{5} Accepted, Active Tickets by Owner (Full Description) (110 matches)

This ticket is meant to collect and prioritize the improvements that need to be done to the OMEdit GUI in order to make it usable with the MSL and with most modern Modelica 3.4 compliant libraries. Each item is covered by a specific ticket.

Features requiring the new front end to be in place

All the features listed here depend on the front-end to provide correct information to OMEdit, and to do so rapidly. This is now possible by using the new front-end (see #4138), which is exploited by new API functions which are much faster than previously. Version 1.14.0 will use these features by default.

• #2079 Replaceable classes and replaceable objects (e.g. the Medium package in all Modelica.Fluid models, or the HeatTransfer model in the Modelica.Fluid.Vessels.ClosedVolume) should show up in the parameter window, with a drop-down menu filled in based on the choices or choicesAllMatching annotations. Once a specific choice is made, it should be possible to change the parameters and/or replaceable classes/models of the redeclared model/component, e.g. by clicking on a button. This feature is essential for the Modelica.Fluid library, as well as for many user libraries such as ThermoPower.

• #2081 Conditional connectors should be displayed correctly. This feature is used extensively in the Modelica.Mechanics and Modelica.Electrical libraries. It requires that the API functions return the correct visibility attribute.

• #2661 Update parameter-dependent Dialog annotations correctly

• #2132 Display conditional connectors correctly by evaluating the corresponding parameters

• #2891 Hierarchical editing of systems for graphical modification of parameters and replaceable classes in sub-models. Requires substantial work to implement some API function, as well as in OMEdit to display the information properly.

• #3915 Rename feature missing

Other features

• #6417 Keyboard shortcuts for frequent actions such as simulate, switch view
• #5529 Insufficient precision when storing sine phases

• #4449 Allow several curves in the same parametric plot window

• #4645 and #4839 Use blank pixels around lines to enhance readability

• #3776 Allow copying/pasting models across diagrams

• #3287 Allow input of common parameters and replaceable classes in a multiple component selection. This can shorten the time to set up a model dramatically

• #2166 See exact simulation values through snap to curve feature

Past improvements

• #4508 Reduce the installation time under Windows by making the installation of contributed open source libraries optional

• #2909, #3788 Drastically reduce the waste of HD space by eliminating useless intermediate output files and reducing the number of useful files

• #3054 Clarify "Revert from previous" button

• #2908 Changes in sub-models should be reflected in models using them immediately, not only when changes to the latter models are made.

• #2906 Drop-down choices for enumeration (and possibly boolean) parameters

• #2845 Consistent view between textual view and library browser view of packages should be always guaranteed. Requires to update the views in OMEdit whenever the focus is changed between the different sub-windows.

• #2391 Improved handling of custom modifers (e.g. start attributes) from the GUI. Modifers, once applied, should still show up when the parameter window is re-opened later.

• #2390 Fully support the showStartAttribute annotation, which is widely used in Modelica.Mechanics.

• #2190 Copy class feature within the same package.

• #2903 Add option to output protected variables to simulation setup window, output tag.

• #2477 Better snap-to-grid features when using blocks from the MSL

• #2506 Keep connecting lines manhattanized and allow no-nonsense editing of models copied from the MSL (e.g. it should be possible to move one component without weird side-effects on the diagram)

• #2611 Zoom issue should be fixed.

• #2988 Zoom In/Out with MouseWheel.

• #2392 Show default values differently in the Component Parameters window.

• #2956 Moving a complete model disassembles everything

• #2596 Re-simulate with changed options. Currently if we want to make a new simulation with different simulations options, e.g. a different final time we have to re-build the whole executable, that is a waste of time.

• #2985 Updating submodel icons where they are used. This ticket is not really urgent. But it is related to #2908, that has already been fixed, so fixing this one is sort of completion of the job. In case it is difficult to fix it could be postponed.

• #1900 Handle connections to expandable connectors in OMEdit

• #2250 Support DisplayUnit and unit conversion for parameter input in OMEdit

• #3558: Flip icons with text boxes correctly

• #2676 Handle models in separate files

• #2905 Comment- and formatting-preserving parsing and unparsing is required to avoid that OMEdit scrambles up the code of existing Modelica classes when saving them on file.

• #2892 Undo/Redo feature in the Diagram view of the model editor.

• #2190 Copy class to a different package.

• #2696: Display enhancement of very large and very small numbers on vertical axis

• #2960 Speed up the exploration of the package tree and the drag-n-drop actions.

• #4504 Allow to install OpenModelica anywhere in the file system, including the default installation directories (such as C:\Program Files) that include spaces

• #2395 - Support the connectorSizing annotation. Required by many libraries including Modelica.Fluid, PNLib, iPSL, OpenIPSL

Steps to reproduce: 1) load TestPackage.mo, and open Test: we have a red box because MyModel is missing 2) load Test1.mo and duplicate its model MyModel into TestPackage. Look at model Test inside TestPackage: the red box should disappear, but it does not. 3) save, unload and re-load TestPackage again 4) open Test inside TextPackage: the red box has disappared and now myModel is displayed.

