__init__.py

"""
pynbody
=======

A light-weight, portable, format-transparent analysis framework
for N-body and SPH astrophysical simulations.

Getting help
------------

Documentation for `pynbody` is maintained in two forms.

 1. Tutorial-style documentation, which can be
    accessed :ref:`here <tutorials>`. If you are viewing this help from
    within python, please visit http://www.itp.uzh.ch/~roskar/pynbody/docs/.
 2. Reference documentation within the code, which can be accessed
    using the standard python help() function, the ipython ? operator,
    or online.

There is also a `user group <https://groups.google.com/forum/?fromgroups=#!forum/pynbody-users>`_
where the developers are happy to help with any problems you encounter.



What's available
----------------

Pynbody handles Gadget, Gadget-HDF, Tipsy, Nchilada and Ramses files. To load
any of these, use

>>> f = pynbody.load(filename)

to create a :class:`pynbody.snapshot.SimSnap` object *f*, which then acts
as a dictionary holding the arrays inside *f*. For more information
see :ref:`data-access`.

Configuration
-------------

Various aspects of the behaviour of pynbody can be controlled.
See <http://code.google.com/p/pynbody/wiki/ConfigFiles>

Subpackages
-----------

:mod:`~pynbody.array`
    Extends numpy arrays with a custom class array.SimArray
    which holds additional information like units.

:mod:`~pynbody.bridge`
    Allows connections to be made between two different
    SimSnap objects in various ways. 
    <http://code.google.com/p/pynbody/wiki/SameSimulationDifferentOutputs>

:mod:`~pynbody.simdict`
    
:mod:`~pynbody.derived`
    Holds procedures for creating new arrays from existing
    ones, e.g. for getting the radial position. 
    <http://code.google.com/p/pynbody/wiki/AutomagicCalculation>

:mod:`~pynbody.family`
    Stores a registry of different particle types like dm,
    star, gas.
    <http://code.google.com/p/pynbody/wiki/TheFamilySystem>

:mod:`~pynbody.filt`
    Defines and implements 'filters' which allow abstract subsets
    of data to be specified.
    <http://code.google.com/p/pynbody/wiki/FiltersAndSubsims>

:mod:`~pynbody.gadget`
    Implements classes and functions for handling gadget files;
    you rarely need to access this module directly as it will
    be invoked automatically via pynbody.load.

:mod:`~pynbody.gadgethdf`
    Implements classes and functions for handling gadget HDF files
    (if h5py is installed); you rarely need to access this module
    directly as it will be invoked automatically via pynbody.load

:mod:`~pynbody.halo`
    Implements halo catalogue functions. If you have a supported
    halo catalogue on disk or a halo finder installed and
    correctly configured, you can access a halo catalogue through
    f.halos() where f is a SimSnap.
    <http://code.google.com/p/pynbody/wiki/HaloCatalogue>

:mod:`~pynbody.kdtree`
    Implements a KD Tree based on Joachim Stadel's smooth.c.
    You are unlikely to need to access this module directly
    as KD Trees are built in higher level analysis code
    automatically.

:mod:`~pynbody.snapshot`
    Implements the basic SimSnap class and also SubSnap classes
    which can represent different views of the same data.
    <http://code.google.com/p/pynbody/wiki/FiltersAndSubsims>
    You rarely need to access this module directly.

:mod:`~pynbody.sph`
    Allows SPH images to be rendered. The easiest interface
    to this module, at least to start with, is through the
    pynbody.plot package.
    <http://code.google.com/p/pynbody/wiki/SphImages>

:mod:`~pynbody.tipsy`
    Implements classes and functions for handling tipsy files.
    You rarely need to access this module directly as it will
    be invoked automatically via pynbody.load.

:mod:`~pynbody.units`
    Implements a light-weight unit class which is
    used to automatically track units of your simulation arrays.
    <http://code.google.com/p/pynbody/wiki/ConvertingUnits>

:mod:`~pynbody.util`
    Various utility routines used internally by pynbody.

