-
Notifications
You must be signed in to change notification settings - Fork 0
3. ReadROOT's IOClasses
The IOClasses that are coded in the ReadROOT package allow for some modification of the C++ files. The current C++ code needs very specific header files and to save time, the IOClasses allow to save a configuration. This means if you use a different computer, you can create and load different configurations of the C++ header files.
Note that you do not have to manually do the configuration as is explained here. Upon importing the package, configurations can easily be added. You can also forcefully reload the configuration in case it wasn't done beforehand.
The FileExplorer class simply uses Tkinter's filedialog. You can either ask the FileExplorer to find files or find a specific folder. In both cases the initial search directory can be selected.
>>> import ReadROOT as r
>>> file_explorer = r.fileexp()
>>> #For files:
>>> file_explorer.FindFiles()
>>> #For folders:
>>> file_explorer.FindFolder()
The Configuration class will allow you to save the paths to the C++ header files needed for the C++ code. In order to create a configuration, a name and a .json file are used to first initialize the configuration. If the configuration already exists in the .json file, it can be loaded back up. If you just made a new configuration, you can save it to a .json file or simply keep it in memory (this will be deleted if Python is closed).
By default creating a Configuration instance will not do anything. After calling find_required_headers the Configuration's required headers and pybind headers will be updated. By default, only 3 header files are needed. If extra files are needed the number of required headers can be changed using set_required_headers. If autosave is set to True, the Configuration instance will automatically be saved to the selected .json file.
>>> import ReadROOT as r
>>> new_config = r.config("Name of your configuration", "config.json") #Here we use "config.json" since this file already exists in the ReadROOT package.
>>> new_config.find_required_headers(autosave=True) #This will allow you to look for the header files and automatically save their path in the json file.
The SetUpCpp class only needs a file path (a C++ file) to update in order to be initialized. Using a selected configuration, one can update the C++ files.
After making a Configuration, it is possible to update two C++ files. The following code will show you how to update both the funcs.hpp and wrap.cpp which are the two C++ files that need to be updated if a different computer or Python version is used.
>>> import ReadROOT as r
>>> new_setup = r.setup("funcs.hpp")
>>> new_setup.load_configuration(new_config)
>>> new_setup = r.setup("wrap.cpp")
>>> new_setup.load_pybind11(new_config)
It is possible to directly create a new configuration after importing the ReadROOT package. In this case, the add_config function will help you create a new configuration. This function only requires a name to be fed to it.
>>> import ReadROOT as r
>>> r.add_config("name_of_your_configuration")
This will then trigger the whole process of making a new configuration, meaning it will force you to look for the numpy.h, pybind11.h and Python.h headers until all 3 have been selected. Note that there is no verification of the files, meaning that if the wrong files have been selected you will not be able to use this configuration.
Currently, the config.json file (file that comes with ReadROOT and contains some saved configurations) contains an extra parameter called LoadConfig. If set to True, upon importing the ReadROOT package, you will be able to select a Configuration in the list of configurations saved in config.json. If you want to force the configuration reload, the following commands can be executed:
>>> import ReadROOT as r
>>> r.do_config()