sacc 0.16__tar.gz → 1.0__tar.gz

{sacc-0.16 → sacc-1.0}/.github/workflows/desc-ci.yml

@@ -25,9 +25,16 @@ jobs:
 
     - name: Dependencies
       run: |
-        python -m pip install --upgrade numpy pytest
+        python -m pip install --upgrade numpy pytest pytest-cov coverage pip
         python -m pip install '.[all]'
 
-    - name: Run tests
+    - name: Run tests and measure coverage
       run: |
-        pytest -vv
+        pytest  --cov sacc --cov-report xml:coverage.xml   ./test
+
+    - name: Report coverage
+      uses: orgoro/coverage@v3.2
+      if: matrix.python-version == '3.11'
+      with:
+        coverageFile: ./coverage.xml
+        token: ${{ secrets.GITHUB_TOKEN }}

{sacc-0.16 → sacc-1.0}/.github/workflows/publish.yml

@@ -17,7 +17,8 @@ jobs:
 
     - name: Build wheel and source tarball
       run: |
-        python setup.py sdist
+        python -m pip install build
+        python -m build . --sdist --wheel
 
     - name: Publish to Real PyPI
       uses: pypa/gh-action-pypi-publish@release/v1

{sacc-0.16/sacc.egg-info → sacc-1.0}/PKG-INFO

@@ -1,17 +1,25 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: sacc
-Version: 0.16
+Version: 1.0
 Summary: SACC - the LSST/DESC summary statistic data format library
-Home-page: https://github.com/LSSTDESC/sacc
-Author: LSST DESC
-Author-email: joezuntz@googlemail.com
-License: UNKNOWN
-Platform: UNKNOWN
+Author-email: LSST DESC <joezuntz@googlemail.com>
+License: BSD-3-Clause
+Project-URL: Homepage, https://github.com/LSSTDESC/sacc
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
 Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: scipy
+Requires-Dist: numpy>=1.20
+Requires-Dist: astropy
 Provides-Extra: all
+Requires-Dist: qp-prob[all]; extra == "all"
+Requires-Dist: numpydoc; extra == "all"
 Provides-Extra: doc
+Requires-Dist: numpydoc; extra == "doc"
 Provides-Extra: qp
-License-File: LICENSE
+Requires-Dist: qp-prob[all]; extra == "qp"
+Dynamic: license-file
 
 Sacc
 ====
@@ -58,5 +66,3 @@ Citation
 Sacc has been submitted to the [Astrophysics Source Code Library](https://ascl.net/code/v/2277); follow the link that will appear there to NASA ADS to export a bibtex citation.
 
 The core developers of Sacc are Joe Zuntz (maintainer), David Alonso, and Matt Becker.
-
-

sacc-1.0/pyproject.toml

@@ -0,0 +1,52 @@
+[project]
+name = "sacc"
+description = "SACC - the LSST/DESC summary statistic data format library"
+readme = "README.md"
+license = { text = "BSD-3-Clause" }
+authors = [
+    { name = "LSST DESC", email = "joezuntz@googlemail.com" }
+]
+dynamic = ["version"]
+
+
+
+classifiers = [
+    "Programming Language :: Python :: 3",
+    "Operating System :: OS Independent"
+]
+
+dependencies = [
+    "scipy",
+    "numpy>=1.20",
+    "astropy"
+]
+
+
+
+[build-system]
+requires = [
+    "setuptools>=42",
+    "setuptools_scm<7.0",
+    "setuptools_scm_git_archive",
+    "wheel"
+]
+build-backend = "setuptools.build_meta"
+
+[project.urls]
+Homepage = "https://github.com/LSSTDESC/sacc"
+
+
+[project.optional-dependencies]
+all = ["qp-prob[all]", "numpydoc"]
+doc = ["numpydoc"]
+qp = ["qp-prob[all]"]
+
+[tool.setuptools.packages.find]
+where = ["."]
+include = ["sacc"]
+
+[tool.setuptools.dynamic]
+version = {attr = "sacc.__version__"}
+
+[tool.coverage.run]
+relative_paths = true

