString().toLowerCase()); >> ASTNode ast = ASTNode.frac(one, two); >> kl.setMath(ast); >> SBMLWriter.write(doc, System.out, ' ', (short) 2); >> } >> >> You'll see that the first unit is not written to SBML at all. The second >> unit is written, but the namespace declaration is missing. I already >> found what is probably the reason, but can't fix it at the moment: The >> recursive check if the ASTNode contains unit declarations doesn't work. > Not sure of what is going on there !! > > The unit of the first astnode is reset somewhere. > > if I print the value of the units in > MathMLXMLStreamCompiler.compileInteger(ASTNode), the first astnode > return 'null'. The unit that get deleted is probably due to the code from org.sbml.jsbml.ASTNode.arithmeticOperation(Type, ASTNode) ?? The swap node does not copy the units ? Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

rodrigue — 2012-05-10T11:49:02-00:00

On 09/05/12 19:42, Andreas Dräger wrote: > Am 5/8/12 3:43 PM, schrieb Nicolas Rodriguez: >> On 08/05/12 14:30, Nicolas Rodriguez wrote: >>> On 08/05/12 14:20, Andreas Dräger wrote: >>>> Am 5/8/12 3:17 PM, schrieb Nicolas Rodriguez: >>>>> On 08/05/12 13:57, Andreas Dräger wrote: >>>>>> |Hi all, >>>>>> >>>>>> It seems the SBML writer does currently write invalid SBML code when >>>>>> making use of the ability of SBML Level 3 to associate plain numbers >>>>>> with units: >>>>>> >>>>>> 1</cn> >>>>>> >>>>>> The above code is only valid if the namespace declaration of the math >>>>>> element also defines >>>>>> >>>>>> xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" >>>>>> >>>>>> Currently, it does only declare the regular MathML namesapce. As a >>>>>> result, JSBML is currently unable to parse models that itself wrote into >>>>>> a file before. Where can we fix this problem? >>>>> The check is in >>>>> org.sbml.jsbml.util.compilers.MathMLXMLStreamCompiler.isSBMLNamespaceNeeded(ASTNode) >>>>> >>>>> the namespace is added in >>>>> org.sbml.jsbml.xml.stax.SBMLWriter.writeMathML(MathContainer, >>>>> SMOutputElement, XMLStreamWriter, int) >>>>> >>>>> line 1084 to 1106. >>>>> >>>>> The compiler org.sbml.jsbml.util.compilers.FindUnitsCompiler is used to >>>>> say if there is a unit declaration or not in the mathML. >>>>> >>>>> This is probably there that the problem happen, and the value returned >>>>> is probably false. >>>>> It is probably due to the way you are declaring the units and then how >>>>> the mathML compiler is calling the methods. >>>>> >>>>> If I read a model with units, then the writing is fine so there is may >>>>> be something missing when we declare units from scratch. >>>>> >>>>> >>>> Hi Nico, >>>> >>>> Yes, it is probably because I simply say >>>> >>>> myASTNode.setUnit(Unit.Kind.DIMENSIONLESS.getName()); >>>> >>>> or something similar. I thought this would be enough and didn't care >>>> about namespaces at the moment. >>> It should be enough but it was probably not tested much until now. >>> >>> We should probably check how the reader is setting things up. >> So, the reader is not doing anything specific, just using the setUnits() >> method ( I suppose it was a typo error on your code as I don't see a >> method setUnit(). >> >> Strange, it was working fine before. >> >> > Hi all, > > here is a minimal example that doesn't work properly with the current > version of JSBML: > > public static void main(String[] args) throws Throwable { > SBMLDocument doc = new SBMLDocument(3, 1); > Model m = doc.createModel("m1"); > Compartment c = m.createCompartment("c1"); > Species s1 = m.createSpecies("s1", c); > Species s2 = m.createSpecies("s2", c); > Reaction r = m.createReaction("r1"); > r.createReactant(s1).setStoichiometry(1d); > r.createProduct(s2).setStoichiometry(1d); > KineticLaw kl = r.createKineticLaw(); > ASTNode one = new ASTNode(1, kl); > ASTNode two = new ASTNode(2, kl); > one.setUnits(Unit.Kind.MOLE.toString().toLowerCase()); > two.setUnits(Unit.Kind.SECOND.toString().toLowerCase()); > ASTNode ast = ASTNode.frac(one, two); > kl.setMath(ast); > SBMLWriter.write(doc, System.out, ' ', (short) 2); > } > > You'll see that the first unit is not written to SBML at all. The second > unit is written, but the namespace declaration is missing. I already > found what is probably the reason, but can't fix it at the moment: The > recursive check if the ASTNode contains unit declarations doesn't work. Not sure of what is going on there !! The unit of the first astnode is reset somewhere. if I print the value of the units in MathMLXMLStreamCompiler.compileInteger(ASTNode), the first astnode return 'null'. The sbml namespace is not added because the method FindUnitsCompiler.compile(int, String) is never called by the ASTNodeCompiler. Again, not sure why it would work sometimes and not others. Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Extensions of extensions

Andreas Dräger — 2012-05-10T05:20:25-00:00

Dear all, In contrast to all other SBML extensions implemented so far, the render extension does not extend the SBML core, but the layout extension. This is why, I would like to start the discussion how we can deal with this in the Java implementation. If we would strictly follow the specification of the render extension, it would have to be embedded within the annotation of the listOfLayouts inside of the layout extension. Currently, the SBasePlugin cannot be extended with additional plugins. And the annotation is very difficult to use based on XMLnode datastructures. I would therefore like to propose to implement the render extension also as a regular extension of the model, i.e., there would be a class such as ExtendedRenderModel that would be similar to the existing ExtendedLayoutModel. Virtually, all data structures beloning to the render extension would be linked as children to this element. On the other hand, however, it would also make sense to implement this "plugin" as a plugin for the ExtendedLayoutModel, because the render extension makes only sense in combination with the layout extension. Should we introduce a mechanism to extend (Abstract)SBasePlugin with further plugins? Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

Andreas Dräger — 2012-05-09T18:42:59-00:00

Am 5/8/12 3:43 PM, schrieb Nicolas Rodriguez: > On 08/05/12 14:30, Nicolas Rodriguez wrote: >> On 08/05/12 14:20, Andreas Dräger wrote: >>> Am 5/8/12 3:17 PM, schrieb Nicolas Rodriguez: >>>> On 08/05/12 13:57, Andreas Dräger wrote: >>>>> |Hi all, >>>>> >>>>> It seems the SBML writer does currently write invalid SBML code when >>>>> making use of the ability of SBML Level 3 to associate plain numbers >>>>> with units: >>>>> >>>>> 1</cn> >>>>> >>>>> The above code is only valid if the namespace declaration of the math >>>>> element also defines >>>>> >>>>> xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" >>>>> >>>>> Currently, it does only declare the regular MathML namesapce. As a >>>>> result, JSBML is currently unable to parse models that itself wrote into >>>>> a file before. Where can we fix this problem? >>>> The check is in >>>> org.sbml.jsbml.util.compilers.MathMLXMLStreamCompiler.isSBMLNamespaceNeeded(ASTNode) >>>> >>>> the namespace is added in >>>> org.sbml.jsbml.xml.stax.SBMLWriter.writeMathML(MathContainer, >>>> SMOutputElement, XMLStreamWriter, int) >>>> >>>> line 1084 to 1106. >>>> >>>> The compiler org.sbml.jsbml.util.compilers.FindUnitsCompiler is used to >>>> say if there is a unit declaration or not in the mathML. >>>> >>>> This is probably there that the problem happen, and the value returned >>>> is probably false. >>>> It is probably due to the way you are declaring the units and then how >>>> the mathML compiler is calling the methods. >>>> >>>> If I read a model with units, then the writing is fine so there is may >>>> be something missing when we declare units from scratch. >>>> >>>> >>> Hi Nico, >>> >>> Yes, it is probably because I simply say >>> >>> myASTNode.setUnit(Unit.Kind.DIMENSIONLESS.getName()); >>> >>> or something similar. I thought this would be enough and didn't care >>> about namespaces at the moment. >> It should be enough but it was probably not tested much until now. >> >> We should probably check how the reader is setting things up. > So, the reader is not doing anything specific, just using the setUnits() > method ( I suppose it was a typo error on your code as I don't see a > method setUnit(). > > Strange, it was working fine before. > > Hi all, here is a minimal example that doesn't work properly with the current version of JSBML: public static void main(String[] args) throws Throwable { SBMLDocument doc = new SBMLDocument(3, 1); Model m = doc.createModel("m1"); Compartment c = m.createCompartment("c1"); Species s1 = m.createSpecies("s1", c); Species s2 = m.createSpecies("s2", c); Reaction r = m.createReaction("r1"); r.createReactant(s1).setStoichiometry(1d); r.createProduct(s2).setStoichiometry(1d); KineticLaw kl = r.createKineticLaw(); ASTNode one = new ASTNode(1, kl); ASTNode two = new ASTNode(2, kl); one.setUnits(Unit.Kind.MOLE.toString().toLowerCase()); two.setUnits(Unit.Kind.SECOND.toString().toLowerCase()); ASTNode ast = ASTNode.frac(one, two); kl.setMath(ast); SBMLWriter.write(doc, System.out, ' ', (short) 2); } You'll see that the first unit is not written to SBML at all. The second unit is written, but the namespace declaration is missing. I already found what is probably the reason, but can't fix it at the moment: The recursive check if the ASTNode contains unit declarations doesn't work. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

rodrigue — 2012-05-08T14:10:18-00:00

On 08/05/12 14:43, Nicolas Rodriguez wrote: > On 08/05/12 14:30, Nicolas Rodriguez wrote: >> On 08/05/12 14:20, Andreas Dräger wrote: >>> Am 5/8/12 3:17 PM, schrieb Nicolas Rodriguez: >>>> On 08/05/12 13:57, Andreas Dräger wrote: >>>>> |Hi all, >>>>> >>>>> It seems the SBML writer does currently write invalid SBML code when >>>>> making use of the ability of SBML Level 3 to associate plain numbers >>>>> with units: >>>>> >>>>> 1</cn> >>>>> >>>>> The above code is only valid if the namespace declaration of the math >>>>> element also defines >>>>> >>>>> xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" >>>>> >>>>> Currently, it does only declare the regular MathML namesapce. As a >>>>> result, JSBML is currently unable to parse models that itself wrote into >>>>> a file before. Where can we fix this problem? >>>> The check is in >>>> org.sbml.jsbml.util.compilers.MathMLXMLStreamCompiler.isSBMLNamespaceNeeded(ASTNode) >>>> >>>> the namespace is added in >>>> org.sbml.jsbml.xml.stax.SBMLWriter.writeMathML(MathContainer, >>>> SMOutputElement, XMLStreamWriter, int) >>>> >>>> line 1084 to 1106. >>>> >>>> The compiler org.sbml.jsbml.util.compilers.FindUnitsCompiler is used to >>>> say if there is a unit declaration or not in the mathML. >>>> >>>> This is probably there that the problem happen, and the value returned >>>> is probably false. >>>> It is probably due to the way you are declaring the units and then how >>>> the mathML compiler is calling the methods. >>>> >>>> If I read a model with units, then the writing is fine so there is may >>>> be something missing when we declare units from scratch. >>>> >>>> >>> Hi Nico, >>> >>> Yes, it is probably because I simply say >>> >>> myASTNode.setUnit(Unit.Kind.DIMENSIONLESS.getName()); >>> >>> or something similar. I thought this would be enough and didn't care >>> about namespaces at the moment. >> It should be enough but it was probably not tested much until now. >> >> We should probably check how the reader is setting things up. > So, the reader is not doing anything specific, just using the setUnits() > method ( I suppose it was a typo error on your code as I don't see a > method setUnit(). > > Strange, it was working fine before. Just checked it again using org.sbml.jsbml.ext.qual.BuildToyModelTest And it is working fine for me. Can you check which version of jsbml did you use to generate the models ? Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