So the issue is in point 2.

Tested with 1.17.0-dev.beta3.

We now have some basic support for MetaModelica in OMEdit but there are still several things missing. Following are the list of features,

1. Code folding.
2. Auto complete.
3. Single & multi-line commenting.
4. Pressing enter should open the selected file in ModelWidget.
5. Shift-Home should not select trailing spaces.
6. Show only name in ctrl-tab window.
7. Find in files.
8. Function Outline.
9. Toggle comment selection for C/C++ Editor.
10. Code folding for C/C++ Editor.

Even if it seems to be a more time-consuming task, I recommend merging the UNDO stack of the Text View with the stack of the Diagram View as discussed in #4334#comment:13.

It is just a strange behavior that you must go the to correct view to undo things. People how use OMEdit assume that the content in Text View and Diagram View is a different representation of the SAME thing and expect a coherent behavior especially for undo/redo.

It would be handy to make ​filterSimulationResults available from OMEdit to export and filter partial result files.

Please consider the following models (also attached):

package P1
  model M1
    extends Modelica.Blocks.Interfaces.SISO;
    Boolean Flag = false;
    Real x = if Flag then u else 0;
  equation
    y = 0.5*x;
  end M1;

  model M2
    extends Modelica.Blocks.Interfaces.SISO;
    replaceable model Mx = P1.M1;
    Mx mx annotation(Placement(visible = true, transformation(origin = {-2, 0}, extent = {{-10, -10}, {10, 10}}, rotation = 0)));
  equation
  connect(u, mx.u) annotation(Line(points = {{-120, 0}, {-14, 0}}, color = {0, 0, 127}));
  connect(mx.y, y) annotation(Line(points = {{9, 0}, {110, 0}}, color = {0, 0, 127}));
  end M2;

  model M3
    P1.M2 m2(redeclare model Mx=P1.M1(Flag=true)) annotation(Placement(visible = true, transformation(origin = {-2, 0}, extent = {{-10, -10}, {10, 10}}, rotation = 0)));
    Modelica.Blocks.Sources.Step step1(startTime = 0.5)  annotation(Placement(visible = true, transformation(origin = {-36, 0}, extent = {{-10, -10}, {10, 10}}, rotation = 0)));
  equation
    connect(step1.y, m2.u) annotation(Line(points = {{-24, 0}, {-14, 0}}, color = {0, 0, 127}));
  end M3;
end P1;

The model M2 defines the model Mx as replaceable and the model M3 uses M2 redeclaring the model Mx in order to change the Flag parameter value from false to true.

Simulating the model M3 the results is according I expect, the Flag parameter value is changed (from false to true) and the step is propagated to the model m2 output.

Now open the model M2 and simply move the model mx

the line:

    Mx mx annotation(Placement(visible = true, transformation(origin = {-2, 0}, extent = {{-10, -10}, {10, 10}}, rotation = 0)));

is changed by OmEdit in:

P1.M2.Mx mx annotation(Placement(visible = true, transformation(origin = {-2, 0}, extent = {{-10, -10}, {10, 10}}, rotation = 0)));

and the model Mx now seems to be no more replaceable, in fact, re-simulating the model M3 the Flag parameter value now is not propagated and the output y of model m2 stays fixed to 0.

Attached are: package P1, output before M1 moving, output after M1 moving.

OMEdit 1.12.0~dev-250-g0f5c07d Connected to OpenModelica 1.12.0~dev-561-gc489db1 sysop: Ubunti 14.04 - 34 bit

Please check the attached simple model, I just dragged a pressure source component boundary in it.

When I double-click on boundary, and then click on the Medium drop-down menu, I get about 30 possible classes for redeclarations. However, there are several issues there

• some are partial packages, which obviously cannot be used, e.g., Modelica.Media.Interfaces.TemplateMedium
• some are found within models but are not encapsulated, so according to ​Section 5.3.2 of the specification, they shouldn't even be looked up, e.g. Modelica.Media.Examples.Tests.Components.ShortPipe.Medium
• some are even found inside partial models, e.g. Modelica.Media.Examples.Tests.Components.PartialTestModel.Medium

On the other hand, not even one of the actually usable models, e.g. Modelica.Media.Air.SimpleAir or Modelica.Media.Water.StandardWater actually show up.

This issue makes OMEdit actually unusable with Modelica Media/Fluid

Hi there,

I am using the XML file generated by Modelica for modeling. I am getting the following error in the format each time-

In bouncing ball, line 162 of generated xml file, <eq: Equation> is generated within the <when>, whereas it should be <equ: Equation> as per the schema.

