Add/remove modules dynamically

“SALOME on demand” feature provides a possibility to dynamically add or remove SALOME modules directly from GUI.

Description file

“SALOME on demand” feature introduces notion of a module description file. This is the file in JSON format that has .salomex extension. The description file provides some parameters of the module; two parameters are mandatory: name of the module and path to its root directory.


  "name": "PYHELLO",
  "root": "/home/user/salome/modules/PYHELLO"


In the description file, the name parameter is an internal name of the module, in contrast to its title which is described in SalomeApp.xml or LightApp.xml resource file.


One .salomex file may describe one SALOME module only.

Modules toolbar

The Modules toolbar provides two additional buttons, which allow adding or removing modules directly from GUI.

Modules toolbar

Adding modules

Pressing the first button add shows a standard Open File dialog that allows selecting one or more .salomex files. Each module correctly configured in the corresponding .salomex file is then added to the SALOME session and appears in the Modules toolbar. As soon as it is done, the user may work with the module in the same way as with any other module: all features are available after activating module from Modules toolbar.

The modules added in this way become persistent. This means that after restarting SALOME these modules are available in the Modules toolbar.

Removing modules

Pressing a small arrow near the second button remove shows a drop-down menu with the list of manually added modules, where the user may select a module to remove. As soon as user selects a module to remove, a message box is shown asking the user if it is necessary also to remove a directory, containing the module, from disk. If user confirms removal of the directory, it will be necessary to save the study, in order to avoid possible data loss.


The module libraries are not removed from RAM until exit from the current SALOME session, so some its functionality may be still accessible, for example, via Python API. However, using module functionality may have unpredicted behavior.