Merge branch 'devel'

.github/workflows/test.yml

@@ -53,6 +53,10 @@ jobs:
  python-version: '3.10'
  tox_env: pyqt6
  name: PyQt6
+ - os: ubuntu-22.04
+ python-version: '3.11'
+ tox_env: beta
+ name: "Scientific Python nightly wheels"
 
  services:
  postgres:
@@ -96,7 +100,7 @@ jobs:
 
  - name: Skip testing workflows at coverage
  if: |
- matrix.python-version == '3.8' && matrix.tox_env == 'orange-released'
+ matrix.python-version == '3.11' && matrix.tox_env == 'orange-released'
  run: |
  echo 'SKIP_EXAMPLE_WORKFLOWS=1' >> $GITHUB_ENV
 
@@ -107,7 +111,7 @@ jobs:
  ORANGE_TEST_DB_URI: postgres://postgres_user:postgres_password@localhost:5432/postgres_db|mssql://SA:sqlServerPassw0rd@localhost:1433
 
  - name: Upload code coverage
- if: matrix.python-version == '3.8' && matrix.tox_env == 'orange-released'
+ if: matrix.python-version == '3.11' && matrix.tox_env == 'orange-released'
  uses: codecov/codecov-action@v3
  with:
  fail_ci_if_error: true

.github/workflows/translations.yml

@@ -0,0 +1,15 @@
+name: Check translations
+
+on:
+ push:
+ branches:
+ - master
+ pull_request:
+ branches:
+ - master
+
+jobs:
+ test-translations:
+ uses: biolab/orange-ci-cd/.github/workflows/test-translations.yml@master
+ with:
+ package-dir: Orange

Orange/canvas/__main__.py

@@ -15,8 +15,8 @@
 from collections import defaultdict
 from datetime import date
 from urllib.request import urlopen, Request
+from packaging.version import Version
 
-import pkg_resources
 import yaml
 
 from AnyQt.QtGui import QColor, QDesktopServices, QIcon, QPalette
@@ -101,9 +101,8 @@ def run(self):
  self.resultReady.emit(contents)
 
  def compare_versions(latest):
- version = pkg_resources.parse_version
  skipped = settings.value('startup/latest-skipped-version', "", type=str)
- if version(latest) <= version(current) or \
+ if Version(latest) <= Version(current) or \
  latest == skipped:
  return
 
@@ -187,8 +186,6 @@ def pull_notifications():
  if not check_notifs:
  return None
 
- Version = pkg_resources.parse_version
-
  # create settings_dict for notif requirements purposes (read-only)
  spec = canvasconfig.spec + config.spec
  settings_dict = canvasconfig.Settings(defaults=spec, store=settings)
@@ -198,7 +195,7 @@ def pull_notifications():
  if ep.dist is not None]
  installed = defaultdict(lambda: "-1")
  for addon in installed_list:
- installed[addon.project_name] = addon.version
+ installed[addon.name] = addon.version
 
  # get set of already displayed notification IDs, stored in settings["notifications/displayed"]
  displayedIDs = literal_eval(settings.value("notifications/displayed", "set()", str))

Orange/canvas/config.py

@@ -2,6 +2,8 @@
 Orange Canvas Configuration
 
 """
+import pkgutil
+
 import random
 import uuid
 import warnings
@@ -11,15 +13,16 @@
 from typing import Dict, Any, Optional, Iterable, List
 
 import packaging.version
-import pkg_resources
 import requests
 
 from AnyQt.QtGui import (
- QPainter, QFont, QFontMetrics, QColor, QPixmap, QIcon, QGuiApplication
+ QPainter, QFont, QFontMetrics, QColor, QImage, QPixmap, QIcon,
+ QGuiApplication
 )
 from AnyQt.QtCore import Qt, QPoint, QRect, QSettings
 
 from orangecanvas import config as occonfig
+from orangecanvas.config import entry_points, EntryPoint
 from orangecanvas.utils.settings import config_slot
 from orangewidget.workflow import config
 from orangewidget.settings import set_widget_settings_dir_components
@@ -61,6 +64,11 @@
 spec = [config_slot(*t) for t in spec]
 
 
+def _pixmap_from_pkg_data(package, path, format):
+ contents = pkgutil.get_data(package, path)
+ return QPixmap.fromImage(QImage.fromData(contents, format))
+
+
 class Config(config.Config):
  """
  Orange application configuration
@@ -96,17 +104,16 @@ def application_icon():
  """
  Return the main application icon.
  """
- path = pkg_resources.resource_filename(
- __name__, "icons/orange-256.png"
+ return QIcon(
+ _pixmap_from_pkg_data(__package__, "icons/orange-256.png", "png")
  )
- return QIcon(path)
 
  @staticmethod
  def splash_screen():
  splash_n = random.randint(1, 3)
- path = pkg_resources.resource_filename(
- __name__, f"icons/orange-splash-screen-{splash_n:02}.png")
- pm = QPixmap(path)
+ pm = _pixmap_from_pkg_data(
+ __name__, f"icons/orange-splash-screen-{splash_n:02}.png", "png"
+ )
 
  version = Config.ApplicationVersion
  if version:
@@ -138,9 +145,9 @@ def widgets_entry_points():
  # Ensure the 'this' distribution's ep is the first. iter_entry_points
  # yields them in unspecified order.
  all_eps = sorted(
- pkg_resources.iter_entry_points(WIDGETS_ENTRY),
+ entry_points(group=WIDGETS_ENTRY),
  key=lambda ep:
- 0 if ep.dist.project_name.lower() == "orange3" else 1
+ 0 if ep.dist.name.lower() == "orange3" else 1
  )
  return iter(all_eps)
 
@@ -173,18 +180,18 @@ def core_packages():
 