Please let me know how/ when this can be resolved. I have attached the modelica model and the generated xml file.

model emptyVector2
  record R
    parameter Real x[:]=zeros(size(x,1));
  end R;
  R r;
end emptyVector2;

here omc.exe freezes. (r15030)

In the following model (also attached) the inner variable v declared in model Test is assigned to in model B where it is declared as outer output. B is instantiated in A and A is instantiated in Test. In model A variable v is declared as inner outer output.

package Bug

  model B
    outer output Real v;
  equation
    v = time;
  end B;

  model A
    inner outer output Real v;
    Real y2;
    B b;
  equation
    y2 = v;
  end A;

  model Test
    inner Real v;
    Real y1;
    A a;
  equation
    y1 = v;
  end Test;
  
end Bug;

The model flattens, but the generated system of equations is wrong (therefore, it doesn't simulate). The flattened Modelica is:

class Bug.Test
  Real v;
  Real y1;
  Real a.v;
  Real a.y2;
equation
  a.v = time;
  a.y2 = v;
  y1 = v;
end Bug.Test;

Note that the equality constraint between v and a.v is lost.

Adding an additional constraint, e.g., Real a.v = v would make the flattened code valid. Alternatively, a.v could be substituted by v everywhere.

Thanks to the solution of a couple of tickets during the last weeks the optimization basing on FMI is working quite well now!

A remaining problem is a memory leak that can be seen in all examples to some extend. Here is an extreme case that leaks about 20 MB/min (depending on the speed of your processor).

The problem can be reproduced by repeatedly running an optimization with HQP. On a machine with omc (r23977), git, gcc and tcl-dev:

$ git clone https://github.com/omuses/hqp.git
$ cd hqp
$ ./configure
$ make
$ cd odc
$ ./odc
% while true {source drumboiler_sp.tcl}

The support for uncertainties (and implicit the connection to OpenTURNS) got broken a while back. We should bring it back as people in MODRIO project need it.

We don't seem to have a clear picture on where to install 3rdParty includes. We seem to push them into one of these places:

• build/include/omc/c
• build/include/omc/cpp

sometimes in their own directory sometimes not.

We should come up with a clear structure on where to put these includes especially of those libraries that are used by both C and CPP runtimes.

For example for ticket #3426 we need sundials but is not clear where to put the includes. If ones looks into the OpenModelica windows installation we seem to have sundials includes twice inside:

• build/include/omc/cpp
• build/include/omc/cpp/includes

Sometimes if the user changes their Windows PATH the build will get to some programs it shouldn't and the build process has issues. Make sure we do some checks on the PATH in the Makefile.omdev.mingw.

Run the attached script.

Currently we put all the dependent dlls in the binary folder but the FMI standard says that it should be only one dlls which has no dependencies. It should be possible to generate this via source code FMUs.

I have a non-public wind power model that sends an array of zeros to a function instead the of the correct array. The model works with +d=evalparam which forces the evaluation in the front-end.

I'm opening this bug so we don't forget about it.

Currently we impose a hardcoded 900 seconds alarm time for tests. To be able to use the same testing framework for big tests this limit should be configurable.

The ExternData library works in Dymola and SimulationX, unfortunately, it is currently not working in OpenModelica. The library developed mentioned this in the documentation and suggest a quick fix (​https://github.com/tbeu/ExternData). However, the fix does not work in OpenModelica. (please check this discussion board ​https://github.com/tbeu/ExternData/issues/9#issuecomment-308656184). By using a stand-alone function as suggested by the library developer-

package Test

extends Modelica.Icons.Package; function OpenModelica_getReal "Get scalar Real value from XML file"

extends Modelica.Icons.Function; input String fileName="" "File where external data is stored"; input String varName "Key"; output Real y "Real value"; protected

ExternData.Types.ExternXMLFile xml = ExternData.Types.ExternXMLFile(fileName) "External XML file object";

algorithm

y := ExternData.Functions.XML.getReal(xml=xml, varName=varName);

annotation(Inline=false);

end OpenModelica_getReal;

model XMLTest "XML file read test"

extends Modelica.Icons.Example; parameter String fileName = Modelica.Utilities.Files.loadResource("modelica://ExternData/Resources/Examples/test.xml"); Real test = OpenModelica_getReal(fileName, "set1.gain.k"); annotation(experiment(StopTime=1));

end XMLTest; annotation(uses(ExternData(version="2.2.0"), Modelica(version="3.2.2")));

end Test;

XML test compiles without any error but when I try to run it gives following errors.

[1] 10:44:56 Translation Notification [ExternData: 512:7-520:18]: Created directory C:\Users\mrn04\AppData\Local\Temp\/omc_compile_ED_XMLFile_422688

[2] 10:44:56 Translation Notification [ExternData: 512:7-520:18]: cp -a "C:/OpenModelica1.12.0-dev-64bit/lib/omlibrary/ExternData 2.2.0/Resources"/* "C:\Users\mrn04\AppData\Local\Temp\/omc_compile_ED_XMLFile_422688"

[3] 10:44:56 Translation Notification [ExternData: 512:7-520:18]: Changed directory to C:/Users/mrn04/AppData/Local/Temp/omc_compile_ED_XMLFile_422688/BuildProjects/autotools

[4] 10:44:56 Scripting Error Error opening file: log: No such file or directory.

