PsrPopPy
========

(For full documentation, see http://samb8s.github.com/PsrPopPy/ or manual.pdf)

Python implementation of PSRPOP (which was written by D Lorimer).
Several of the old models from that (e.g. NE2001) are still included in their native fortran, since re-writing those is beyond the scope of this work. Currently, only a rudimentary makefile is included. This is something that needs work from a willing volunteer!

The main external requirements are [matplotlib](matplotlib.sourceforge.net) and [wxPython](http://wxpython.org/), which are used for the visualization stuff. It has a very useful API for making simple GUIs, as well as making beautiful plots. I've had difficulty compiliing wxPython from scratch on more recent versions of Mac OS X, but it should be straightforward to install via macports or similar.

Thanks
------

Many thanks to recent suggestions from Manjari Bagchi and Anirban Chakraborty. Their work https://arxiv.org/abs/2012.13243 used PsrPopPy, and they found some issues which I've been happy to correct.

If you spot any issues, please either let me know, or make a pull request - I'm not actively developing the code anymore, but I'm happy to keep it reasonably maintained.

Dev Notes
---------

I've just added scintillation effects to `dosurvey`! They can be switched on by adding the
flag --scint, but are off by default. The code uses equations from Lorimer & Kramer and 
the NE2001 code to calculate modulation indices for pulsars in the population. The S/N of each
pulsar is then scaled up or down using the modulation index.

Compiling
---------

I've just incorporated a real makefile for the first time - be sure to edit `makefile.linux` or `makefile.darwin` as appropriate to point to the correct location of your gfortran compiler. After that, it should be as simple as typing `make`.

If this fails, try using the scripts. Inside the lib/fortran directory, edit `make_mac.csh` or `make_linux.csh` (as appropriate) -- change the variable `gf' to point to your local gfortran compiler, then run the script. Fingers crossed, it should all work.

Note mac users should be sure to use a suitable version of gfortran - available from http://gcc.gnu.org/wiki/GFortranBinaries

### Compiling using `enscons`

An experimental compiling option is available to install everything as a python package, called `psrpoppy`, using
[`enscons`](https://bitbucket.org/dholth/enscons) (available on PyPi via `pip install enscons`). To compile and install the package system-wide just use:

```
sudo python setup.py install
```

After this you can import the python module with:

```python
import psrpoppy
```

or, following [this example](examples/populate_and_survey.py), you could do:

```python
from psrpoppy import populate

pop = populate.generate(1038, 
                        surveyList=['PMSURV'],
                        radialDistType='lfl06',
                        siDistPars=[-1.41, 0.96], # non-standard SI distribution
                        duty_percent=6.,
                        electronModel='lmt85',
                        nostdout=True # switches off output to stdout
                       )
```

Usage
=====

If not installing the package via `enscons` I'd recommend adding the `psrpoppy` directory to your PYTHONPATH and adding the `bin` directory to your PATH. This should then leave you set up to run the code from wherever you like.


A brief description of the "executables" follows.

populate
--------

Create a population mode using user-defined parameters using the snapshot method

evolve
------

Create a populate model using the Ridley & Lorimer evolution method

dosurvey 
--------

Run a population model through a survey. Pre-defined surveys are given, but a user may also create their own.

wxView
------

More detailed population model viewer. Make histograms, scatter plots, etc. All based off the wx backend for matplotlib.

wxHist
------

Make more intricate histograms, including histograms of multiple population models.