Move the pickle download out of setup.py into decomposer_base.py

.github/workflows/ci.yml

@@ -4,8 +4,8 @@ jobs:
   build:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
-      - uses: actions/setup-python@v4
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
         with:
           python-version: 3.9
       - name: Install dependencies
@@ -26,7 +26,7 @@ jobs:
           coverage report -m
           coverage xml
       - name: Upload coverage report to Codecov
-        uses: codecov/codecov-action@v3
+        uses: codecov/codecov-action@v4
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
           files: coverage.xml

README.md

@@ -51,6 +51,8 @@ pyPAHdb can be directly installed from the
 
 `pip install git+https://github.com/PAHdb/pyPAHdb.git`
 
+Note that upon first run a pre-computed matrix will be automatically downloaded.
+
 ## Supported data formats
 
 pyPAHdb supports reading IPAC tables, _Spitzer_ FITS files and _JWST_ FITS files.

pypahdb/decomposer_base.py

@@ -15,9 +15,11 @@
 import os
 import pickle
 from functools import partial
+from urllib.request import urlretrieve
 
+import importlib_resources
 import numpy as np
-import pkg_resources
+from tqdm import tqdm
 from astropy import units as u
 from scipy import optimize
 from specutils import Spectrum1D
@@ -115,16 +117,38 @@ def __init__(self, spectrum):
             print("spectral data is all zeros.")
             return None
 
-        # Retrieve the precomputed data and raise error if file is
-        # not found.
-        file_name = "resources/precomputed.pkl"
-        file_path = pkg_resources.resource_filename("pypahdb", file_name)
-        with open(file_path, "rb") as f:
-            try:
-                self._precomputed = pickle.load(f, encoding="latin1")
-            except Exception as e:
-                print("Python 3 is required for pypahdb.")
-                raise (e)
+        # Download the precomputed data if not present
+        remote_pkl = "https://www.astrochemistry.org/pahdb/pypahdb/pickle.php"
+        if os.getenv("GITHUB_ACTIONS") == "true":
+            remote_pkl += "?github_actions=true"
+        local_pkl = (
+            importlib_resources.files("pypahdb") / "resources/precomputed.pkl"
+        )
+        if not os.path.isfile(local_pkl):
+
+            def hook(t):
+                last_b = [0]
+
+                def inner(b=1, bsize=1, tsize=None):
+                    if tsize is not None:
+                        t.total = tsize
+                    t.update((b - last_b[0]) * bsize)
+                    last_b[0] = b
+                return inner
+
+            print("downloading pre-computed matrix")
+            with tqdm(
+                unit="B",
+                unit_scale=True,
+                leave=True,
+                miniters=1,
+            ) as t:
+                urlretrieve(
+                    remote_pkl, filename=local_pkl, reporthook=hook(t), data=None
+                )
+
+        with open(local_pkl, "rb") as f:
+            self._precomputed = pickle.load(f, encoding="latin1")
 
         # Linearly interpolate the precomputed spectra onto the
         # frequency grid of the input spectrum.

requirements.txt

@@ -2,3 +2,5 @@ scipy
 astropy
 specutils
 matplotlib
+importlib_resources
+tqdm

setup.py

@@ -1,13 +1,7 @@
-from os import path
-from os import getenv
 from setuptools import setup, find_packages
-from setuptools.command.build_py import build_py
 import versioneer
 
-try:
-    from urllib.request import urlopen
-except ImportError:
-    from urllib2 import urlopen
+from os import path
 
 """
 NOTE: This file must remain Python 2 compatible for the foreseeable future,
@@ -39,21 +33,6 @@
     requirements = [line for line in requirements_file.read().splitlines()
                     if not line.startswith('#')]
 
-
-class BuildPyCommand(build_py):
-    def run(self):
-        remote_pkl = 'https://www.astrochemistry.org/pahdb/pypahdb/pickle.php'
-        if getenv('GITHUB_ACTIONS') == 'true':
-            remote_pkl += '?github_actions=true'
-        local_pkl = 'pypahdb/resources/precomputed.pkl'
-        # honor the --dry-run flag
-        if not self.dry_run and not path.isfile(local_pkl):
-            response = urlopen(remote_pkl)
-            with open(path.join(here, local_pkl), 'wb') as f:
-                f.write(response.read())
-        versioneer.get_cmdclass()['build_py'].run(self)
-        build_py.run(self)
-
 # Arguments marked as "Required" below must be included for upload to PyPI.
 # Fields marked as "Optional" may be commented out.
 
@@ -222,6 +201,5 @@ def run(self):
 
     # Run custom commands
     cmdclass={  # Optional
-        'build_py': BuildPyCommand,
     },
 )