[5] 10:44:56 Translation Error [C:/dev/OpenModelica64bit/OMCompiler/Compiler/SimCode/SimCodeFunctionUtil.mo: 742:9-742:129]: Internal error function elaborateFunction failed for function: function ExternData.Functions.XML.getReal"Inline never" "Get scalar Real value from XML file" input String varName "Key"; output Real y "Real value"; input ExternalObject ExternData.Types.ExternXMLFile xml "External XML file object"; external "C" y = ED_getDoubleFromXML(xml, varName); end ExternData.Functions.XML.getReal;

[6] 10:44:56 Translation Error [C:/dev/OpenModelica64bit/OMCompiler/Compiler/SimCode/SimCodeUtil.mo: 603:5-603:83]: Internal error Creation of Modelica functions failed.

Any kind of help on this issue will be very helpful. Thanks.

For Windows release use scripts from: ​https://github.com/OpenModelica/OpenModelicaSetup instead of old SVN: ​https://openmodelica.org/svn/OpenModelica/installers/windows/OpenModelicaSetup

The memory pool frees the string parameters which it shouldn't do. We need a way to alloc and save the string parameters so that we can free them properly, but only when the FMU is terminated.

Currently there seems to be 4 separate implementations for the csv files:

1. input file externalInput.csv uses space as a delimiter
2. input file for optimization uses space as a delimiter
3. writing output results to the csv file is done using delimiter ","
4. reading results from the results file uses comma as delimiter (libcsv.c)

All these 4 different implementations should be based on libcsv.c which is the most configurable and one should be able to set the csv delimiter char and the quote char from command line.

At the moment one needs to call instantiateModel to get the flat Modelica output. A flag to dump the flat Modelica during the model checking or model build for simulation is desirable as the instantiation can take some time for large models so is not preferable to be repeated.

Previous ticket text

I used to generate flat modelica outputs and dae dumps directly form OMEdit by the following compiler flags assigned as "Additional Translation Flags" in the Options Dialog:

-m +d=evalOutputOnly --postOptModules+=dumpDAE

Currently there are no additional files created.

Maybe I'm missing something, but at least it used work.

More or less a copy of this issue: ​https://github.com/tbeu/netCDF-DataReader/issues/6 Not sure whether this has to be fixed on OM side or in the library, one of the two issues can be closed.

The netCDF library comes with three examples, all three work on Linux, but fail on Windows. Tested with Windows 10 64 bit, OM 1.11 32 bit release and OM 1.12 64 bit nightly.

For the Simple example the error is:

[3] 10:30:07 Fehler
Failed to write to file (not open)

[4] 10:30:07 Übersetzung Fehler
Template error: A template call failed (a call with 0 parameters: dladdr failed: not available on Windows). One possible reason could be that a template imported function call failed (which should not happen for functions called from within template code; templates preserve pure 'match'/non-failing semantics).

For the other two examples, the error is:

NMAKE : fatal error U1077: "cl": R�ckgabe-Code "0x2"
Stop.
Compilation process failed. Exited with code 2.

The full log is pasted here: ​http://pastebin.com/3k3e7LMB

Side note: There seems to be some encoding wrong with the word R�ckgabe

I had several issues with co-simulation FMUs to be run under Vector CANoe. Debugging them was not that easy, because I was not able to use Visual Studio, but had to resort to the gdb binary, which I found in the OpenModelica directory. Even when working with this, I first had to modify the FMU's Makefile to generate debug information. It would be helpful, if there would be a switch to generate debugging information and eventually generating mixed source and (dis)assembler listings. The debug information should also be made available as PDB files. ​https://github.com/rainers/cv2pdb might be helpful for this.

If a simulation fails, it is not possible to change the solver parameters and re-simulate, one has to recompile the model from scratch every time.

It would be nice if one could re-simulate without re-compiling also in this case.

@adeas31, would this be easy to implement?

I tried this on an old version and it used to work, but got broken at some point. I haven't tried it with -overrideFile, but it may have the same issue.

loadString("
model restartOverrideBug
  parameter Real a = 1;
  Real x(start=1);
  Real y;
equation
  y = der(x);
  der(y) = a * x;
end restartOverrideBug;
");
getErrorString();

// Standard run: returns a=1
simulate(restartOverrideBug, simflags="-output a,y");

// Restart without override: returns a=1
simulate(restartOverrideBug, simflags="-r=tmp.mat -iit=1.0 -iif=restartOverrideBug_res.mat -iim=none -output a,y");

// Restart with override gives the same answer: returns a=1 (incorrect)
simulate(restartOverrideBug, simflags="-r=tmp.mat -iit=1.0 -iif=restartOverrideBug_res.mat -iim=none -output a,y -override a=2");