{sacc-0.16 → sacc-1.0}/sacc/__init__.py

@@ -3,4 +3,4 @@ from .windows import Window, BandpowerWindow, TopHatWindow, LogTopHatWindow  # n
 from .data_types import standard_types, parse_data_type_name, build_data_type_name  # noqa
 from .tracers import BaseTracer  # noqa
 from .covariance import BaseCovariance  # noqa
-__version__ = '0.16' #noqa
+__version__ = '1.0' #noqa

{sacc-0.16 → sacc-1.0}/sacc/data_types.py

@@ -1,4 +1,3 @@
-import warnings
 from collections import namedtuple
 from astropy.table import Table
 
@@ -298,13 +297,6 @@ class DataPoint:
                                      f"{data_type} "
                                      "(ignore_missing_tags=False)")
 
-        # We encourage people to use existing type names, and issue a
-        # warning if they do not to prod them in the right direction.
-        if data_type not in standard_types:
-            warnings.warn(f"Unknown data_type value {data_type}. "
-                          "If possible use a pre-defined type, or "
-                          "add to the list.")
-
     def __repr__(self):
         t = ", ".join(f'{k}={v}' for (k, v) in self.tags.items())
         st = f"DataPoint(data_type='{self.data_type}', "

{sacc-0.16 → sacc-1.0}/sacc/sacc.py

@@ -371,12 +371,14 @@ class Sacc:
             # Record this index
             if ok:
                 indices.append(i)
+
         if len(indices) == 0 and warn_empty:
             if tracers is None:
                 warnings.warn("Empty index selected")
             else:
                 warnings.warn("Empty index selected - maybe you "
                               "should check the tracer order?")
+
         return np.array(indices, dtype=int)
 
     def remove_selection(self, data_type=None, tracers=None,
@@ -603,6 +605,40 @@ class Sacc:
         indices = self.indices(data_type=data_type, tracers=tracers, **select)
         return self.mean[indices]
 
+    def get_standard_deviation(self, data_type=None, tracers=None, **select):
+        """
+        Get standard deviation values for each data point matching the criteria.
+
+        This requires the covariance matrix to be set.
+
+        Parameters
+        ----------
+
+        data_type: str
+            Select only data points which are of this data type.
+            If None (the default) then match any data types
+
+        tracers: tuple
+            Select only data points which match this tracer combination.
+            If None (the default) then match any tracer combinations.
+
+        **select:
+            Select only data points with tag names and values matching
+            all values provided in this kwargs option.
+            You can also use the syntax name__lt=value or
+            name__gt=value in the selection to select points
+            less or greater than a threshold
+
+        Returns
+        -------
+        values: array
+            The standard deviation values for each matching data point,
+            in the order they were added.
+
+        """
+        indices = self.indices(data_type=data_type, tracers=tracers, **select)
+        return np.sqrt(self.covariance.get_block(indices).diagonal())
+
     def get_data_types(self, tracers=None):
         """
         Get a list of the different data types stored in the Sacc
@@ -796,20 +832,19 @@ class Sacc:
 
         # Since we don't want to re-order the file as a side effect
         # we first make a copy of ourself and re-order that.
-        S = self.copy()
-        S.to_canonical_order()
-
         # Tables for the windows
-        tables, window_ids = S._make_window_tables()
+        tables, window_ids = self._make_window_tables()
         lookup = {'window': window_ids}
 
         # Tables for the tracers
-        tables += BaseTracer.to_tables(S.tracers.values())
+        tables += BaseTracer.to_tables(self.tracers.values())
 