    
"""

from . import backcompat

# Import basic dependencies
import ConfigParser
import os
import imp
import numpy
import warnings
import sys

# Create config dictionaries which will be required by subpackages
# We use the OrderedDict, which is default in 2.7, but provided here for 2.6/2.5 by
# the backcompat module. This keeps things in the order they were parsed (important
# for units module, for instance).
config_parser = ConfigParser.ConfigParser(dict_type = backcompat.OrderedDict)
config = {}


# Process configuration options
config_parser.optionxform = str
config_parser.read(os.path.join(os.path.dirname(__file__),"default_config.ini"))
config_parser.read(os.path.join(os.path.dirname(__file__),"config.ini"))
config_parser.read(os.path.expanduser("~/.pynbodyrc"))
config_parser.read("config.ini")



config= {'verbose': config_parser.getboolean('general','verbose'),
         'tracktime': config_parser.getboolean('general','tracktime'),
         'centering-scheme': config_parser.get('general','centering-scheme')}

config['snap-class-priority'] = map(str.strip,
                                    config_parser.get('general', 'snap-class-priority').split(","))
config['halo-class-priority'] = map(str.strip,
                                    config_parser.get('general', 'halo-class-priority').split(","))


config['default-cosmology'] = {}
for k in config_parser.options('default-cosmology') :
    config['default-cosmology'][k] = float(config_parser.get('default-cosmology', k))

config['sph'] = {}
for k in config_parser.options('sph') :
    try:
        config['sph'][k] = int(config_parser.get('sph', k))
    except ValueError:
        pass
    
config['threading'] = config_parser.get('general', 'threading')
config['number_of_threads'] = int(config_parser.get('general', 'number_of_threads'))

config['gravity_calculation_mode'] = config_parser.get('general', 'gravity_calculation_mode')
config['disk-fit-function'] = config_parser.get('general', 'disk-fit-function')

# Import subpackages
from . import util, filt, array, family, snapshot,  tipsy, gadget, gadgethdf, ramses, grafic, analysis, halo, derived, bridge, gravity, sph, nchilada

try: 
    from . import plot
except: 
    warnings.warn("Unable to import plotting package (missing matplotlib or running from a text-only terminal? Plotting is disabled.", RuntimeWarning)

# The following code resolves inter-dependencies when reloading
imp.reload(array)
imp.reload(util)
# imp.reload(family) # reloading this causes problems for active snapshots
imp.reload(snapshot)
imp.reload(nchilada)
imp.reload(tipsy)
imp.reload(gadget)
imp.reload(gadgethdf)
imp.reload(ramses)
imp.reload(filt)
imp.reload(analysis)
imp.reload(halo)
imp.reload(derived)
imp.reload(bridge)
imp.reload(gravity)
imp.reload(sph)
imp.reload(grafic)

try : 
    imp.reload(plot)
except : 
    pass

# from analysis import profile


# This is our definitive list of classes which are able to
# load snapshots
_snap_classes = [gadgethdf.GadgetHDFSnap, nchilada.NchiladaSnap, gadget.GadgetSnap, tipsy.TipsySnap, ramses.RamsesSnap, grafic.GrafICSnap]

# Turn the config strings for snapshot/halo classes into lists of
# actual classes
_snap_classes_dict = dict([(x.__name__,x) for x in _snap_classes])
_halo_classes_dict = dict([(x.__name__,x) for x in halo._halo_classes])
config['snap-class-priority'] = [_snap_classes_dict[x] for x in config['snap-class-priority']]
config['halo-class-priority'] = [_halo_classes_dict[x] for x in config['halo-class-priority']]


def load(filename, *args, **kwargs) :
    """Loads a file using the appropriate class, returning a SimSnap
    instance."""
   
    for c in config['snap-class-priority'] :
        if c._can_load(filename) :
            if config['verbose'] : print>>sys.stderr, "Loading using backend",str(c)
            return c(filename,*args,**kwargs)

    raise IOError("File %r: format not understood or does not exist"%filename)


from snapshot import _new as new

derived_array = snapshot.SimSnap.derived_quantity

__all__ = ['load', 'new', 'derived_array']