rodrigue — 2012-05-08T13:43:37-00:00

On 08/05/12 14:30, Nicolas Rodriguez wrote: > On 08/05/12 14:20, Andreas Dräger wrote: >> Am 5/8/12 3:17 PM, schrieb Nicolas Rodriguez: >>> On 08/05/12 13:57, Andreas Dräger wrote: >>>> |Hi all, >>>> >>>> It seems the SBML writer does currently write invalid SBML code when >>>> making use of the ability of SBML Level 3 to associate plain numbers >>>> with units: >>>> >>>> 1</cn> >>>> >>>> The above code is only valid if the namespace declaration of the math >>>> element also defines >>>> >>>> xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" >>>> >>>> Currently, it does only declare the regular MathML namesapce. As a >>>> result, JSBML is currently unable to parse models that itself wrote into >>>> a file before. Where can we fix this problem? >>> The check is in >>> org.sbml.jsbml.util.compilers.MathMLXMLStreamCompiler.isSBMLNamespaceNeeded(ASTNode) >>> >>> the namespace is added in >>> org.sbml.jsbml.xml.stax.SBMLWriter.writeMathML(MathContainer, >>> SMOutputElement, XMLStreamWriter, int) >>> >>> line 1084 to 1106. >>> >>> The compiler org.sbml.jsbml.util.compilers.FindUnitsCompiler is used to >>> say if there is a unit declaration or not in the mathML. >>> >>> This is probably there that the problem happen, and the value returned >>> is probably false. >>> It is probably due to the way you are declaring the units and then how >>> the mathML compiler is calling the methods. >>> >>> If I read a model with units, then the writing is fine so there is may >>> be something missing when we declare units from scratch. >>> >>> >> Hi Nico, >> >> Yes, it is probably because I simply say >> >> myASTNode.setUnit(Unit.Kind.DIMENSIONLESS.getName()); >> >> or something similar. I thought this would be enough and didn't care >> about namespaces at the moment. > It should be enough but it was probably not tested much until now. > > We should probably check how the reader is setting things up. So, the reader is not doing anything specific, just using the setUnits() method ( I suppose it was a typo error on your code as I don't see a method setUnit(). Strange, it was working fine before. Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

rodrigue — 2012-05-08T13:30:25-00:00

On 08/05/12 14:20, Andreas Dräger wrote: > Am 5/8/12 3:17 PM, schrieb Nicolas Rodriguez: >> On 08/05/12 13:57, Andreas Dräger wrote: >>> |Hi all, >>> >>> It seems the SBML writer does currently write invalid SBML code when >>> making use of the ability of SBML Level 3 to associate plain numbers >>> with units: >>> >>> 1</cn> >>> >>> The above code is only valid if the namespace declaration of the math >>> element also defines >>> >>> xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" >>> >>> Currently, it does only declare the regular MathML namesapce. As a >>> result, JSBML is currently unable to parse models that itself wrote into >>> a file before. Where can we fix this problem? >> The check is in >> org.sbml.jsbml.util.compilers.MathMLXMLStreamCompiler.isSBMLNamespaceNeeded(ASTNode) >> >> the namespace is added in >> org.sbml.jsbml.xml.stax.SBMLWriter.writeMathML(MathContainer, >> SMOutputElement, XMLStreamWriter, int) >> >> line 1084 to 1106. >> >> The compiler org.sbml.jsbml.util.compilers.FindUnitsCompiler is used to >> say if there is a unit declaration or not in the mathML. >> >> This is probably there that the problem happen, and the value returned >> is probably false. >> It is probably due to the way you are declaring the units and then how >> the mathML compiler is calling the methods. >> >> If I read a model with units, then the writing is fine so there is may >> be something missing when we declare units from scratch. >> >> > Hi Nico, > > Yes, it is probably because I simply say > > myASTNode.setUnit(Unit.Kind.DIMENSIONLESS.getName()); > > or something similar. I thought this would be enough and didn't care > about namespaces at the moment. It should be enough but it was probably not tested much until now. We should probably check how the reader is setting things up. Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

Andreas Dräger — 2012-05-08T13:20:58-00:00

Am 5/8/12 3:17 PM, schrieb Nicolas Rodriguez: > On 08/05/12 13:57, Andreas Dräger wrote: >> |Hi all, >> >> It seems the SBML writer does currently write invalid SBML code when >> making use of the ability of SBML Level 3 to associate plain numbers >> with units: >> >> 1</cn> >> >> The above code is only valid if the namespace declaration of the math >> element also defines >> >> xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" >> >> Currently, it does only declare the regular MathML namesapce. As a >> result, JSBML is currently unable to parse models that itself wrote into >> a file before. Where can we fix this problem? > The check is in > org.sbml.jsbml.util.compilers.MathMLXMLStreamCompiler.isSBMLNamespaceNeeded(ASTNode) > > the namespace is added in > org.sbml.jsbml.xml.stax.SBMLWriter.writeMathML(MathContainer, > SMOutputElement, XMLStreamWriter, int) > > line 1084 to 1106. > > The compiler org.sbml.jsbml.util.compilers.FindUnitsCompiler is used to > say if there is a unit declaration or not in the mathML. > > This is probably there that the problem happen, and the value returned > is probably false. > It is probably due to the way you are declaring the units and then how > the mathML compiler is calling the methods. > > If I read a model with units, then the writing is fine so there is may > be something missing when we declare units from scratch. > > Hi Nico, Yes, it is probably because I simply say myASTNode.setUnit(Unit.Kind.DIMENSIONLESS.getName()); or something similar. I thought this would be enough and didn't care about namespaces at the moment. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Bug in SBMLWriter

rodrigue — 2012-05-08T13:17:59-00:00

On 08/05/12 13:57, Andreas Dräger wrote: > |Hi all, > > It seems the SBML writer does currently write invalid SBML code when > making use of the ability of SBML Level 3 to associate plain numbers > with units: > > 1</cn> > > The above code is only valid if the namespace declaration of the math > element also defines > > xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" > > Currently, it does only declare the regular MathML namesapce. As a > result, JSBML is currently unable to parse models that itself wrote into > a file before. Where can we fix this problem? The check is in org.sbml.jsbml.util.compilers.MathMLXMLStreamCompiler.isSBMLNamespaceNeeded(ASTNode) the namespace is added in org.sbml.jsbml.xml.stax.SBMLWriter.writeMathML(MathContainer, SMOutputElement, XMLStreamWriter, int) line 1084 to 1106. The compiler org.sbml.jsbml.util.compilers.FindUnitsCompiler is used to say if there is a unit declaration or not in the mathML. This is probably there that the problem happen, and the value returned is probably false. It is probably due to the way you are declaring the units and then how the mathML compiler is calling the methods. If I read a model with units, then the writing is fine so there is may be something missing when we declare units from scratch. Nico > Cheers > Andreas > | > ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Bug in SBMLWriter

Andreas Dräger — 2012-05-08T12:57:57-00:00

|Hi all, It seems the SBML writer does currently write invalid SBML code when making use of the ability of SBML Level 3 to associate plain numbers with units: 1 </cn> The above code is only valid if the namespace declaration of the math element also defines xmlns:sbml="<a href="http://www.sbml.org/sbml/level3/version1/core" target="_blank">http://www.sbml.org/sbml/level3/version1/core</a>" Currently, it does only declare the regular MathML namesapce. As a result, JSBML is currently unable to parse models that itself wrote into a file before. Where can we fix this problem? Cheers Andreas | -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: jsbml-svn commit list added

Mike Hucka — 2012-05-08T11:00:02-00:00

On Tue, 08 May 2012 11:30:21 +0100, Nicolas Rodriguez wrote: > And the bounces emails are just that the list might not be properly > configured to accept mailing coming from the committers with their > sourceforge user mail. > We have to look into that but I have not much time for that at the moment. Whoops, sorry about that! I forgot to set the necessary setting for this aspect. It should be now fixed, and you guy and Andreas should not receive moderator requests for those messages. MH ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: jsbml-svn commit list added

rodrigue — 2012-05-08T10:30:21-00:00

On 08/05/12 10:51, Andreas Dräger wrote: > Am 5/2/12 3:57 PM, schrieb Nicolas Rodriguez: >> On 01/05/12 18:19, Michael Hucka wrote: >>> FYI: >>> >>> I added an SVN commit hook and a mailing list, <a href="mailto:jsbml-svn@lists.sourceforge.net" target="_blank">jsbml-svn@lists.sourceforge.net</a>, so that people can subscribe to the list and receive email notifications of every commit made to the SVN repo for JSBML. >>> >>> We use this for libSBML and other projects, and I personally find it immensely useful to learn about changes as they are made. >> Thanks. >> >> I am using and I think Andreas as well the sourceforge feeds for the >> project which is quite nice as well. >> >> <a href="https://sourceforge.net/export/rss2_keepsake.php?group_id=279608" target="_blank">https://sourceforge.net/export/rss2_keepsake.php?group_id=279608</a> >> >> Then, if you subscribe to it, you get a message for each commit but also >> for other activities happening to the project like >> when a tracker item is created. >> > Hi Nico, > > How to switch this off again? It is really annoying. Also, I now always > get e-mails that I should confirm some mail in the bounces and really > don't have the time to do all that again and again. What do you want to switch off ? you can unsusbscribe went you want to the mailing list. And the bounces emails are just that the list might not be properly configured to accept mailing coming from the committers with their sourceforge user mail. We have to look into that but I have not much time for that at the moment. Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: jsbml-svn commit list added

Andreas Dräger — 2012-05-08T09:51:38-00:00

Am 5/2/12 3:57 PM, schrieb Nicolas Rodriguez: > On 01/05/12 18:19, Michael Hucka wrote: >> FYI: >> >> I added an SVN commit hook and a mailing list, <a href="mailto:jsbml-svn@lists.sourceforge.net" target="_blank">jsbml-svn@lists.sourceforge.net</a>, so that people can subscribe to the list and receive email notifications of every commit made to the SVN repo for JSBML. >> >> We use this for libSBML and other projects, and I personally find it immensely useful to learn about changes as they are made. > Thanks. > > I am using and I think Andreas as well the sourceforge feeds for the > project which is quite nice as well. > > <a href="https://sourceforge.net/export/rss2_keepsake.php?group_id=279608" target="_blank">https://sourceforge.net/export/rss2_keepsake.php?group_id=279608</a> > > Then, if you subscribe to it, you get a message for each commit but also > for other activities happening to the project like > when a tracker item is created. > Hi Nico, How to switch this off again? It is really annoying. Also, I now always get e-mails that I should confirm some mail in the bounces and really don't have the time to do all that again and again. Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: jsbml-svn commit list added