         # Tables for the data sets
-        for dt in S.get_data_types():
-            data = S.get_data_points(dt)
+        for dt in self.get_data_types():
+            indices = self.indices(dt)
+            data = [self.data[i] for i in indices]
             table = DataPoint.to_table(data, lookup)
+            table.add_column(indices, name='sacc_ordering')
             # Could move this inside to_table?
             table.meta['SACCTYPE'] = 'data'
             table.meta['SACCNAME'] = dt
@@ -822,8 +857,8 @@ class Sacc:
         # save any global metadata in the header.
         # We save the keys and values as separate header cards,
         # because otherwise the keys are all forced to upper case
-        hdr['NMETA'] = len(S.metadata)
-        for i, (k, v) in enumerate(S.metadata.items()):
+        hdr['NMETA'] = len(self.metadata)
+        for i, (k, v) in enumerate(self.metadata.items()):
             hdr[f'KEY{i}'] = k
             hdr[f'VAL{i}'] = v
         hdus = [fits.PrimaryHDU(header=hdr)] + \
@@ -833,8 +868,8 @@ class Sacc:
         # All the other data elements become astropy tables first,
         # But covariances are a bit more complicated and dense, so we
         # allow them to convert straight to
-        if S.covariance is not None:
-            hdus.append(S.covariance.to_hdu())
+        if self.covariance is not None:
+            hdus.append(self.covariance.to_hdu())
 
         # Make and save the final FITS data
         hdu_list = fits.HDUList(hdus)
@@ -896,9 +931,18 @@ class Sacc:
         lookup = {'window': windows}
 
         # Collect together all the data points from the different sections
-        data = []
+        data_unordered = []
+        index = []
         for table in data_tables:
-            data += DataPoint.from_table(table, lookup)
+            index += table["sacc_ordering"].tolist()
+            table.remove_column('sacc_ordering')
+            data_unordered += DataPoint.from_table(table, lookup)
+
+        # Put the data back in its original order, matching the
+        # covariance.
+        data = [None for i in range(len(data_unordered))]
+        for i, d in zip(index, data_unordered):
+            data[i] = d
 
         # Finally, take all the pieces that we have collected
         # and add them all into this data set.
@@ -986,7 +1030,7 @@ class Sacc:
                              "later.")
         ws = ws[0]
         if not isinstance(ws, BandpowerWindow):
-            warnings.warn("No bandpower windows associated to these data")
+            warnings.warn("No bandpower windows associated with these data")
             return None
         else:
             w_inds = np.array(self._get_tags_by_index(['window_ind'],

{sacc-0.16 → sacc-1.0}/sacc/tracers.py

@@ -2,7 +2,6 @@ import numpy as np
 from astropy.table import Table
 from .utils import (Namespace, hide_null_values,
                     remove_dict_null_values, unique_list)
-import warnings
 
 standard_quantities = Namespace('galaxy_shear',
                                 'galaxy_density',
@@ -40,10 +39,6 @@ class BaseTracer:
         # We encourage people to use existing quantity names, and issue a
         # warning if they do not to prod them in the right direction.
         quantity = kwargs.pop('quantity', 'generic')
-        if quantity not in standard_quantities:
-            warnings.warn(f"Unknown quantity {quantity}. "
-                          "If possible use a pre-defined quantity, or "
-                          "add to the list.")
         self.name = name
         self.quantity = quantity
         self.metadata = kwargs.pop('metadata', {})

{sacc-0.16 → sacc-1.0/sacc.egg-info}/PKG-INFO

@@ -1,17 +1,25 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: sacc
-Version: 0.16
+Version: 1.0
 Summary: SACC - the LSST/DESC summary statistic data format library
-Home-page: https://github.com/LSSTDESC/sacc
-Author: LSST DESC
-Author-email: joezuntz@googlemail.com
-License: UNKNOWN
-Platform: UNKNOWN
+Author-email: LSST DESC <joezuntz@googlemail.com>
+License: BSD-3-Clause
+Project-URL: Homepage, https://github.com/LSSTDESC/sacc
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
 Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: scipy
+Requires-Dist: numpy>=1.20
+Requires-Dist: astropy
 Provides-Extra: all
+Requires-Dist: qp-prob[all]; extra == "all"
+Requires-Dist: numpydoc; extra == "all"
 Provides-Extra: doc
+Requires-Dist: numpydoc; extra == "doc"
 Provides-Extra: qp
-License-File: LICENSE
+Requires-Dist: qp-prob[all]; extra == "qp"
+Dynamic: license-file
 
 Sacc
 ====
@@ -58,5 +66,3 @@ Citation
 Sacc has been submitted to the [Astrophysics Source Code Library](https://ascl.net/code/v/2277); follow the link that will appear there to NASA ADS to export a bibtex citation.
 