// No restart, but with override: returns a=2 (correct override)
simulate(restartOverrideBug, simflags="-r=tmp.mat -output a,y -override a=2");

The initialization problem of the following model is not handled correctly:

model PreMatrix
  Integer x[2,2] = integer(time*fill(1,2,2));
  Integer y[2,2];
algorithm 
  for i in 1:size(x,1),j in 1:size(x,2) loop
    y[i,j] := pre(x[i,j]);
  end for;
end PreMatrix;

The initialization passes without any message, but there should be some warnings about missing fixed start attribute for discrete states:

Warning: Assuming fixed start value for the following 4 variables:
         x[2,2]:DISCRETE(fixed = false ) .PreMatrix, .Integer type: Integer 
         x[2,1]:DISCRETE(fixed = false ) .PreMatrix, .Integer type: Integer 
         x[1,2]:DISCRETE(fixed = false ) .PreMatrix, .Integer type: Integer 
         x[1,1]:DISCRETE(fixed = false ) .PreMatrix, .Integer type: Integer

​https://test.openmodelica.org/libraries/Chemical_Experimental/BuildModelRecursive.html says that Chemical.Examples.AcidBase.AlbuminTitration takes 5400 seconds for the backend, which is incredibly much. The C-code also does not compile due to going beyond the pre-defined 256 level bracket nesting depth.

The web page ​https://openmodelica.org/doc/OpenModelicaUsersGuide/latest/omchelptext.html lists all compiler flags, and is automatically generated from omc help info.

The situation has improved dramatically once #2913 was addressed. However, there are four things that still need to be fixed.

1. The documentation gives -d and --debug as debug flags. However, in the page there are still three references to +d, which are very confusing because it is not clear whether that is a separate flag with different effects or not. Please always use the same notation -d everywhere. You may also mention in the documentation that the +d notation is deprecated but can still be found in existing code

2. The documentation ​https://openmodelica.org/doc/OpenModelicaUsersGuide/latest/omchelptext.html#omcflag-debug-section explains how to set up the -d flag. However, there are no examples. I would suggest to add one or two, and at least one where multiple flags are set and some are reset, for better clarity.

3. The documentation of --preOptModules, --initOptModules and --postOptModules does not contain any example, so it is not clear how one can actually set the options that are listed there. I would suggest to add a couple of examples from which it is clear what the syntax is.

4. The new flags introduced in ​https://github.com/OpenModelica/OMCompiler/pull/282 are not documented at all, please add the documentation of --forceRecommendedOrdering and also add the documentation about += and -= to --preOptModules, and --postOptModules. BTW, can this also be used with --initOptModules, or should a special section dedicated to flags used in initOptModules be also introduced in the documentation?

It is really weird that states and their derivatives have the same unit information in OMEdit/OMPlot. Because time is always in seconds, it should be rather easy to determine state derivative unit information by dividing the state unit information by seconds.

I try to load a very light and simple Dymola FMU in an SSP model and I get these errors: [4] 11:19:43 Ecriture Erreur [loadSnapshot] loading snapshot failed

[5] 11:21:35 Ecriture Erreur [initializeDependencyGraph_initialUnknowns] tankwin3264: erroneous dependencies detected in modelDescription.xml

[6] 11:21:35 Ecriture Erreur [NewComponent] tankwin3264: Couldn't initialize dependency graph for initial unknowns.

[7] 11:21:35 Ecriture Erreur [loadSnapshot] loading snapshot failed

When doing this a second time, a new error comes like this: [8] 11:26:19 Ecriture Erreur [addSubModel] "tankYPipeDym.Root.tankwin3264" already exists in the scope

Then I save the .ssp file and reopen it with an external browser. The FMU appears. There should not be dependencies troubles in Dymola FMUs. The FMU is attached to this ticket.

This thread of discussion is related to SSP runs (OMSimulator). I'm not able to identify the .csv file containing all variables after the run of s SSP model. I see the results in the viewer but where is the .csv file (name, location) for post-morten visualization?

Second question: is there somewhere some information related to a SSP model run (simulation time, time for FMU loading, etc.)?

Proposed here: ​http://stackoverflow.com/questions/42673420/global-measure-to-test-when-steady-state-of-system-is-reached

The first point sounds like a good to have feature and doesn't seem that hard to implement, basically a terminate when all the ders are 0 (within an epsilon) for a number of steps.

For the second point we would need a global operator (that returns true when all ders are within an epsilon interval) that one can use in equations to enable / disable parts of the system.

I run a SSP model, and after the successful execution, the model is taggued as modified in the browser. This is not undertandable as nothing has been changed in the ssp file due to the run.

Refer to the enclosed video.