Mike Hucka — 2012-05-03T00:50:56-00:00

On Wed, 02 May 2012 14:57:03 +0100, Nicolas Rodriguez wrote: > On 01/05/12 18:19, Michael Hucka wrote: > I am using and I think Andreas as well the sourceforge feeds for the > project which is quite nice as well. > > <a href="https://sourceforge.net/export/rss2_keepsake.php?group_id=279608" target="_blank">https://sourceforge.net/export/rss2_keepsake.php?group_id=279608</a> Good thinking. So, now there are 2 ways for people to be notified of changes, which is a good thing. (I stopped using RSS some time ago after switching to Twitter for news & notifications, so for me it's more convenient to use email notification for this stuff.) MH ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: jsbml-svn commit list added

rodrigue — 2012-05-02T13:57:03-00:00

On 01/05/12 18:19, Michael Hucka wrote: > FYI: > > I added an SVN commit hook and a mailing list, <a href="mailto:jsbml-svn@lists.sourceforge.net" target="_blank">jsbml-svn@lists.sourceforge.net</a>, so that people can subscribe to the list and receive email notifications of every commit made to the SVN repo for JSBML. > > We use this for libSBML and other projects, and I personally find it immensely useful to learn about changes as they are made. Thanks. I am using and I think Andreas as well the sourceforge feeds for the project which is quite nice as well. <a href="https://sourceforge.net/export/rss2_keepsake.php?group_id=279608" target="_blank">https://sourceforge.net/export/rss2_keepsake.php?group_id=279608</a> Then, if you subscribe to it, you get a message for each commit but also for other activities happening to the project like when a tracker item is created. Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

jsbml-svn commit list added

Mike Hucka — 2012-05-01T17:19:53-00:00

FYI: I added an SVN commit hook and a mailing list, <a href="mailto:jsbml-svn@lists.sourceforge.net" target="_blank">jsbml-svn@lists.sourceforge.net</a>, so that people can subscribe to the list and receive email notifications of every commit made to the SVN repo for JSBML. We use this for libSBML and other projects, and I personally find it immensely useful to learn about changes as they are made. MH -- Mike Hucka, Ph.D. -- <a href="mailto:mhucka@caltech.edu" target="_blank">mhucka@caltech.edu</a> -- <a href="http://www.cds.caltech.edu/~mhucka" target="_blank">http://www.cds.caltech.edu/~mhucka</a> Computing and Mathematical Sciences, California Institute of Technology, Pasadena, California, USA -- Twitter: @mhucka -- Skype: michaelhucka ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Eclipse templates for new JSBML classes

Andreas Dräger — 2012-04-26T05:22:02-00:00

Am 4/24/12 12:14 PM, schrieb Nicolas Rodriguez: > Not a big deal but sometime, you have classes that do not have id so may > be we could have two constructor templates ? > > Also, in writeXMLAttributes, you don't need to do > > // attributes.remove("validationXMLAllchecks"); > > for each attribute, only for the attribute that are already present in > the extended core classes. > So only for id and name in fact. We could do an AbstractExtNamedSBase > that would implement only the > writeXMLAttributes for id and name so that we don't need to do that on > every NamedSBase in the extensions. > > It is still on my todo list to update the user guide about these > read/write XML attributes methods btw, I haven't forgotten :-) > > Hi, Maybe it won't always be possible to extend AbstractExtNamedSBase in extension packages because the specifications sometimes may define an inheritance structure that might require a different inheritance relationship. But I agree, if possible we should use such a class. By the way, some time ago I also proposed another change. Here is my previous text: I would like to ask why we decided that the interface SBasePlugin does not extend the interface TreeNodeWithChangeSupport and why the abstract class AbstractSBasePlugin does not extend AbstractTreeNode. I understand that extensions to some SBase are not a separate TreeNode, but since the getChildAt(int) method simply excludes extension elements anyway (but not their children), we could avoid a duplication of code by making AbstractSBasePlugin a derivative of AbstractTreeNode (all other methods seem to be simimilar). In this way, it would also become possible to maintain userObjects within instances of SBasePlugin, which is not possible at the moment (it would again require a duplication of code). Cheers Andreas -- Dr. Andreas DrÃ¤ger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 TÃ¼bingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Eclipse templates for new JSBML classes

Mike Hucka — 2012-04-24T14:02:32-00:00

On Tue, 24 Apr 2012 10:20:18 +0100, Nicolas Rodriguez wrote: > dev/eclipse sound fine to me. > Also, it could be put outside of the core as the latest template was > made for the extensions You're right; good point. MH ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Eclipse templates for new JSBML classes

rodrigue — 2012-04-24T10:14:53-00:00

Not a big deal but sometime, you have classes that do not have id so may be we could have two constructor templates ? Also, in writeXMLAttributes, you don't need to do // attributes.remove("validationXMLAllchecks"); for each attribute, only for the attribute that are already present in the extended core classes. So only for id and name in fact. We could do an AbstractExtNamedSBase that would implement only the writeXMLAttributes for id and name so that we don't need to do that on every NamedSBase in the extensions. It is still on my todo list to update the user guide about these read/write XML attributes methods btw, I haven't forgotten :-) Nico On 24/04/12 10:54, Finja BÃ¼chel wrote: > Hi Mike, > > thanks for your feedback. I like the idea of combining all templates in > one folder. The folder dev and subfoler eclipse would be fine for me. If > nobody is against creating such a folder, I will create and commit it at > the end of the week. Additionally, I will include a README.txt, because > we have four different templates. Unfortunately, we can't combine them > in one file: > > 1) the core/files/java.java contains the header which should be included > in all java class files. We can also delete this file, because if > everybody includes the core/files/codetemplates.xml (see next point 2) > the header is also included. But I'm not sure if everybody includes this > template... > > 2) the core/files/codetemplates.xml can be included in eclipse under the > project properties in "Java" -> "Code Style" -> "Code templates". The > file defines the header and some standard comments. > > 3) the core/files/SBML_Project_Java_style_for_Eclipse_3.6.xml should be > included in "Java" -> "Code Style" -> "Formatter". > > 4) the core/doc/templates should be included in "Java" -> "Editor" -> > "Templates", and can be used for easy code creation. > > While creating the templates Florian and I observed that the code isn't > indented properly. Did you see other coding style mistakes? > > We tried to fix the indentation problem, but found no other solution > than pressing "ctr + Shift + F" after automatic code generation. > Especially, if we started templates with a variable declaration the > indentation was even worse. That's the reason for defining the variables > at the end of the "JSBML_Constructors"-template. If somebody has a > solution for this problem I would be happy to change the templates. > > Best wishes, > Finja > > > > > Am 23.04.2012 19:06, schrieb Michael Hucka: >> Hi, >> >> Thanks for doing this! This is a great start, and I'd like to ask for a few changes that I think would make this even better: >> >> 1) There were already some templates in core/files/codetemplates.xml. Could the new templates be merged or somehow integrated with the old ones? >> >> 2) The new templates don't actually seem to follow the coding style established by core/files/SBML_Project_Java_style_for_Eclipse_3.6.xml. Could they be updated, to be consistent? (To be clear, I'm not so much requiring that we stick to this proposed coding style as I am requesting adherence to *some* style. ) >> >> 3) Neither location (core/files/ and doc/templates/) seems best for Eclipse configuration files like this. I'd like to propose that we introduce a directory "dev" within "core" at the same level as "doc" and "files", and then have "eclipse" within that: >> >> dev/ >> eclipse/ >> doc/ >> files/ >> lib/ >> â€¦ >> >> (Unless people can think of an even better location.) The reason is that these files are not really documentation files (so don't belong in "doc") and since there are now two of them (the templates and the code style for Eclipse), it seems like they should go in their own subdirectory. The name "dev" makes it more clear what this is for, and in the future, there might be other things that go into "dev". While we are at it, I think I would remove doc/templates altogether, because the only other file in there is subsumed by the Eclipse code templates. >> >> Thanks, >> MH >> -- >> Mike Hucka, Ph.D. -- <a href="mailto:mhucka@caltech.edu" target="_blank">mhucka@caltech.edu</a> -- <a href="http://www.cds.caltech.edu/~mhucka" target="_blank">http://www.cds.caltech.edu/~mhucka</a> >> Computing and Mathematical Sciences, California Institute of Technology, >> Pasadena, California, USA -- Twitter: @mhucka -- Skype: michaelhucka >> >> >> On Mon, 16 Apr 2012 15:48:04 +0200, Finja BÃ¼chel wrote: >>> Hi everybody, >>> >>> Florian and I implemented some JSBML templates for eclipse to facilitate >>> programming of new JSBML classes. We created templates for the standard >>> constructors, get-, set-, unset- methods for class variables, methods >>> for ListOf element handling and XML attribute writing. >>> >>> The template xml can be found in core/doc/templates/JSBML_templates.xml. >>> >>> Just import the templates in your eclipse project templates. >>> >>> We are looking forward to some feedback. >>> >>> Best wishes >>> Florian and Finja >>> >>> >>> -- >>> Dipl.-Inform. (Bioinform.) Finja BÃ¼chel >>> Eberhard-Karls-UniversitÃ¤t TÃ¼bingen >>> Lehrstuhl Kognitive Systeme >>> Sand 1 >>> 72076 TÃ¼bingen >>> Germany >>> >>> Tel: (+49/0) 7071 / 29 70436 >>> Fax: (+49/0) 7071 / 29 5091 >>> <a href="mailto:finja.buechel@uni-tuebingen.de" target="_blank">finja.buechel@uni-tuebingen.de</a> >>> ____________________________________________________________ >>> To manage your jsbml-development list subscription, visit >>> <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> >>> >>> For a web interface to the jsbml-development mailing list, visit >>> <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> >>> >>> For questions or feedback about the jsbml-development list, >>> contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> >> ____________________________________________________________ >> To manage your jsbml-development list subscription, visit >> <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> >> >> For a web interface to the jsbml-development mailing list, visit >> <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> >> >> For questions or feedback about the jsbml-development list, >> contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Eclipse templates for new JSBML classes