 The core developers of Sacc are Joe Zuntz (maintainer), David Alonso, and Matt Becker.
-
-

{sacc-0.16 → sacc-1.0}/sacc.egg-info/SOURCES.txt

@@ -5,8 +5,8 @@
 LICENSE
 MANIFEST.in
 README.md
+pyproject.toml
 requirements.txt
-setup.py
 .github/workflows/desc-ci.yml
 .github/workflows/publish.yml
 doc/Makefile
@@ -47,7 +47,8 @@ sacc.egg-info/SOURCES.txt
 sacc.egg-info/dependency_links.txt
 sacc.egg-info/requires.txt
 sacc.egg-info/top_level.txt
+test/__init__.py
 test/make_test_data.py
 test/test_cluster_data_tracers.py
-test/test_sacc2.py
+test/test_sacc.py
 test/data/.gitignore

sacc-0.16/test/test_sacc2.py → sacc-1.0/test/test_sacc.py

@@ -8,7 +8,11 @@ import os
 import pathlib
 import urllib
 import time
-import qp
+import warnings
+try:
+    import qp
+except:
+    pass
 
 test_dir = pathlib.Path(__file__).resolve().parent
 test_data_dir = test_dir / 'data'
@@ -79,20 +83,58 @@ def test_add_covariance():
     s.add_covariance(0 * cov, overwrite=True)
     assert np.all(s.covariance.covmat == 0 * cov)
 
+def test_get_sigma():
+    s_base = get_filled_sacc()
+    n = len(s_base)
+
+    # first check it works with a diagonal cov
+    # for the full matrix
+    s = s_base.copy()
+    variance = np.random.uniform(size=n)
+    s.add_covariance(variance)
+    assert isinstance(s.covariance, sacc.covariance.DiagonalCovariance)
+    sigma = s.get_standard_deviation()
+    assert np.allclose(sigma, variance ** 0.5)
+
+    # Now check the last 20 data points work
+    tracers = ('source_2', 'source_2')
+    dt = sacc.standard_types.galaxy_shear_cl_bb
+    sigma = s.get_standard_deviation(tracers=tracers, data_type=dt)
+    assert len(sigma) == 20
+    assert np.allclose(sigma, variance[-20:] ** 0.5)
+
+    # Now check with a dense covariance
+    s = s_base.copy()
+    cov = np.random.uniform(size=(n, n))
+    # make symmetric out of habit
+    cov = (cov + cov.T) / 2
+    s.add_covariance(cov)
+    assert isinstance(s.covariance, sacc.covariance.FullCovariance)
+
+    sigma = s.get_standard_deviation()
+    assert np.allclose(sigma, np.sqrt(np.diagonal(cov)))
+
+    tracers = ('source_2', 'source_2')
+    dt = sacc.standard_types.galaxy_shear_cl_bb
+    sigma = s.get_standard_deviation(tracers=tracers, data_type=dt)
+    assert len(sigma) == 20
+    assert np.allclose(sigma, np.sqrt(np.diagonal(cov))[-20:])
+
+    # now block-diagonal
+    s = s_base.copy()
+    cov = [
+        np.random.uniform(size=(n//4, n//4))
+        for i in range(4)
+    ]
+    s.add_covariance(cov)
+    assert isinstance(s.covariance, sacc.covariance.BlockDiagonalCovariance)
+    sigma = s.get_standard_deviation()
+    expected = np.concatenate([
+        np.sqrt(np.diagonal(c))
+        for c in cov
+    ])
+    assert np.allclose(sigma, expected)
 
