lhcbstyle 0.1.0

LHCb style for ROOT plots

LHCbStyle

Packaged version of the lhcbStyle.C macro with a python wrapper and utilities.

Looking for just the macro itself? Click here

Installation

NB: if using lb-conda default, this package is already available.

Via package managers

The python module can be installed using pip:

pip install lhcbstyle

or conda (via conda-forge):

conda install -c conda-forge lhcbstyle

Installing the macro

NB: conda users can ignore this step. It is done automatically.

After installing the package with pip or from source, there is one last step to install the lhcbStyle.C macro to ROOT's macro path. This is done using the console command:

lhcbstyle install

The default directory is found using the PyROOT function gROOT.GetMacroDir(), which is typically somewhere like /usr/share/root/macros or $ROOTSYS/macros. If you don't have permission to write to this directory, or you want to install the macro to a different ROOT installation, you can set the destination with an environment variable:

ROOT_MACRO_DIR=/path/to/macros lhcbstyle install

or use the command-line option --macro-dir:

lhcbstyle install --macro-dir /path/to/macros

If your chosen path does not correspond to $ROOTSYS/macros, then you can add it to ROOT's macro paths using your ~/.rootrc file. For example, if you installed lhcbStyle.C to ~/.local/share/root/macros then add this line to ~/.rootrc:

Unix.*.Root.MacroPath: .:$(HOME)/.local/share/root/macros:$(ROOTSYS)/macros

Usage

In C++

With the macro successfully installed, you can call it in the usual way:

gROOT->ProcessLine(".L lhcbStyle.C");
lhcbStyle();

The function lhcbStyle() takes a boolean argument print_msg which controls whether a message is printed to stdout upon setting the style.

The TPaveText (lhcbName), TText (lhcbLabel) and TLatex (lhcbLatex) objects are now located in the LHCbStyle namespace.

For example, to draw the "LHCb" blurb on a plot, using the default position:

LHCbStyle::lhcbName->Draw("same");

The TStyle object (lhcbStyle) is also available in the same namespace.

For example, to adjust the style before calling lhcbStyle():

LHCbStyle::lhcbStyle->SetPadLeftMargin(0.16);

or after calling lhcbStyle():

gStyle->SetPadLeftMargin(0.16);

In python

Naturally, in PyROOT, you can execute the macro in much the same way as in C++:

ROOT.gROOT.ProcessLine(".L lhcbStyle.C")
ROOT.lhcbStyle()

However, the package includes a python class called LHCbStyle which adds context management, allowing the use of the with keyword, e.g.:

from lhcbstyle import LHCbStyle

with LHCbStyle() as lbs:
    can = ROOT.TCanvas()
    hist = make_plot()
    hist.Draw("E1")
    lbs.lhcbName.Draw("same")
    can.SaveAs("hist_lhcb.png")

From the command line

There is also a console command lhcbstyle plot which draws a plotable object saved to a ROOT file.

The positional arguments are:

1. the ROOT file to open
2. the name of the TObject to draw
3. the output filename, without extension (optional, defaults to "plot")

The optional arguments are:

• --draw-style, -D: option string to pass to the Draw() function
• --formats, -F: file extensions to save the plot as

Additionally an label can be added to the plot with one of:

• --official: draw "LHCb" label
• --preliminary: draw "LHCb Preliminary" label
• --simulation: draw "LHCb Simulation" label
• --unofficial: draw "LHCb Unofficial" label
• --no-label: don't draw any label (default)

For example, to open plot.root, extract a TH1D called hist and draw with error bars and save it as Fig1.pdf, Fig1.eps and Fig1.png:

lhcbstyle plot plot.root hist Fig1 --formats pdf eps png

See the full usage notes using

lhcbstyle plot -h

Contributing

Creating a development environment

ssh://[e-mail]:7999/lhcb-docs/lhcbstyle.git
cd lhcbstyle
mamba create --name test-env root pytest pip setuptools_scm
pip install -e .[testing]
pre-commit install
curl -o lb-check-copyright "https://gitlab.cern.ch/lhcb-core/LbDevTools/raw/master/LbDevTools/SourceTools.py?inline=false"
chmod +x lb-check-copyright

Running the tests:

pre-commit run --all-files
pytest