finja.buechel — 2012-04-24T09:54:38-00:00

Hi Mike, thanks for your feedback. I like the idea of combining all templates in one folder. The folder dev and subfoler eclipse would be fine for me. If nobody is against creating such a folder, I will create and commit it at the end of the week. Additionally, I will include a README.txt, because we have four different templates. Unfortunately, we can't combine them in one file: 1) the core/files/java.java contains the header which should be included in all java class files. We can also delete this file, because if everybody includes the core/files/codetemplates.xml (see next point 2) the header is also included. But I'm not sure if everybody includes this template... 2) the core/files/codetemplates.xml can be included in eclipse under the project properties in "Java" -> "Code Style" -> "Code templates". The file defines the header and some standard comments. 3) the core/files/SBML_Project_Java_style_for_Eclipse_3.6.xml should be included in "Java" -> "Code Style" -> "Formatter". 4) the core/doc/templates should be included in "Java" -> "Editor" -> "Templates", and can be used for easy code creation. While creating the templates Florian and I observed that the code isn't indented properly. Did you see other coding style mistakes? We tried to fix the indentation problem, but found no other solution than pressing "ctr + Shift + F" after automatic code generation. Especially, if we started templates with a variable declaration the indentation was even worse. That's the reason for defining the variables at the end of the "JSBML_Constructors"-template. If somebody has a solution for this problem I would be happy to change the templates. Best wishes, Finja Am 23.04.2012 19:06, schrieb Michael Hucka: > Hi, > > Thanks for doing this! This is a great start, and I'd like to ask for a few changes that I think would make this even better: > > 1) There were already some templates in core/files/codetemplates.xml. Could the new templates be merged or somehow integrated with the old ones? > > 2) The new templates don't actually seem to follow the coding style established by core/files/SBML_Project_Java_style_for_Eclipse_3.6.xml. Could they be updated, to be consistent? (To be clear, I'm not so much requiring that we stick to this proposed coding style as I am requesting adherence to *some* style. ) > > 3) Neither location (core/files/ and doc/templates/) seems best for Eclipse configuration files like this. I'd like to propose that we introduce a directory "dev" within "core" at the same level as "doc" and "files", and then have "eclipse" within that: > > dev/ > eclipse/ > doc/ > files/ > lib/ > â€¦ > > (Unless people can think of an even better location.) The reason is that these files are not really documentation files (so don't belong in "doc") and since there are now two of them (the templates and the code style for Eclipse), it seems like they should go in their own subdirectory. The name "dev" makes it more clear what this is for, and in the future, there might be other things that go into "dev". While we are at it, I think I would remove doc/templates altogether, because the only other file in there is subsumed by the Eclipse code templates. > > Thanks, > MH > -- > Mike Hucka, Ph.D. -- <a href="mailto:mhucka@caltech.edu" target="_blank">mhucka@caltech.edu</a> -- <a href="http://www.cds.caltech.edu/~mhucka" target="_blank">http://www.cds.caltech.edu/~mhucka</a> > Computing and Mathematical Sciences, California Institute of Technology, > Pasadena, California, USA -- Twitter: @mhucka -- Skype: michaelhucka > > > On Mon, 16 Apr 2012 15:48:04 +0200, Finja BÃ¼chel wrote: >> Hi everybody, >> >> Florian and I implemented some JSBML templates for eclipse to facilitate >> programming of new JSBML classes. We created templates for the standard >> constructors, get-, set-, unset- methods for class variables, methods >> for ListOf element handling and XML attribute writing. >> >> The template xml can be found in core/doc/templates/JSBML_templates.xml. >> >> Just import the templates in your eclipse project templates. >> >> We are looking forward to some feedback. >> >> Best wishes >> Florian and Finja >> >> >> -- >> Dipl.-Inform. (Bioinform.) Finja BÃ¼chel >> Eberhard-Karls-UniversitÃ¤t TÃ¼bingen >> Lehrstuhl Kognitive Systeme >> Sand 1 >> 72076 TÃ¼bingen >> Germany >> >> Tel: (+49/0) 7071 / 29 70436 >> Fax: (+49/0) 7071 / 29 5091 >> <a href="mailto:finja.buechel@uni-tuebingen.de" target="_blank">finja.buechel@uni-tuebingen.de</a> >> ____________________________________________________________ >> To manage your jsbml-development list subscription, visit >> <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> >> >> For a web interface to the jsbml-development mailing list, visit >> <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> >> >> For questions or feedback about the jsbml-development list, >> contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> -- Dipl.-Inform. (Bioinform.) Finja BÃ¼chel Eberhard-Karls-UniversitÃ¤t TÃ¼bingen Lehrstuhl Kognitive Systeme Sand 1 72076 TÃ¼bingen Germany Tel: (+49/0) 7071 / 29 70436 Fax: (+49/0) 7071 / 29 5091 <a href="mailto:finja.buechel@uni-tuebingen.de" target="_blank">finja.buechel@uni-tuebingen.de</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Eclipse templates for new JSBML classes

rodrigue — 2012-04-24T09:20:18-00:00

On 23/04/12 18:06, Michael Hucka wrote: > Hi, > > Thanks for doing this! This is a great start, and I'd like to ask for a few changes that I think would make this even better: > > 1) There were already some templates in core/files/codetemplates.xml. Could the new templates be merged or somehow integrated with the old ones? > > 2) The new templates don't actually seem to follow the coding style established by core/files/SBML_Project_Java_style_for_Eclipse_3.6.xml. Could they be updated, to be consistent? (To be clear, I'm not so much requiring that we stick to this proposed coding style as I am requesting adherence to *some* style. ) > > 3) Neither location (core/files/ and doc/templates/) seems best for Eclipse configuration files like this. I'd like to propose that we introduce a directory "dev" within "core" at the same level as "doc" and "files", and then have "eclipse" within that: > > dev/ > eclipse/ > doc/ > files/ > lib/ > â€¦ > > (Unless people can think of an even better location.) The reason is that these files are not really documentation files (so don't belong in "doc") and since there are now two of them (the templates and the code style for Eclipse), it seems like they should go in their own subdirectory. The name "dev" makes it more clear what this is for, and in the future, there might be other things that go into "dev". While we are at it, I think I would remove doc/templates altogether, because the only other file in there is subsumed by the Eclipse code templates. dev/eclipse sound fine to me. Also, it could be put outside of the core as the latest template was made for the extensions Nico > > On Mon, 16 Apr 2012 15:48:04 +0200, Finja BÃ¼chel wrote: >> Hi everybody, >> >> Florian and I implemented some JSBML templates for eclipse to facilitate >> programming of new JSBML classes. We created templates for the standard >> constructors, get-, set-, unset- methods for class variables, methods >> for ListOf element handling and XML attribute writing. >> >> The template xml can be found in core/doc/templates/JSBML_templates.xml. >> >> Just import the templates in your eclipse project templates. >> >> We are looking forward to some feedback. >> >> Best wishes >> Florian and Finja >> >> >> -- >> Dipl.-Inform. (Bioinform.) Finja BÃ¼chel >> Eberhard-Karls-UniversitÃ¤t TÃ¼bingen >> Lehrstuhl Kognitive Systeme >> Sand 1 >> 72076 TÃ¼bingen >> Germany >> >> Tel: (+49/0) 7071 / 29 70436 >> Fax: (+49/0) 7071 / 29 5091 >> <a href="mailto:finja.buechel@uni-tuebingen.de" target="_blank">finja.buechel@uni-tuebingen.de</a> >> ____________________________________________________________ >> To manage your jsbml-development list subscription, visit >> <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> >> >> For a web interface to the jsbml-development mailing list, visit >> <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> >> >> For questions or feedback about the jsbml-development list, >> contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Eclipse templates for new JSBML classes

Mike Hucka — 2012-04-23T17:06:01-00:00

Hi, Thanks for doing this! This is a great start, and I'd like to ask for a few changes that I think would make this even better: 1) There were already some templates in core/files/codetemplates.xml. Could the new templates be merged or somehow integrated with the old ones? 2) The new templates don't actually seem to follow the coding style established by core/files/SBML_Project_Java_style_for_Eclipse_3.6.xml. Could they be updated, to be consistent? (To be clear, I'm not so much requiring that we stick to this proposed coding style as I am requesting adherence to *some* style. ) 3) Neither location (core/files/ and doc/templates/) seems best for Eclipse configuration files like this. I'd like to propose that we introduce a directory "dev" within "core" at the same level as "doc" and "files", and then have "eclipse" within that: dev/ eclipse/ doc/ files/ lib/ â€¦ (Unless people can think of an even better location.) The reason is that these files are not really documentation files (so don't belong in "doc") and since there are now two of them (the templates and the code style for Eclipse), it seems like they should go in their own subdirectory. The name "dev" makes it more clear what this is for, and in the future, there might be other things that go into "dev". While we are at it, I think I would remove doc/templates altogether, because the only other file in there is subsumed by the Eclipse code templates. Thanks, MH -- Mike Hucka, Ph.D. -- <a href="mailto:mhucka@caltech.edu" target="_blank">mhucka@caltech.edu</a> -- <a href="http://www.cds.caltech.edu/~mhucka" target="_blank">http://www.cds.caltech.edu/~mhucka</a> Computing and Mathematical Sciences, California Institute of Technology, Pasadena, California, USA -- Twitter: @mhucka -- Skype: michaelhucka On Mon, 16 Apr 2012 15:48:04 +0200, Finja BÃ¼chel wrote: > Hi everybody, > > Florian and I implemented some JSBML templates for eclipse to facilitate > programming of new JSBML classes. We created templates for the standard > constructors, get-, set-, unset- methods for class variables, methods > for ListOf element handling and XML attribute writing. > > The template xml can be found in core/doc/templates/JSBML_templates.xml. > > Just import the templates in your eclipse project templates. > > We are looking forward to some feedback. > > Best wishes > Florian and Finja > > > -- > Dipl.-Inform. (Bioinform.) Finja BÃ¼chel > Eberhard-Karls-UniversitÃ¤t TÃ¼bingen > Lehrstuhl Kognitive Systeme > Sand 1 > 72076 TÃ¼bingen > Germany > > Tel: (+49/0) 7071 / 29 70436 > Fax: (+49/0) 7071 / 29 5091 > <a href="mailto:finja.buechel@uni-tuebingen.de" target="_blank">finja.buechel@uni-tuebingen.de</a> > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: SBML Layout Extension: New convenient abstract class NamedSBaseGlyph

rodrigue — 2012-04-23T09:00:42-00:00

On 23/04/12 07:40, Andreas Dräger wrote: > Dear all, > > Matthias König suggests to implement a new abstract class > NamedSBaseGlyph that would inherit from GraphicalObject and be the new > super class of SpeciesGlyph, ReactionGlyph, ComparmtentGlyph, and > SpeciesReferenceGlyph. The most important methods would deal with > setting/getting the identifier of the > Species/Compartment/Reaction/SpeciesReference to be drawn by the glyph. > This would simplify the implementation of layout algorithms. If you're > happy with this, I would like to implement this class. That's sound good to me. BTW, we still need to make sure that we are doing the same layout extension as libSBML so that there is not two slightly different versions out there. Cheers, Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

SBML Layout Extension: New convenient abstract class NamedSBaseGlyph

Andreas Dräger — 2012-04-23T06:40:12-00:00

Dear all, Matthias König suggests to implement a new abstract class NamedSBaseGlyph that would inherit from GraphicalObject and be the new super class of SpeciesGlyph, ReactionGlyph, ComparmtentGlyph, and SpeciesReferenceGlyph. The most important methods would deal with setting/getting the identifier of the Species/Compartment/Reaction/SpeciesReference to be drawn by the glyph. This would simplify the implementation of layout algorithms. If you're happy with this, I would like to implement this class. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