There already exists the option to save values at events. If one has event iterations it also would be interesting to see the values at intermediate events in an event iteration. E.g., the model below starts an event iteration at time > 2 that results in the output

D x: 2
A x: 5
B x: 6

but in the (csv) result file at t=2 there are only two entries (x=2 and x=6).

model WhenPriority
 import S = Modelica.Utilities.Streams;
 Integer x(start=0,fixed=true);
equation
  when pre(x) == 2 then
    S.print("A x: "+String(x));
    x = pre(x) + 3;
  elsewhen pre(x) > 4 then
    S.print("B x: "+String(x));
    x = pre(x) + 1;
  elsewhen pre(x) > 3 then
    S.print("C x: "+String(x));
    x = pre(x) + 2;
  elsewhen time > 2 then
    S.print("D x: "+String(x));
    x = 2;
  end when;
end WhenPriority;

Of course that doesn't improve the insight to the model if plotting the variable with current OMPlot, but "customized" visualization solutions could profit from it and even try to visualize the event order.

The VS compiler is extremely fast, because it is specifically optimized for the Windows/Intel combination. We should enable the possiblity of using it to compile simulation executables, to reduce the lag from the pressing of the simulation button to the start of the actual simulation.

I tentatively assign this to mahge930, since he's already been working with CMAKE and has some experience on this topic.

If I run (using OM r20199) "gridTest21" of the enclosed package "AutoCreateBug" I get the following message:

"C:\Programmi\OpenModelica1.9.1Nightly\\MinGW\bin\mingw32-make.exe" -f AutoCreateBug.gridTest21.makefile
gcc   -falign-functions -msse2 -mfpmath=sse     -I"C:/Programmi/OpenModelica1.9.1Nightly//include/omc/c" -I. -DOPENMODELICA_XML_FROM_FILE_AT_RUNTIME  -c -o AutoCreateBug.gridTest21.o AutoCreateBug.gridTest21.c
AutoCreateBug.gridTest21.c: In function 'AutoCreateBug_gridTest21_eqFunction_27':
AutoCreateBug.gridTest21.c:458: error: expected expression before 'modelica_integer'
AutoCreateBug.gridTest21.c:458: error: expected ')' before '$Pi$rB$Pv'
AutoCreateBug.gridTest21.c:460: error: expected expression before 'modelica_integer'
AutoCreateBug.gridTest21.c:460: error: expected ')' before '$Pi$rB$Pv'
mingw32-make: *** [AutoCreateBug.gridTest21.o] Error 1
Compilation process exited with code 2

It seems that OM creates for this model a C file that is invalid.

OM counts just 2 equations for following model, but it should be 4.

model TestSimpleEqu
 record foo
   Real a;
   Real b;
 end foo;

 function foofoo2
   input foo x[2];
   output foo y[2];
 algorithm
   y[1].a := x[2].b;
   y[1].b := x[2].a;
   y[2].a := x[1].b;
   y[2].b := x[1].a;
 end foofoo2;
 
 foo g1[2];
equation
  //{foo(g1[1].a, g1[1].b), foo(g1[2].a, g1[2].b)} = foofoo2({foo(1.0, 2.0),foo(1.0, 2.0)});
  g1 = foofoo2({foo(1.0, 2.0),foo(1.0, 2.0)});
end TestSimpleEqu;

I get the following output by the simulation of a Petri net model:

C:/Users/TKLEIN~1/AppData/Local/Temp/OpenModelica/OMEdit/PNlib2.Examples.TestWkeit.exe -port=51187 -logFormat=xml -override=startTime=0,stopTime=5,stepSize=0.01,tolerance=1e-6,solver=dassl,outputFormat=mat,variableFilter=.* -r=PNlib2.Examples.TestWkeit_res.mat -dasslJacobian=coloredNumerical -w -lv=LOG_STATS

scalar s->index[0][0] == 2456744 incorrect, a->dim_size[0] == 2

This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.

Simulation process failed. Exited with code 255.

The following model (Top in the Bug package) fails compilation. For some reason, if z is 1, omc produces C code including calls to a nonexistent function Bug_Component_array_get. The calls to this function also reference a parameter components which does not exist in the caller's context.

Setting z to anything greater than 1 produces working code, where the Bug_Component_array_get is not declared not used anywhere.

Consider the following test model

class matrixAdd
  parameter Real[3,2] m={{1,2},{2,3},{4,5}};
  parameter Real[3,2] a={{2,2},{2,2},{2,2}};
  Real[3,2] b;
  Real max;
algorithm
  max:=0;
  for i in 1:3 loop
    for j in 1:2 loop
      b[i,j]:=m[i,j]+a[i,j];
      if (max<b[i,j]) then
          max:=b[i,j];
      end if;
    end for;
  end for;
end matrixAdd;

It causes the C compilation to break with

matrixAdd_05evt.c: In function 'matrixAdd_function_ZeroCrossings':
matrixAdd_05evt.c:51:161: error: 'i' undeclared (first use in this function)
   tmp0 = LessZC(data->localData[0]->realVars[6] /* max variable */, (&data->localData[0]->realVars[0] /* b[1,1] variable */)[calc_base_index_dims_subs(2, 3, 2, i, ((modelica_integer) 1))], data->simulationInfo->storedRelations[0]);
                                                                                                                                                                 ^
matrixAdd_05evt.c:51:161: note: each undeclared identifier is reported only once for each function it appears in
matrixAdd_05evt.c: In function 'matrixAdd_function_updateRelations':
matrixAdd_05evt.c:78:163: error: 'i' undeclared (first use in this function)
     tmp2 = LessZC(data->localData[0]->realVars[6] /* max variable */, (&data->localData[0]->realVars[0] /* b[1,1] variable */)[calc_base_index_dims_subs(2, 3, 2, i, ((modelica_integer) 1))], data->simulationInfo->storedRelations[0]);

Currently it's not possible to formulation final constraints in the optimization. In the moment we support path constraints with Otimica Extension.