-def test_quantity_warning():
-    s = sacc.Sacc()
-    with pytest.warns(UserWarning):
-        s.add_tracer('Misc', 'source_0',
-                     quantity='dummy')
-
-
-def test_data_type_warning():
-    s = sacc.Sacc()
-    s.add_tracer('Misc', 'source_0')
-    with pytest.warns(UserWarning):
-        s.add_data_point('cl_wrong', ('source_0', 'source_0'),
-                         0.1, ell=10.)
 
 
 def test_get_data_types():
@@ -891,3 +933,46 @@ def test_sacc_has_tracer():
     assert not s.has_tracer("this_is_not_a_tracer")
     for tracer_name in ['source_0', 'source_1', 'source_2']:
         assert s.has_tracer(tracer_name)
+
+def test_save_order_maintained():
+    s = sacc.Sacc()
+
+    # Tracer
+    s.add_tracer('misc', 'source_0')
+
+    # add a series of data points with alternating
+    # types so they will be saved in separate table
+    s.add_data_point("dt1", ('source_0',), 0.1, a=1)
+    s.add_data_point("dt2", ('source_0',), 0.1, b=2)
+    s.add_data_point("dt1", ('source_0',), 0.1, a=-1)
+    s.add_data_point("dt2", ('source_0',), 0.1, b=-2)
+
+    with tempfile.TemporaryDirectory() as tmpdir:
+        filename = os.path.join(tmpdir, 'test.sacc')
+        s.save_fits(filename)
+        s2 = sacc.Sacc.load_fits(filename)
+
+    # check that the order of the data points is maintained
+    assert len(s2) == 4
+    assert s2.data[0].data_type == "dt1"
+    assert s2.data[1].data_type == "dt2"
+    assert s2.data[2].data_type == "dt1"
+    assert s2.data[3].data_type == "dt2"
+    # and that the tags are all okay
+    assert s2.data[0].get_tag('a') == 1
+    assert s2.data[1].get_tag('b') == 2
+    assert s2.data[2].get_tag('a') == -1
+    assert s2.data[3].get_tag('b') == -2
+
+def test_warn_empty():
+    s = sacc.Sacc()
+
+    # First check that no warning is raised if warn_empty is False
+    # which is the default
+    with warnings.catch_warnings():
+        warnings.simplefilter("error")
+        s.indices(data_type='non_existent_data_type', warn_empty=False)
+
+    # Now check that a warning is raised if warn_empty is True
+    with pytest.warns(UserWarning, match="Empty index selected"):
+        s.indices(data_type='non_existent_data_type', warn_empty=True)

sacc-0.16/setup.py

@@ -1,31 +0,0 @@
-#!/usr/bin/env python
-from setuptools import setup
-
-description = "SACC - the LSST/DESC summary statistic data format library"
-
-with open('requirements.txt') as f:
-    requirements = f.read().splitlines()
-
-requirements.remove("qp-prob")
-requirements.remove("numpydoc")
-
-with open('README.md') as f:
-    long_description = f.read()
-
-setup(name="sacc", 
-      description=description,
-      long_description=long_description,
-      long_description_content_type='text/markdown',
-      url="https://github.com/LSSTDESC/sacc",
-      author="LSST DESC",
-      author_email="joezuntz@googlemail.com",
-      use_scm_version=True,
-      setup_requires=["setuptools_scm<7.0", "setuptools_scm_git_archive"],
-      install_requires=requirements,
-      extras_require = {
-        'all':  ["qp-prob[all]", "numpydoc"],
-        'doc': ["numpydoc"],
-        'qp':  ["qp-prob[all]"],
-      },
-      packages=['sacc'],
-)

{sacc-0.16 → sacc-1.0}/sacc.egg-info/requires.txt

@@ -1,10 +1,10 @@
-astropy
-numpy>=1.20
 scipy
+numpy>=1.20
+astropy
 
 [all]
-numpydoc
 qp-prob[all]
+numpydoc
 
 [doc]
 numpydoc