[jsbml-team] jsbml-rc2 to test

rodrigue — 2012-04-17T10:41:04-00:00

------------------------- JSBML-0.8-rc2 ------------------------- We are pleased to announce the second release candidate of JSBML-0.8 which is available for download from <a href="https://sourceforge.net/projects/jsbml/files/jsbml/0.8-rc2" target="_blank">https://sourceforge.net/projects/jsbml/files/jsbml/0.8-rc2</a> This will be a short release and we are planning to do the final 0.8 during the next few weeks if no major problems are found. ============================================= Version 0.8-rc2 (11-04-2012) ============================================= Please try it and report any problems you encounter using JSBML either to the mailing list <a href="mailto:jsbml-development@caltech.edu" target="_blank">jsbml-development@caltech.edu</a> or using the issue tracker for JSBML at <a href="http://sourceforge.net/tracker/?group_id=279608&atid=1186776" target="_blank">http://sourceforge.net/tracker/?group_id=279608&atid=1186776</a> Below is a summary of the changes visible to the users : * User visible changes : - Major improvements of the reading and writing speed for large models. * Bug Fixes: - It was forgotten to implement an unsetSubstanceUnits method on the Model class. Thanks to Robert Byrnes for reporting this problem. - In the Reaction class, the setReversible method only accepted Boolean, no boolean. - Several bug fixes and improvements to units derivation, merging and simplifying. - There was a bug in KineticLaw that prevented a correct iteration through all child elements. - When cloning JSBML wrongly assigned the same TreeNodeChangeListeners to the new data structure. Now, the TreeNodeChangeListeners are not copied when cloning a JSBML object. - The clone methods were corrected to register properly all entities with an id. - QuantityWithUnit did not fire update events when changing its value because of an incorrect comparison between the previous and the new value. - Corrected a bug that made it impossible to set a kinetic law with an existing list of local parameters as the new kinetic law in a different reaction. - Meike Aichele detected and fixed the problem that some Listeners were set too soon in the Model class, so that no Event was detected. - One rdf namespace declaration was missing when creating an History element from scratch. - [Tracker item #3511439] : Model.getSpeciesType(String) was corrected and it's speed improved. Thanks to Thomas Bernard who noticed and reported the problem. - Annotation elements that are missing a proper namespace declaration will be correctly read and not considered as improper SBML elements. - [Tracker item #3487517] : If one empty rdf Bag element was present in an annotation read by jsbml, an XML element was not closed properly, resulting in an invalid generated SBML. Thanks to Anna Zhukova to report this problem. - [Tracker item #3434930] : The parsing of a Message as an XML String was incorrect as the method was expecting only a notes element to enclose the HTML. * Known issues : - Reading of more than one RDF Description elements in the RDF annotation block (the official SBML annotation scheme) is not supported at the moment. Planned for the next major release. - The RDF block at the top level of the annotation is considered to be the official SBML annotation. - SBML L3 packages are not supported and not kept. ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Eclipse templates for new JSBML classes

rodrigue — 2012-04-16T16:11:18-00:00

On 16/04/12 16:57, Nicolas Rodriguez wrote: > On 16/04/12 16:54, Nicolas Rodriguez wrote: >> On 16/04/12 14:48, Finja Büchel wrote: >>> Hi everybody, >>> >>> Florian and I implemented some JSBML templates for eclipse to facilitate >>> programming of new JSBML classes. We created templates for the standard >>> constructors, get-, set-, unset- methods for class variables, methods >>> for ListOf element handling and XML attribute writing. >>> >>> The template xml can be found in core/doc/templates/JSBML_templates.xml. >>> >>> Just import the templates in your eclipse project templates. >> So I tried to go to my project properties, then 'Java Code Style' -> >> 'Code Templates'. From there I tried to import >> you xml files but I don't see any new or updated templates in the list. >> >> I am using eclipse Version: 3.7.2. >> >> Am I doing something wrong ? How do we use the templates afterward ? > Ok, found the right place (I thing) : 'Java' -> 'Editor' -> 'Templates' It seems to be really nice for SBML L3 packages developers. Just a note, as I did not knew that. When you have several fields to rename, you have to use 'tab' and not 'enter' to be able rename them all in one go. Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Eclipse templates for new JSBML classes

rodrigue — 2012-04-16T15:57:33-00:00

On 16/04/12 16:54, Nicolas Rodriguez wrote: > On 16/04/12 14:48, Finja Büchel wrote: >> Hi everybody, >> >> Florian and I implemented some JSBML templates for eclipse to facilitate >> programming of new JSBML classes. We created templates for the standard >> constructors, get-, set-, unset- methods for class variables, methods >> for ListOf element handling and XML attribute writing. >> >> The template xml can be found in core/doc/templates/JSBML_templates.xml. >> >> Just import the templates in your eclipse project templates. > So I tried to go to my project properties, then 'Java Code Style' -> > 'Code Templates'. From there I tried to import > you xml files but I don't see any new or updated templates in the list. > > I am using eclipse Version: 3.7.2. > > Am I doing something wrong ? How do we use the templates afterward ? Ok, found the right place (I thing) : 'Java' -> 'Editor' -> 'Templates' Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Eclipse templates for new JSBML classes

rodrigue — 2012-04-16T15:54:04-00:00

On 16/04/12 14:48, Finja Büchel wrote: > Hi everybody, > > Florian and I implemented some JSBML templates for eclipse to facilitate > programming of new JSBML classes. We created templates for the standard > constructors, get-, set-, unset- methods for class variables, methods > for ListOf element handling and XML attribute writing. > > The template xml can be found in core/doc/templates/JSBML_templates.xml. > > Just import the templates in your eclipse project templates. So I tried to go to my project properties, then 'Java Code Style' -> 'Code Templates'. From there I tried to import you xml files but I don't see any new or updated templates in the list. I am using eclipse Version: 3.7.2. Am I doing something wrong ? How do we use the templates afterward ? Thanks, Nico ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Eclipse templates for new JSBML classes

finja.buechel — 2012-04-16T13:48:04-00:00

Hi everybody, Florian and I implemented some JSBML templates for eclipse to facilitate programming of new JSBML classes. We created templates for the standard constructors, get-, set-, unset- methods for class variables, methods for ListOf element handling and XML attribute writing. The template xml can be found in core/doc/templates/JSBML_templates.xml. Just import the templates in your eclipse project templates. We are looking forward to some feedback. Best wishes Florian and Finja -- Dipl.-Inform. (Bioinform.) Finja Büchel Eberhard-Karls-Universität Tübingen Lehrstuhl Kognitive Systeme Sand 1 72076 Tübingen Germany Tel: (+49/0) 7071 / 29 70436 Fax: (+49/0) 7071 / 29 5091 <a href="mailto:finja.buechel@uni-tuebingen.de" target="_blank">finja.buechel@uni-tuebingen.de</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Please register for HARMONY 2012

Mike Hucka — 2012-04-16T00:27:12-00:00

For all who are planning on attending HARMONY 2012, please register soon. Please don't wait until the last minute! The organizers need to make plans for the number of people who will be attending. Here is a direct link to the registration page: <a href="http://harmony2012.eventbrite.com/" target="_blank">http://harmony2012.eventbrite.com/</a> Here is the event home page: <a href="http://co.mbine.org/events/HARMONY_2012" target="_blank">http://co.mbine.org/events/HARMONY_2012</a> H A R M O N Y 2012 May 21-25, 2012 Maastricht, The Netherlands Thanks! -- Mike Hucka, Ph.D. -- <a href="mailto:mhucka@caltech.edu" target="_blank">mhucka@caltech.edu</a> -- <a href="http://www.cds.caltech.edu/~mhucka" target="_blank">http://www.cds.caltech.edu/~mhucka</a> Computing and Mathematical Sciences, California Institute of Technology, Pasadena, California, USA -- Twitter: @mhucka -- Skype: michaelhucka ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Constructors in extension elements

Andreas Dräger — 2012-04-12T05:03:28-00:00

Hi Michael, > On Tue, 10 Apr 2012 12:12:12 +0200, Andreas Dräger wrote: >> Is there already a general rule which constructors are required when >> implementing extension elements? >> It seems there should there always be an empty constructor and one that >> takes the usual arguments Level and Version. But how about the package >> version? > If this doesn't exist already, there should be common guidelines for every extension -- i.e., rules that say there should always be a constructor, always be methods X, Y, and Z, etc. All extensions should follow a minimum interface set. For methods we have it already. The user guide nicely explains which methods should be present in extension elements. For constructors it is missing/unclear at the moment. >> Should constructors of package elements throw an exception when trying >> to initialize them with Level Version combinations before 3.1 (but with >> a defined LV, i.e., other than -1.-1)? > This makes sense to me. It could be a IllegalArgumentException. > > There is, however, an exception with the Layout extension that can be stored within the annotation element prior to L3. Would it actually make sense to follow this approach with all extension packages? I mean, could we store arbitrary SBML extensions within annotation elements if the Level is less than 3? Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Constructors in extension elements

Mike Hucka — 2012-04-11T22:47:01-00:00

On Tue, 10 Apr 2012 12:12:12 +0200, Andreas Dräger wrote: > Is there already a general rule which constructors are required when > implementing extension elements? > It seems there should there always be an empty constructor and one that > takes the usual arguments Level and Version. But how about the package > version? If this doesn't exist already, there should be common guidelines for every extension -- i.e., rules that say there should always be a constructor, always be methods X, Y, and Z, etc. All extensions should follow a minimum interface set. > Should constructors of package elements throw an exception when trying > to initialize them with Level Version combinations before 3.1 (but with > a defined LV, i.e., other than -1.-1)? This makes sense to me. It could be a IllegalArgumentException. MH ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: building extensions

rodrigue — 2012-04-11T15:46:14-00:00

Thanks Martijn, I applied the changes to the extension build files although slightly differently as I did it before you mail went through. Nico On 11/04/12 16:36, mvaniersel wrote: > In my setup, the SBML-qual build file can't find the core library anymore. > > Here is what I did: > > svn co <a href="https://jsbml.svn.sourceforge.net/svnroot/jsbml/trunk" target="_blank">https://jsbml.svn.sourceforge.net/svnroot/jsbml/trunk</a> jsbml > cd jsbml/core > ant jar > cd ../extensions/qual > ant jar > > > After the last command, I get a complaint that the directories jsbml/build and jsbml/lib don't exist. However, it should use jsbml/core/build and jsbml/core/lib. > > > I propose the following patch, which lets you override the core directory location more easily with e.g. "ant jar -Djsbml.core.dir=../../core" > > > Index: extensions/qual/build.xml > =================================================================== > --- extensions/qual/build.xml (revision 1214) > +++ extensions/qual/build.xml (working copy) > @@ -46,9 +46,10 @@ > > > > - > - > - > + > + > + > + > > > > > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

building extensions

mvaniersel — 2012-04-11T10:27:58-00:00

In my setup, the SBML-qual build file can't find the core library anymore. Here is what I did: svn co <a href="https://jsbml.svn.sourceforge.net/svnroot/jsbml/trunk" target="_blank">https://jsbml.svn.sourceforge.net/svnroot/jsbml/trunk</a> jsbml cd jsbml/core ant jar cd ../extensions/qual ant jar After the last command, I get a complaint that the directories jsbml/build and jsbml/lib don't exist. However, it should use jsbml/core/build and jsbml/core/lib. I propose the following patch, which lets you override the core directory location more easily with e.g. "ant jar -Djsbml.core.dir=../../core" <pre> Index: extensions/qual/build.xml =================================================================== --- extensions/qual/build.xml (revision 1214) +++ extensions/qual/build.xml (working copy) @@ -46,9 +46,10 @@ - - - + + + +