 Real a,b,c,d,e;
.....
constraint
 a*b + c/d < e

In Optimica it's possible to write for final constraints.

 Real a,b,c,d,e;
.....
constraint
 a(finaTime)*b(finalTime) + c(1.0)*d(finalTime) < b(1.0)

where finalTime is a real parameter 1.0, which is unknown sometimes. Note: stopTime not equal finalTime!

There some issues with different time point and the case if finalTime is unknown.

It seem like the currently way in Modelica is to set marker via (Custom) annotations see e.g. ​http://www.ep.liu.se/ecp/096/018/ecp14096018.pdf

It's would be nice to have something like

 Real a,b,c,d,e;
.....
constraint
 a*b + c/d < e annotation(finalConstraint = true)

The issue is that a*b + c/d < e is an equation and the annotation going missing in the BackEnd. Alternative it should be possible to write

 Real a,b,c,d,e;
 Boolean f (finalConstraint = true);
.....
  equation
 f = a*b + c/d < e;

However custom annotations not really finished yet! Therefore, I prefer for the first prototype similar implementation like constraint with

 Real a,b,c,d,e;
.....
equation
....
final constraint
 a*b + c/d < e;
constraint
...

The model Modelica.Media.Incompressible.Examples.Glycol47 gets erroneous values for the poly_rho and poly_Cp constants, because the input array B to the Lapack function dgelsx is overwritten with the unassigned output argument x. This only happens when code generation is used to generate dgelsx.

The prefixes input, output, inner, outer, stream and flow are not allowed inside records, but we do not check this currently.

The MSL32 model Modelica.Electrical.Machines.Examples.AsynchronousInductionMachines.AIMC_YD takes over 100 minutes for simulation with Cellier Tearing because of way too much steps:

LOG_STATS         | info    | ### STATISTICS ###
|                 | |       | | timer
|                 | |       | | |    0.0322984s          reading init.xml
|                 | |       | | |     0.003696s          reading info.xml
|                 | |       | | |  0.000564877s          pre-initialization
|                 | |       | | |  0.000243327s [  0.0%] initialization
|                 | |       | | |   5.6153e-05s [  0.0%] steps
|                 | |       | | |    0.0105037s [  0.0%] creating output-file
|                 | |       | | |   0.00208371s [  0.0%] event-handling
|                 | |       | | |    0.0116676s [  0.0%] overhead
|                 | |       | | |      6586.88s [100.0%] simulation
|                 | |       | | |       6586.9s [100.0%] total
|                 | |       | | events
|                 | |       | | |     2 state events
|                 | |       | | |     0 time events
|                 | |       | | solver: DASSL
|                 | |       | | | 15358920 steps taken
|                 | |       | | | 30979010 calls of functionODE
|                 | |       | | | 23021038 evaluations of jacobian
|                 | |       | | |   217 error test failures
|                 | |       | | | 7673606 convergence test failures

With omcTearing:

LOG_STATS         | info    | ### STATISTICS ###
|                 | |       | | timer
|                 | |       | | |     0.031817s          reading init.xml
|                 | |       | | |   0.00360088s          reading info.xml
|                 | |       | | |  0.000564458s          pre-initialization
|                 | |       | | |  0.000215112s [  0.0%] initialization
|                 | |       | | |  0.000122851s [  0.0%] steps
|                 | |       | | |   0.00647291s [  0.7%] creating output-file
|                 | |       | | |   0.00109345s [  0.1%] event-handling
|                 | |       | | |   0.00640162s [  0.7%] overhead
|                 | |       | | |     0.849506s [ 98.3%] simulation
|                 | |       | | |     0.864376s [100.0%] total
|                 | |       | | events
|                 | |       | | |     2 state events
|                 | |       | | |     0 time events
|                 | |       | | solver: DASSL
|                 | |       | | |  8942 steps taken
|                 | |       | | | 12840 calls of functionODE
|                 | |       | | |   595 evaluations of jacobian
|                 | |       | | |   176 error test failures
|                 | |       | | |    83 convergence test failures

With calling of functions through components, identical functions are often generated. The C compiler will generate equivalent code for both, which is not desirable. We should figure out which functions are equivalent and replace calls to one with calls to another (or simply alias them via function pointer).

model N
  function f
    input Real r;
    output Real o := r;
  end f;
end N;

model M
  N n1,n2;
  Real r1 = n1.f(time), r2 = n2.f(time);
end M;

A Hash from DAE.Function to Absyn.Path should be sufficient (or String->Path and manually DAEDump the function to speed up any hash conflicts).

