"""
Module for setting up species in the working folder.
"""
import json
import socket
import urllib.request
from configparser import ConfigParser
from os import environ
from pathlib import Path, PosixPath, WindowsPath
import h5py
from beartype import beartype
from beartype.typing import Optional, Union
from .._version import __version__, __version_tuple__
[docs]
class SpeciesInit:
"""
Class for initiating species by creating the database and
configuration file in case they are not present in the working
folder, and creating the data folder for storage of input data.
"""
@beartype
def __init__(
self,
config_file: Optional[Union[str, PosixPath, WindowsPath]] = None,
database_file: Optional[Union[str, PosixPath, WindowsPath]] = None,
) -> None:
"""
Parameters
----------
config_file : str, PosixPath, WindowsPath, None
Path to the configuration file that is stored as
`species_config.ini`. Setting the argument enables
reading a configuration file at a different path
than the current working directory. The default
location (i.e. the working folder of your script or
notebook) is used by setting the argument to ``None``.
database_file : str, PosixPath, WindowsPath, None
Path to the HDF5 database that is stored as
`species_database.hdf5`. Setting the argument will
overwrite the database path in the configuration file.
The path from the configuration file is used if the
argument of ``database_file`` is set to ``None``.
Returns
-------
NoneType
None
"""
print("=======\nspecies\n=======")
# Check if there is a new version available
species_version = (
f"{__version_tuple__[0]}."
f"{__version_tuple__[1]}."
f"{__version_tuple__[2]}"
)
try:
pypi_url = "https://pypi.org/pypi/species/json"
with urllib.request.urlopen(pypi_url, timeout=1.0) as open_url:
url_content = open_url.read()
url_data = json.loads(url_content)
pypi_version = url_data["info"]["version"]
except (urllib.error.URLError, socket.timeout):
pypi_version = None
if pypi_version is not None:
pypi_split = pypi_version.split(".")
current_split = species_version.split(".")
new_major = (pypi_split[0] == current_split[0]) & (
pypi_split[1] > current_split[1]
)
new_minor = (
(pypi_split[0] == current_split[0])
& (pypi_split[1] == current_split[1])
& (pypi_split[2] > current_split[2])
)
if new_major | new_minor:
print(f"\n-> species v{pypi_version} is available!")
print(f"\nVersion: {__version__}")
working_folder = Path.cwd()
print(f"Working folder: {working_folder}")
if config_file is None:
config_file = working_folder / "species_config.ini"
elif isinstance(config_file, str):
config_file = Path(config_file)
environ["SPECIES_CONFIG"] = str(config_file)
config = ConfigParser(allow_no_value=True)
if config_file.exists():
print(f"\nConfiguration file: {config_file}")
else:
print("\nCreating species_config.ini...", end="", flush=True)
config.add_section("species")
# config.set('species', '; File with the HDF5 database')
config.set("species", "database", "species_database.hdf5")
# config.set('species', '; File with the HDF5 database')
config.set("species", "data_folder", "./data/")
# config.set('species', '; File with the HDF5 database')
config.set("species", "vega_mag", "0.03")
with open(config_file, "w", encoding="utf-8") as file_obj:
config.write(file_obj)
print(" [DONE]")
config.read(config_file)
config_update = False
if database_file is None:
if "database" in config["species"]:
database_file = Path(config["species"]["database"])
else:
database_file = Path("./species_database.hdf5")
config.set("species", "database", "species_database.hdf5")
config_update = True
else:
if isinstance(database_file, str):
config.set("species", "database", database_file)
database_file = Path(database_file)
else:
config.set("species", "database", str(database_file))
config_update = True
if "data_folder" in config["species"]:
data_folder = Path(config["species"]["data_folder"])
else:
data_folder = Path("./data/")
config.set("species", "data_folder", "./data/")
config_update = True
if "vega_mag" in config["species"]:
vega_mag = config["species"]["vega_mag"]
else:
vega_mag = 0.03
config.set("species", "vega_mag", "0.03")
config_update = True
if config_update:
# If condition is needed because the file should
# not be opened when using MPI and the config
# file is already present and no need to update
with open(config_file, "w", encoding="utf-8") as file_obj:
config.write(file_obj)
if database_file.exists():
print(f"Database file: {database_file}")
else:
print("Creating species_database.hdf5...", end="", flush=True)
h5_file = h5py.File(database_file, "w")
h5_file.close()
print(" [DONE]")
try:
from mpi4py import MPI
mpi_rank = MPI.COMM_WORLD.Get_rank()
MPI.COMM_WORLD.Barrier()
except ImportError:
mpi_rank = 0
# Add samples to the database
if mpi_rank == 0:
with h5py.File(database_file, "a") as hdf5_file:
if "configuration" in hdf5_file:
del hdf5_file["configuration"]
config_group = hdf5_file.create_group("configuration")
config_group.attrs["config_file"] = str(config_file)
config_group.attrs["database_file"] = str(database_file)
config_group.attrs["data_folder"] = str(data_folder)
if data_folder.exists():
print(f"Data folder: {data_folder}")
else:
print("Creating data folder...", end="", flush=True)
data_folder.mkdir()
print(" [DONE]")
print("\nConfiguration settings:")
print(f" - Database: {database_file}")
print(f" - Data folder: {data_folder}")
print(f" - Magnitude of Vega: {vega_mag}")
try:
from mpi4py import MPI
# Rank of this process in a communicator
mpi_rank = MPI.COMM_WORLD.Get_rank()
# Number of processes in a communicator
mpi_size = MPI.COMM_WORLD.Get_size()
print("\nMultiprocessing: mpi4py installed")
print(f"Process number {mpi_rank+1:d} out of {mpi_size:d}...")
except ImportError:
print("\nMultiprocessing: mpi4py not installed")