</pre>

Constructors in extension elements

Andreas Dräger — 2012-04-10T10:12:12-00:00

Dear all, Is there already a general rule which constructors are required when implementing extension elements? It seems there should there always be an empty constructor and one that takes the usual arguments Level and Version. But how about the package version? How about more convenient constructors that take some arguments to initialize the data structure? Should constructors of package elements throw an exception when trying to initialize them with Level Version combinations before 3.1 (but with a defined LV, i.e., other than -1.-1)? At the moment it seems there are very diverse solutions in the packages that have already been implemented. Since we already have such a very nice tutorial that explains how to implement extension packages, we should also fix some rules about constructors. Finally, I would like to ask why we decided that the interface SBasePlugin does not extend the interface TreeNodeWithChangeSupport and why the abstract class AbstractSBasePlugin does not extend AbstractTreeNode. I understand that extensions to some SBase are not a separate TreeNode, but since the getChildAt(int) method simply excludes extension elements anyway (but not their children), we could avoid a duplication of code by making AbstractSBasePlugin a derivative of AbstractTreeNode (all other methods seem to be simimilar). In this way, it would also become possible to maintain userObjects within instances of SBasePlugin, which is not possible at the moment (it would again require a duplication of code). Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: More issues with history elements

rbyrnes — 2012-04-04T16:40:43-00:00

Thanks, Nico! --JavaBob

infix survey

Lucian Smith — 2012-04-03T17:59:39-00:00

Andreas asked me to bring this up here: we are writing a new level 3 infix-to-mathml parser for libsbml, and have a survey about it: <a href="http://www.surveymonkey.com/s/YG76BDG" target="_blank">http://www.surveymonkey.com/s/YG76BDG</a> One of the main questions there is 'what do we do about 'log(x)'? Some traditions set the base as 10, some as e, and (Andreas tells me) still others as 2! I understand there's a jsbml infix parser; what does it do? If anyone here would like to fill out the survey, please do. -Lucian ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: More issues with history elements

rodrigue — 2012-04-02T08:53:57-00:00

Hi, The latest trunk should correct this, one test was missing for the rdf namespace for the History element. Cheers, Nico On 23/03/12 23:32, rbyrnes wrote: > I have a test program that constructs a history for a model and a compartment. Here is the program and its output. > > The validator at www.sbml.org reports that the output is valid SBML, but there is a problem with the model history. > > I found that the warning could be avoided by hand editing all the Bag and li elements to rdf:Bag and rdf:li. Also, I needed to add a modification date element to the history for the compartment to avoid a second warning. > > Are there easier ways to avoid these issues? > > --JavaBob > > import java.io.*; > > import org.sbml.jsbml.*; > > public class TestCompartmentHistoryGeneration > { > public static void main( String[] args ) > { > try > { > // Construct a model. > SBMLDocument document = new SBMLDocument( 3, 1 ); > Model model = document.createModel( "modelid" ); > model.setMetaId( "modelmetaid" ); > > // Add to the model history. > History mh = model.getHistory( ); > Creator creator = new Creator( ); > creator.setGivenName( "Steve" ); > creator.setFamilyName( "Jobs" ); > mh.addCreator( creator ); > mh.setCreatedDate( new java.util.Date( ) ); > > // Create a compartment and add to the compartment history. > Compartment c1 = model.createCompartment( "c1" ); > c1.setMetaId( "compartmentmetaid" ); > c1.setConstant( true ); > History h = c1.getHistory( ); > Creator cCreator = new Creator( ); > cCreator.setGivenName( "Steve" ); > cCreator.setFamilyName( "Jobs" ); > h.addCreator( cCreator ); > h.setCreatedDate( new java.util.Date( ) ); > c1.setHistory( h ); > > // Save to file. > File saveF = new File( "E:\\testsbml\\testcompartmenthistory.sbml" ); > SBMLWriter writer = new SBMLWriter( ); > FileOutputStream stream = new FileOutputStream( saveF ); > writer.write( document, stream ); > } > catch( Exception ex ) > { > ex.printStackTrace( ); > } > System.out.println( "Done" ); > } > } > > <?xml version='1.0' encoding='UTF-8' standalone='no'?> > > > > > > > > > > Jobs</vCard:Family> > Steve</vCard:Given> > </vCard:N> > </li> > </Bag> > </dc:creator> > > 2012-03-23T21:25:07Z</dcterms:W3CDTF> > </dcterms:created> > > 2012-03-23T21:25:07Z</dcterms:W3CDTF> > </dcterms:modified> > </rdf:Description> > </rdf:RDF> > </annotation> > > > > > > > > > > Jobs</vCard:Family> > Steve</vCard:Given> > </vCard:N> > </li> > </Bag> > </dc:creator> > > 2012-03-23T21:25:07Z</dcterms:W3CDTF> > </dcterms:created> > </rdf:Description> > </rdf:RDF> > </annotation> > </compartment> > </listOfCompartments> > </model> > </sbml> > > > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: More issues with history elements

rodrigue — 2012-03-28T15:35:40-00:00

On 23/03/12 23:32, rbyrnes wrote: > I have a test program that constructs a history for a model and a compartment. Here is the program and its output. > > The validator at www.sbml.org reports that the output is valid SBML, but there is a problem with the model history. > > I found that the warning could be avoided by hand editing all the Bag and li elements to rdf:Bag and rdf:li. Also, I needed to add a modification date element to the history for the compartment to avoid a second warning. > A quick reply to say that this is not forgotten and is probably a bug. I will have a look at it a bit later, probably next week. Thanks, Nico > > import java.io.*; > > import org.sbml.jsbml.*; > > public class TestCompartmentHistoryGeneration > { > public static void main( String[] args ) > { > try > { > // Construct a model. > SBMLDocument document = new SBMLDocument( 3, 1 ); > Model model = document.createModel( "modelid" ); > model.setMetaId( "modelmetaid" ); > > // Add to the model history. > History mh = model.getHistory( ); > Creator creator = new Creator( ); > creator.setGivenName( "Steve" ); > creator.setFamilyName( "Jobs" ); > mh.addCreator( creator ); > mh.setCreatedDate( new java.util.Date( ) ); > > // Create a compartment and add to the compartment history. > Compartment c1 = model.createCompartment( "c1" ); > c1.setMetaId( "compartmentmetaid" ); > c1.setConstant( true ); > History h = c1.getHistory( ); > Creator cCreator = new Creator( ); > cCreator.setGivenName( "Steve" ); > cCreator.setFamilyName( "Jobs" ); > h.addCreator( cCreator ); > h.setCreatedDate( new java.util.Date( ) ); > c1.setHistory( h ); > > // Save to file. > File saveF = new File( "E:\\testsbml\\testcompartmenthistory.sbml" ); > SBMLWriter writer = new SBMLWriter( ); > FileOutputStream stream = new FileOutputStream( saveF ); > writer.write( document, stream ); > } > catch( Exception ex ) > { > ex.printStackTrace( ); > } > System.out.println( "Done" ); > } > } > > <?xml version='1.0' encoding='UTF-8' standalone='no'?> > > > > > > > > > > Jobs</vCard:Family> > Steve</vCard:Given> > </vCard:N> > </li> > </Bag> > </dc:creator> > > 2012-03-23T21:25:07Z</dcterms:W3CDTF> > </dcterms:created> > > 2012-03-23T21:25:07Z</dcterms:W3CDTF> > </dcterms:modified> > </rdf:Description> > </rdf:RDF> > </annotation> > > > > > > > > > > Jobs</vCard:Family> > Steve</vCard:Given> > </vCard:N> > </li> > </Bag> > </dc:creator> > > 2012-03-23T21:25:07Z</dcterms:W3CDTF> > </dcterms:created> > </rdf:Description> > </rdf:RDF> > </annotation> > </compartment> > </listOfCompartments> > </model> > </sbml> > > > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: SBML namespace in math elements

rodrigue — 2012-03-28T15:33:56-00:00

On 28/03/12 14:37, rbyrnes wrote: > The version of JSBML that I am testing has a problem with the SBML namespace in math elements. Here is a test program, along with a test file containing a math element, and the output of the program. > > JSBML reads the math element, but it doesn't keep the SBML namespace, at least at the level of the math element. Coding would be easier if was kept. Is there a workaround for this? In your example file, the sbml namespace is mispelled (there is a typo). I think this might make the sbml:units being ignored but I did not try to run the program yet. In general, if they are units define inside an ASTNode, the SBMLWriter would add automatically the sbml namespace on the math element. Cheers, Nico > > > import java.io.*; > import java.util.*; > import javax.xml.stream.XMLStreamException; > import org.sbml.jsbml.*; > > public class ConstraintMathMLTest > { > public static void main( String[] args ) > { > try > { > File f = null; > SBMLDocument document = null; > try > { > f = new File( "E:\\testsbml\\testConstraintMathML.sbml"); > document = SBMLReader.read( f); > Model m = document.getModel( ); > ListOf cList = m.getListOfConstraints( ); > Iterator cIt = cList.iterator( ); > while ( cIt.hasNext( ) ) > { > Constraint c = (Constraint) cIt.next( ); > System.out.println( c.getMathMLString( ) ); > } > } > catch( java.util.EmptyStackException e ) > { > throw new IOException( e.getMessage( ) ); > } > catch( XMLStreamException x ) > { > throw new IOException( x.getMessage( ) ); > } > } > catch( Exception ex ) > { > ex.printStackTrace( ); > } > System.out.println( "Done" ); > } > } > > <?xml version='1.0' encoding='UTF-8' standalone='no'?> > > > > > > > 2012-03-22T18:21:20Z</dcterms:W3CDTF> > </dcterms:created> > > 2012-03-22T18:24:00Z</dcterms:W3CDTF> > </dcterms:modified> > </rdf:Description> > </rdf:RDF> > </annotation> > > > > > </rdf:Description> > </rdf:RDF> > </annotation> > > > > > </rdf:Description> > </rdf:RDF> > </annotation> > > > > > > 1</cn> > S1</ci> > </apply> > > > S1</ci> > 100</cn> > </apply> > </apply> > </math> > </constraint> > </listOfConstraints> > </model> > </sbml> > > Here is the output of the program: > > <?xml version='1.0' encoding='UTF-8'?> > > > > > > 1</cn> > S1</ci> > </apply> > > > S1</ci> > 100</cn> > </apply> > </apply> > </math> > Done > ____________________________________________________________ > To manage your jsbml-development list subscription, visit > <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> > > For a web interface to the jsbml-development mailing list, visit > <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> > > For questions or feedback about the jsbml-development list, > contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a> ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

SBML namespace in math elements

rbyrnes — 2012-03-27T20:53:32-00:00

The version of JSBML that I am testing has a problem with the SBML namespace in math elements. Here is a test program, along with a test file containing a math element, and the output of the program. JSBML reads the math element, but it doesn't keep the SBML namespace, at least at the level of the math element. Coding would be easier if was kept. Is there a workaround for this? --JavaBob import java.io.*; import java.util.*; import javax.xml.stream.XMLStreamException; import org.sbml.jsbml.*; public class ConstraintMathMLTest { public static void main( String[] args ) { try { File f = null; SBMLDocument document = null; try { f = new File( "E:\\testsbml\\testConstraintMathML.sbml"); document = SBMLReader.read( f); Model m = document.getModel( ); ListOf cList = m.getListOfConstraints( ); Iterator cIt = cList.iterator( ); while ( cIt.hasNext( ) ) { Constraint c = (Constraint) cIt.next( ); System.out.println( c.getMathMLString( ) ); } } catch( java.util.EmptyStackException e ) { throw new IOException( e.getMessage( ) ); } catch( XMLStreamException x ) { throw new IOException( x.getMessage( ) ); } } catch( Exception ex ) { ex.printStackTrace( ); } System.out.println( "Done" ); } } <?xml version='1.0' encoding='UTF-8' standalone='no'?> 2012-03-22T18:21:20Z</dcterms:W3CDTF> </dcterms:created> 2012-03-22T18:24:00Z</dcterms:W3CDTF> </dcterms:modified> </rdf:Description> </rdf:RDF> </annotation> </rdf:Description> </rdf:RDF> </annotation> </rdf:Description> </rdf:RDF> </annotation> 1 </cn> S1 </ci> </apply> S1 </ci> 100 </cn> </apply> </apply> </math> </constraint> </listOfConstraints> </model> </sbml> Here is the output of the program: <?xml version='1.0' encoding='UTF-8'?> 1 </cn> S1 </ci> </apply> S1 </ci> 100 </cn> </apply> </apply> </math> Done