When working on ​Modelica_DeviceDrivers gh#114 I tried to run (on OpenModelica 1.11.0~dev-42-g1a73b84 under Ubuntu 16.04 64bit):

model Try4
  parameter String mem_name = "COM4";
  import Modelica_DeviceDrivers.Communication.SharedMemory;
  import Modelica_DeviceDrivers.Communication.SharedMemory_;
  import Modelica_DeviceDrivers.Packaging.SerialPackager;
  import Modelica_DeviceDrivers.Packaging.SerialPackager_;
protected
  SharedMemory sm = SharedMemory(mem_name, 16 * 1024);
  SerialPackager pkg = SerialPackager();
equation
  SharedMemory_.read(sm, pkg);
  annotation(uses(Modelica_DeviceDrivers(version="1.4.4")));
end Try4;

which compiles fine and even seems to execute but I got a translation warning:

[4] 09:08:37 Translation Warning
[Modelica_DeviceDrivers.Communication.SharedMemory: 17:3-25:17]: 
Could not find library pthread in either of:
  /usr/lib/x86_64-linux-gnu/omc/libpthread.a
  /usr/lib/x86_64-linux-gnu/omc/libpthread.so
  /usr/lib//libpthread.a
  /usr/lib//libpthread.so
 ...

the lib actually resides in:

$ locate libpthread.a
/usr/lib/x86_64-linux-gnu/libpthread.a
/usr/lib32/libpthread.a
/usr/libx32/libpthread.a

Same goes for librt.

So OMC should first search /usr/lib/x86_64-linux-gnu/ then /usr/lib/x86_64-linux-gnu/omc/

Please check the execstat log of the ​DistributionSystemModelicaActiveLoads_N_80_M_80 model. The preOpt clockPartitioning phase takes over 5 seconds and half a gigabye net memory to find out that the system has in fact no clocked partitions at all.

I guess it shouldn't be hard to check from NF data that there are no instances of clocks in the model, so as to skip this phase entirely. Is there any reason not to do that?

We need to change our data structures slightly in order to achieve a better performance and perhaps remove the need for boxptr-functions (marshalling, box/unbox).

(16:22:18) Martin Sjölund: I think in general the way we tag pointers is wrong (16:22:32) Martin Sjölund: we should tag integers instead (16:22:33) Adrian Pop: yes, probably tagging pointers is confusing the hell out of GC (16:22:49) Adrian Pop: we could probably not tag anything anymore (16:23:03) Martin Sjölund: maybe (16:23:11) Adrian Pop: as that is needed only for the previous RML gc (16:23:19) Adrian Pop: and have 32 bit integers (16:23:49) Martin Sjölund: !!! (16:23:55) Martin Sjölund: that would be great actually (16:24:19) Martin Sjölund: would probably be a good performance boost, too (16:24:23) Adrian Pop: the only problem i think it might be with debugging (16:24:40) Martin Sjölund: only problem is knowing what is integer and what is not (16:25:15) Martin Sjölund: or we box integers too (16:28:56) Martin Sjölund: I mean... allocating data could be as simple as: void*[2] = {header,double_word}; 64-bit void*[3] = {header,double_word1,double_word2}; 32-bit void*[2] = {header,word}; 32-bit, 64-bit char*[2] = {headerbyte1,headerbyte2,headerbyte3,headerbyte4,char1,...,charn}; 32-bit, 64-bit (16:30:28) Martin Sjölund: since GC is simplified, we need less bits for book-keeping too (16:31:22) Martin Sjölund: could just be a bitset for int|double|enum|string|list|option|record = 3 bits to tag what data it is (16:31:32) Adrian Pop: true (16:32:21) Martin Sjölund: runtime could be quite simple with that (16:32:47) Martin Sjölund: oh, I forgot array as type (of course) (16:33:14) Martin Sjölund: but still 3 bits... and arrays could have 3 bits to say what is stored in it (same 8 types) (16:33:51) Martin Sjölund: and same for lists (16:34:09) Martin Sjölund: lists of unboxed data could be cool (16:35:06) Martin Sjölund: and I guess 28-bit arrays still fine length (16:37:03) Martin Sjölund: and uniontypes with only atomic members (many of them) could be allocated atomic too... (16:37:10) Martin Sjölund: so many things that can be done... (16:37:20) Adrian Pop: :) (16:37:35) Martin Sjölund: do you think we should move in this direction? (16:38:09) Martin Sjölund: before we rely too much on the eclipse debugger :D (16:38:26) Adrian Pop: i think we should not do that yet (16:38:38) Adrian Pop: wait until we move to the bootstrapped compiler (16:38:46) Adrian Pop: and then we can do changes