More issues with history elements

rbyrnes — 2012-03-23T21:39:19-00:00

I have a test program that constructs a history for a model and a compartment. Here is the program and its output. The validator at www.sbml.org reports that the output is valid SBML, but there is a problem with the model history. I found that the warning could be avoided by hand editing all the Bag and li elements to rdf:Bag and rdf:li. Also, I needed to add a modification date element to the history for the compartment to avoid a second warning. Are there easier ways to avoid these issues? --JavaBob import java.io.*; import org.sbml.jsbml.*; public class TestCompartmentHistoryGeneration { public static void main( String[] args ) { try { // Construct a model. SBMLDocument document = new SBMLDocument( 3, 1 ); Model model = document.createModel( "modelid" ); model.setMetaId( "modelmetaid" ); // Add to the model history. History mh = model.getHistory( ); Creator creator = new Creator( ); creator.setGivenName( "Steve" ); creator.setFamilyName( "Jobs" ); mh.addCreator( creator ); mh.setCreatedDate( new java.util.Date( ) ); // Create a compartment and add to the compartment history. Compartment c1 = model.createCompartment( "c1" ); c1.setMetaId( "compartmentmetaid" ); c1.setConstant( true ); History h = c1.getHistory( ); Creator cCreator = new Creator( ); cCreator.setGivenName( "Steve" ); cCreator.setFamilyName( "Jobs" ); h.addCreator( cCreator ); h.setCreatedDate( new java.util.Date( ) ); c1.setHistory( h ); // Save to file. File saveF = new File( "E:\\testsbml\\testcompartmenthistory.sbml" ); SBMLWriter writer = new SBMLWriter( ); FileOutputStream stream = new FileOutputStream( saveF ); writer.write( document, stream ); } catch( Exception ex ) { ex.printStackTrace( ); } System.out.println( "Done" ); } } <?xml version='1.0' encoding='UTF-8' standalone='no'?> Jobs</vCard:Family> Steve</vCard:Given> </vCard:N> </li> </Bag> </dc:creator> 2012-03-23T21:25:07Z</dcterms:W3CDTF> </dcterms:created> 2012-03-23T21:25:07Z</dcterms:W3CDTF> </dcterms:modified> </rdf:Description> </rdf:RDF> </annotation> Jobs</vCard:Family> Steve</vCard:Given> </vCard:N> </li> </Bag> </dc:creator> 2012-03-23T21:25:07Z</dcterms:W3CDTF> </dcterms:created> </rdf:Description> </rdf:RDF> </annotation> </compartment> </listOfCompartments> </model> </sbml>

Re: Problem reading history element?

rodrigue — 2012-03-22T15:09:45-00:00

The problem is coming from the test file, many namespaces are forgotten. The following file is working fine : <?xml version='1.0' encoding='UTF-8' standalone='no'?> 2012-03-19T17:17:33Z</dcterms:W3CDTF> </dcterms:created> 2012-03-19T17:17:33Z</dcterms:W3CDTF> </dcterms:modified> </rdf:Description> </rdf:RDF> </annotation> aa</vCard:Family> bb</vCard:Given> </vCard:N> <a href="mailto:seymour@ssss.edu" target="_blank">seymour@ssss.edu</a></vCard:EMAIL> ssss</vCard:Orgname> </vCard:ORG> </rdf:li> </rdf:Bag> </dc:creator> </rdf:Description> </rdf:RDF> </annotation> </rdf:Description> </rdf:RDF> </annotation> </compartment> </listOfCompartments> </model> </sbml> Cheers, Nico On 20/03/12 07:27, Andreas Dräger wrote: > Am 3/20/12 5:45 AM, schrieb rbyrnes: >> Andreas- >> >> I am seeing another problem, while reading histories of non-model elements. Here is a test file, a sample program, and output. I am using a jar file that I compiled from the source code as "jsbml-1.0-a1-with-dependencies.jar". >> >> --JavaBob >> >> <?xml version='1.0' encoding='UTF-8' standalone='no'?> >> >> >> >> >> >> >> 2012-03-19T17:17:33Z</W3CDTF> >> </created> >> >> 2012-03-19T17:17:33Z</W3CDTF> >> </modified> >> </rdf:Description> >> </rdf:RDF> >> </annotation> >> >> >> >> >> >> >> >> >> aa</vCard:Family> >> bb</vCard:Given> >> </vCard:N> >> <a href="mailto:seymour@ssss.edu" target="_blank">seymour@ssss.edu</a></vCard:EMAIL> >> >> ssss</vCard:Orgname> >> </vCard:ORG> >> </li> >> </Bag> >> </dc:creator> >> </rdf:Description> >> </rdf:RDF> >> </annotation> >> >> >> >> >> </rdf:Description> >> </rdf:RDF> >> </annotation> >> </compartment> >> </listOfCompartments> >> </model> >> </sbml> >> >> import java.io.*; >> import javax.xml.stream.XMLStreamException; >> import org.sbml.jsbml.*; >> >> public class TestCompartmentHistory >> { >> public static void main( String[] args ) >> { >> try >> { >> File f = null; >> SBMLDocument document = null; >> try >> { >> f = new File( "E:\\testsbml\\testcompartmenthistory.sbml" ); >> document = SBMLReader.read( f ); >> Model m = document.getModel( ); >> History c1History = m.getListOfCompartments( ).get( 0 ).getHistory( ); >> java.util.List list = (java.util.List) c1History.getListOfCreators( ); >> System.out.println( "Number of creators: " + list.size( ) ); >> } >> catch( java.util.EmptyStackException e ) >> { >> throw new IOException( e.getMessage( ) ); >> } >> catch( XMLStreamException x ) >> { >> throw new IOException( x.getMessage( ) ); >> } >> } >> catch( Exception ex ) >> { >> ex.printStackTrace( ); >> } >> System.out.println( "Done" ); >> } >> } >> >> WARN (VCardParser.java:250) - Lost Information : the element 'N' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:131) - Lost Information : the characters ' >> ' on the element 'N' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:250) - Lost Information : the element 'Family' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:131) - Lost Information : the characters 'aa' on the element 'Family' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:250) - Lost Information : the element 'Given' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:131) - Lost Information : the characters 'bb' on the element 'Given' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:250) - Lost Information : the element 'EMAIL' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:131) - Lost Information : the characters <a href="mailto:'seymour@ssss.edu" target="_blank">'seymour@ssss.edu</a>' on the element 'EMAIL' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:250) - Lost Information : the element 'ORG' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:131) - Lost Information : the characters ' >> ' on the element 'ORG' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:250) - Lost Information : the element 'Orgname' might be lost as the context object is not a Creator. >> WARN (VCardParser.java:131) - Lost Information : the characters 'ssss' on the element 'Orgname' might be lost as the context object is not a Creator. >> Number of creators: 0 >> Done >> >> > Hi Robert, > > You are right. It seems there is a bug. I modified your program a little > bit and tried to directly read a newly created SBMLDocument again. > Writing works, but reading fails at the moment. It seems there is > something wrong with the history parsing: > > import java.io.ByteArrayOutputStream; > import java.io.IOException; > > import javax.xml.stream.XMLStreamException; > > import org.sbml.jsbml.Creator; > import org.sbml.jsbml.History; > import org.sbml.jsbml.Model; > import org.sbml.jsbml.SBMLDocument; > import org.sbml.jsbml.SBMLReader; > import org.sbml.jsbml.SBMLWriter; > > public class TestCompartmentHistory { > > public static void main(String[] args) { > try { > SBMLDocument document = new SBMLDocument(3, 1); > Creator creator = new Creator("Andreas", "Draeger", "University > of Tuebingen", "<a href="mailto:andreas.draeger@uni-tuebingen.de" target="_blank">andreas.draeger@uni-tuebingen.de</a>"); > document.getHistory().addCreator(creator); > document.createModel("m").getHistory().addCreator(creator.clone()); > ByteArrayOutputStream stream = new ByteArrayOutputStream(); > SBMLWriter.write(document, stream, ' ', (short) 2); > stream.flush(); > System.out.println(stream.toString()); > try { > document = SBMLReader.read(stream.toString()); > Model m = document.getModel(); > History c1History = m.getListOfCompartments().get(0).getHistory(); > java.util.List list = (java.util.List) > c1History.getListOfCreators(); > System.out.println("Number of creators: " + list.size()); > } catch (java.util.EmptyStackException e) { > throw new IOException(e.getMessage()); > } catch (XMLStreamException x) { > throw new IOException(x.getMessage()); > } > } catch (Exception ex) { > ex.printStackTrace(); > } > System.out.println("Done"); > } > > } > > I suggest to wait for an answer by Nicolas Rodriguez, who implemented > the parsing methods together with Marine Dumousseau. > > Cheers > Andreas > ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Units inheritance in Level 3

Andreas Dräger — 2012-03-21T07:13:35-00:00

Am 3/20/12 12:56 PM, schrieb Andreas Dräger: > Hi Sarah, > >>> Yes. That's what I would like to do. It is actually very easy using a >>> filter for listOf that accepts the first identical >>> UnitDefinition (using a call like >>> UnitDefinition.areIdentical(myJustDerivedUnit, unitNumberXXfromTheList), >>> with XX being the counter in the for-each loop). >> My main thought about this is to consider what the user might want to >> do with the unitDefinition returned from getDerivedUnits(). They may >> wish to change or use it somewhere else and may not actually want it >> to be part of the original model. >> > In deed, there are two different ways to go. > > 1. The user always receives a newly created UnitDefinition and can > manipulate it in any way, but if it is necessary to ensure that the unit > is already part of the model, the user must add it by him/herself. > Similarly, the user would have to filter the Model's list of units for > an identical unit by him/herself. > > 2. The user gets a unit that may already belong to the model. A check > can be done by asking myModel.getListOfUnits().contains(myUnit). If it > is already in the model and the user wants to do some change, a call to > the clone() method might be necessary. > > It now depends on what the users would prefer to have. Maybe you're > right, and way #2 is less streightforward because the user does not know > if the unit belongs to the model or not. In idea #1 the returned unit > never belongs to the model. > > Are there any other opinions on that topic? > Hi everybody, Actually, when calling getDerivedUnitsInstance() on an element such as Species or Compartment, the user can never know if the returned UnitDefinition object is part of the model or a newly created element. If, for instance, there is a unit defined for the species, the method looks the corresponding UnitDefinition up in the model and returns it. If it is a base unit, a new UnitDefinition will be created that only contains the base unit in its list of units. Similarly, if the species inherits a unit from the model, also the model's inherited unit definition object will be returned. So in some cases the user receives a UnitDefinition that is not (yet) part of the Model, and in some cases the returned value is a UnitDefintion from the Model's list of units. Therfore, it seems more logical to preferably return a UnitDefinition from the Model rather than a completely new one. The approach to search for identical UnitDefinition objects in the model after creating some UnitDefintion (for a base unit or something different) seems to be the better choice, I think. In any way, the method we use to derive units is not straightforward at the moment. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Units inheritance in Level 3

Andreas Dräger — 2012-03-20T11:56:52-00:00

Hi Sarah, >> Yes. That's what I would like to do. It is actually very easy using a >> filter for listOf that accepts the first identical >> UnitDefinition (using a call like >> UnitDefinition.areIdentical(myJustDerivedUnit, unitNumberXXfromTheList), >> with XX being the counter in the for-each loop). > > My main thought about this is to consider what the user might want to > do with the unitDefinition returned from getDerivedUnits(). They may > wish to change or use it somewhere else and may not actually want it > to be part of the original model. > In deed, there are two different ways to go. 1. The user always receives a newly created UnitDefinition and can manipulate it in any way, but if it is necessary to ensure that the unit is already part of the model, the user must add it by him/herself. Similarly, the user would have to filter the Model's list of units for an identical unit by him/herself. 2. The user gets a unit that may already belong to the model. A check can be done by asking myModel.getListOfUnits().contains(myUnit). If it is already in the model and the user wants to do some change, a call to the clone() method might be necessary. It now depends on what the users would prefer to have. Maybe you're right, and way #2 is less streightforward because the user does not know if the unit belongs to the model or not. In idea #1 the returned unit never belongs to the model. Are there any other opinions on that topic? Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Units inheritance in Level 3

Sarah Keating — 2012-03-20T10:16:23-00:00

Hi Andreas > Yes. That's what I would like to do. It is actually very easy using a > filter for listOf that accepts the first identical > UnitDefinition (using a call like > UnitDefinition.areIdentical(myJustDerivedUnit, unitNumberXXfromTheList), > with XX being the counter in the for-each loop). My main thought about this is to consider what the user might want to do with the unitDefinition returned from getDerivedUnits(). They may wish to change or use it somewhere else and may not actually want it to be part of the original model. Sarah ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Units inheritance in Level 3

Andreas Dräger — 2012-03-20T10:00:29-00:00

Am 3/20/12 10:50 AM, schrieb Sarah Keating: > I know we talked about this, but I thought it would be good to clarify > on the list. > > You are right, you can declare 'default' units using the attributes on > the Model element. So if a Species or Compartment does not explicitly > declare units it can inherit them from the model. But these 'default' > units must be fully declared within the model. Deriving the units for > compartments and species should take this into account. > > Note that this also applies to time. A model that does not have a > value for the timeUnits attribute does not have any default time unit. Hi Sarah, Yes, that's how I understood it. My question therefore was whether the, e.g., timeUnits (or other units attribute of the model) must be defined in the listOfUnits within the model in addition to have and you explained that there are two different options: 1. The timeUnits (or substance,...) might be a base unit, such as second. In this case no additional entry in the listOfUnits for the timeUnits is necessary. 2. The timeUnits can be a more complicated unit, e.g., hour = 3600 * second. Then there must be an element in the listOfUnits in addition to the field. That's how I understood the specification and your explanation. I already changed the implementation for getDerivedUnits() in Species and Compartment to be able to fall back to the default unit from the Model if there is one. If there is no own reference to a unitDefinition in Compartment or Species and also such fully specified unit in the model, null will be returned when deriving a unit for these element. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Units inheritance in Level 3

Andreas Dräger — 2012-03-20T09:55:21-00:00

Am 3/20/12 10:50 AM, schrieb Sarah Keating: > Hi Andreas > >> compartment and fix this problem for Level 3 models. Furthermore, when >> deriving units, I would also like to have a mechanism that goes through >> all units in the model's list of units and identifies the first unit >> that is identically defined as the derived unit. In this way, the >> derived unit would preferably be part of the SBML model, only in some >> cases it would not be part of the model. Currently, it might happen that >> the getDerivedUnits() method returns some unit definition that is >> equally defined as some other unit definition within the model, but the >> returned element is not part of the model. > > Can I clarify if I understand this correctly ? > > The function getDerivedUnits() returns a unitDefinition object (as > defined by SBML). It is possible that the Model already contains a > unitDefinition object which is identical to the returned > unitDefinition. If this is the case you want to return the actual > instance of the unitDefinition from the model rather than create a new > unitDefinition object that is identical to one that already exists. > > Is that correct ? Hi Sarah, Yes. That's what I would like to do. It is actually very easy using a filter for listOf that accepts the first identical UnitDefinition (using a call like UnitDefinition.areIdentical(myJustDerivedUnit, unitNumberXXfromTheList), with XX being the counter in the for-each loop). Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Units inheritance in Level 3

Sarah Keating — 2012-03-20T09:50:40-00:00

Hi Andreas > compartment and fix this problem for Level 3 models. Furthermore, when > deriving units, I would also like to have a mechanism that goes through > all units in the model's list of units and identifies the first unit > that is identically defined as the derived unit. In this way, the > derived unit would preferably be part of the SBML model, only in some > cases it would not be part of the model. Currently, it might happen that > the getDerivedUnits() method returns some unit definition that is > equally defined as some other unit definition within the model, but the > returned element is not part of the model. Can I clarify if I understand this correctly ? The function getDerivedUnits() returns a unitDefinition object (as defined by SBML). It is possible that the Model already contains a unitDefinition object which is identical to the returned unitDefinition. If this is the case you want to return the actual instance of the unitDefinition from the model rather than create a new unitDefinition object that is identical to one that already exists. Is that correct ? Sarah ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Units inheritance in Level 3

Sarah Keating — 2012-03-20T09:50:06-00:00

Hi Andreas I know we talked about this, but I thought it would be good to clarify on the list. You are right, you can declare 'default' units using the attributes on the Model element. So if a Species or Compartment does not explicitly declare units it can inherit them from the model. But these 'default' units must be fully declared within the model. Deriving the units for compartments and species should take this into account. Note that this also applies to time. A model that does not have a value for the timeUnits attribute does not have any default time unit. Sarah ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>

Re: Problem reading history element?

Andreas Dräger — 2012-03-20T07:27:47-00:00

Am 3/20/12 5:45 AM, schrieb rbyrnes: > Andreas- > > I am seeing another problem, while reading histories of non-model elements. Here is a test file, a sample program, and output. I am using a jar file that I compiled from the source code as "jsbml-1.0-a1-with-dependencies.jar". > > --JavaBob > > <?xml version='1.0' encoding='UTF-8' standalone='no'?> > > > > > > > 2012-03-19T17:17:33Z</W3CDTF> > </created> > > 2012-03-19T17:17:33Z</W3CDTF> > </modified> > </rdf:Description> > </rdf:RDF> > </annotation> > > > > > > > > > aa</vCard:Family> > bb</vCard:Given> > </vCard:N> > <a href="mailto:seymour@ssss.edu" target="_blank">seymour@ssss.edu</a></vCard:EMAIL> > > ssss</vCard:Orgname> > </vCard:ORG> > </li> > </Bag> > </dc:creator> > </rdf:Description> > </rdf:RDF> > </annotation> > > > > > </rdf:Description> > </rdf:RDF> > </annotation> > </compartment> > </listOfCompartments> > </model> > </sbml> > > import java.io.*; > import javax.xml.stream.XMLStreamException; > import org.sbml.jsbml.*; > > public class TestCompartmentHistory > { > public static void main( String[] args ) > { > try > { > File f = null; > SBMLDocument document = null; > try > { > f = new File( "E:\\testsbml\\testcompartmenthistory.sbml" ); > document = SBMLReader.read( f ); > Model m = document.getModel( ); > History c1History = m.getListOfCompartments( ).get( 0 ).getHistory( ); > java.util.List list = (java.util.List) c1History.getListOfCreators( ); > System.out.println( "Number of creators: " + list.size( ) ); > } > catch( java.util.EmptyStackException e ) > { > throw new IOException( e.getMessage( ) ); > } > catch( XMLStreamException x ) > { > throw new IOException( x.getMessage( ) ); > } > } > catch( Exception ex ) > { > ex.printStackTrace( ); > } > System.out.println( "Done" ); > } > } > > WARN (VCardParser.java:250) - Lost Information : the element 'N' might be lost as the context object is not a Creator. > WARN (VCardParser.java:131) - Lost Information : the characters ' > ' on the element 'N' might be lost as the context object is not a Creator. > WARN (VCardParser.java:250) - Lost Information : the element 'Family' might be lost as the context object is not a Creator. > WARN (VCardParser.java:131) - Lost Information : the characters 'aa' on the element 'Family' might be lost as the context object is not a Creator. > WARN (VCardParser.java:250) - Lost Information : the element 'Given' might be lost as the context object is not a Creator. > WARN (VCardParser.java:131) - Lost Information : the characters 'bb' on the element 'Given' might be lost as the context object is not a Creator. > WARN (VCardParser.java:250) - Lost Information : the element 'EMAIL' might be lost as the context object is not a Creator. > WARN (VCardParser.java:131) - Lost Information : the characters <a href="mailto:'seymour@ssss.edu" target="_blank">'seymour@ssss.edu</a>' on the element 'EMAIL' might be lost as the context object is not a Creator. > WARN (VCardParser.java:250) - Lost Information : the element 'ORG' might be lost as the context object is not a Creator. > WARN (VCardParser.java:131) - Lost Information : the characters ' > ' on the element 'ORG' might be lost as the context object is not a Creator. > WARN (VCardParser.java:250) - Lost Information : the element 'Orgname' might be lost as the context object is not a Creator. > WARN (VCardParser.java:131) - Lost Information : the characters 'ssss' on the element 'Orgname' might be lost as the context object is not a Creator. > Number of creators: 0 > Done > > Hi Robert, You are right. It seems there is a bug. I modified your program a little bit and tried to directly read a newly created SBMLDocument again. Writing works, but reading fails at the moment. It seems there is something wrong with the history parsing: import java.io.ByteArrayOutputStream; import java.io.IOException; import javax.xml.stream.XMLStreamException; import org.sbml.jsbml.Creator; import org.sbml.jsbml.History; import org.sbml.jsbml.Model; import org.sbml.jsbml.SBMLDocument; import org.sbml.jsbml.SBMLReader; import org.sbml.jsbml.SBMLWriter; public class TestCompartmentHistory { public static void main(String[] args) { try { SBMLDocument document = new SBMLDocument(3, 1); Creator creator = new Creator("Andreas", "Draeger", "University of Tuebingen", "<a href="mailto:andreas.draeger@uni-tuebingen.de" target="_blank">andreas.draeger@uni-tuebingen.de</a>"); document.getHistory().addCreator(creator); document.createModel("m").getHistory().addCreator(creator.clone()); ByteArrayOutputStream stream = new ByteArrayOutputStream(); SBMLWriter.write(document, stream, ' ', (short) 2); stream.flush(); System.out.println(stream.toString()); try { document = SBMLReader.read(stream.toString()); Model m = document.getModel(); History c1History = m.getListOfCompartments().get(0).getHistory(); java.util.List list = (java.util.List) c1History.getListOfCreators(); System.out.println("Number of creators: " + list.size()); } catch (java.util.EmptyStackException e) { throw new IOException(e.getMessage()); } catch (XMLStreamException x) { throw new IOException(x.getMessage()); } } catch (Exception ex) { ex.printStackTrace(); } System.out.println("Done"); } } I suggest to wait for an answer by Nicolas Rodriguez, who implemented the parsing methods together with Marine Dumousseau. Cheers Andreas -- Dr. Andreas Dräger University of Tuebingen Center for Bioinformatics Tuebingen (ZBIT) Sand 1 72076 Tübingen Germany Phone: +49-7071-29-78982 Fax: +49-7071-29-5091 ____________________________________________________________ To manage your jsbml-development list subscription, visit <a href="https://utils.its.caltech.edu/mailman/listinfo/jsbml-development" target="_blank">https://utils.its.caltech.edu/mailman/listinfo/jsbml-development</a> For a web interface to the jsbml-development mailing list, visit <a href="http://sbml.org/Forums/" target="_blank">http://sbml.org/Forums/</a> For questions or feedback about the jsbml-development list, contact <a href="mailto:sbml-team@caltech.edu" target="_blank">sbml-team@caltech.edu</a>