  @staticmethod
  def examples_entry_points():
- # type: () -> Iterable[pkg_resources.EntryPoint]
+ # type: () -> Iterable[EntryPoint]
  """
  Return an iterator over the entry points yielding 'Example Workflows'
  """
  # `iter_entry_points` yields them in unspecified order, so we order
  # them by name. The default is at the beginning, unless another
  # entrypoint precedes it alphabetically (e.g. starting with '!').
- default_ep = pkg_resources.EntryPoint(
- "000-Orange3", "Orange.canvas.workflows",
-  dist=pkg_resources.get_distribution("Orange3"))
+ default_ep = EntryPoint(
+ "000-Orange3", "Orange.canvas.workflows", "orange.widgets.tutorials",
+ )
 
- all_ep = list(pkg_resources.iter_entry_points("orange.widgets.tutorials"))
+ all_ep = list(entry_points(group="orange.widgets.tutorials"))
  all_ep.append(default_ep)
  all_ep.sort(key=lambda x: x.name)
  return iter(all_ep)

Orange/data/io.py

@@ -21,10 +21,10 @@
 import numpy as np
 import pandas as pd
 
-
 import xlrd
 import xlsxwriter
 import openpyxl
+import h5py
 
 from Orange.data import _io, Table, Domain, ContinuousVariable, update_origin
 from Orange.data import Compression, open_compressed, detect_encoding, \
@@ -33,7 +33,6 @@
 
 from Orange.util import flatten
 
-
 # Support values longer than 128K (i.e. text contents features)
 csv.field_size_limit(100*1024*1024)
 
@@ -542,34 +541,56 @@ class GenericHDF5Reader(FileFormat):
 
  def __init__(self, filename):
  super().__init__(filename=filename)
- self.data = None
 
+ self.h5_file = h5py.File(filename)
+
+ self.datasets = {}
+ self._load_group("/", self.h5_file)
+
+ @property
+ def sheets(self) -> List:
+ """List of datasets in the file.
+
+ Returns
+ -------
+ List of dataset paths
+ """
+ return list(self.datasets.keys())
+
+ def select_sheet(self, sheet):
+ """Select dataset to be read
+
+ Parameters
+ ----------
+ sheet : str
+ dataset path
+ """
+ if sheet is None:
+ sheet = self.sheets[0]
+ self.sheet = sheet
+
  def read(self):
- """Processes the data stored in self.data and returns it as an Orange
+ """Process data stored in self.data and returns it as an Orange
  Table object.
 
  Returns
  -------
  table (Orange.Table object): 
  Contains the information of the chosen dataset in the hdf5 file.
-
- Raises
- ------
- Exception: If the self.data variable has not been filled yet.
  """
- if self.data is None:
- raise ValueError("The data has not been loaded correctly")
-
- if self.data.name is not None:
- name = self.data.name.split('/')[-1]
+
+ if self.sheet is not None:
+ name = self.sheet.split('/')[-1]
  else:
  name = "Data"
 
+ data = self.datasets[self.sheet]
+
  # Standard names for the columns of the dataset, can be changed manually
  # in the widget itself
- columns = [str(i) for i in range(len(self.data.shape))]
+ columns = [str(i) for i in range(len(data.shape))]
 
- dataset = np.array(self.data)
+ dataset = np.array(data)
 
  # Indexs are created to keep track of the position of the values in the
  # original data file
@@ -583,3 +604,17 @@ def read(self):
  table = Table.from_numpy(domain=Domain(attributes=attrs), X=df.values)
 
  return table
+
+ def _load_group(self, root, group): 
+ """Recursive procedure that constructs the list of datasets
+ stored in the .hdf5 file. 
+
+ Given a root, iterates over all its children to decide whether 
+ they are a dataset or another group of data.
+ """
+ for name, obj in group.items():
+ path = root + name
+ if isinstance(obj, h5py.Group):
+ self._load_group(path + "/", group[name])
+ elif isinstance(obj, h5py.Dataset):
+ self.datasets[path] = obj

Orange/data/io_base.py

@@ -15,6 +15,7 @@
 from glob import glob
 
 import numpy as np
+import pandas
 
 from Orange.data import Table, Domain, Variable, DiscreteVariable, \
  StringVariable, ContinuousVariable, TimeVariable
@@ -665,7 +666,7 @@ def formatter(cls, var):
  elif var.is_discrete:
  return lambda value: "" if isnan(value) else var.values[int(value)]
  elif var.is_string:
- return lambda value: value
+ return lambda value: "" if pandas.isnull(value) else value
  else:
  return var.repr_val
 

Orange/data/table.py

@@ -1551,18 +1551,47 @@ def has_missing_class(self):
  return bn.anynan(self._Y)
 
  @staticmethod
- def __get_nan_frequency(data):
+ def __get_nan_count(data):
  if data.size == 0:
  return 0
  dense = data if not sp.issparse(data) else data.data
- return np.isnan(dense).sum() / np.prod(data.shape)
+ return np.isnan(dense).sum()
+
+ @classmethod
+ def __get_nan_frequency(cls, data):
+ return cls.__get_nan_count(data) / (np.prod(data.shape) or 1)
+
+ def get_nan_count_attribute(self):
+ return self.__get_nan_count(self.X)
+
+ def get_nan_count_class(self):
+ return self.__get_nan_count(self.Y)
+
+ def get_nan_count_metas(self):
+ if self.metas.dtype != object:
+ return self.__get_nan_count(self.metas)
+
+ data = self.metas
+ if sp.issparse(data):
+ data = data.tocsc()
+
+ count = 0
+ for i, attr in enumerate(self.domain.metas):
+ col = data[:, i]
+ missing = np.isnan(col.astype(float)) \
+ if not isinstance(attr, StringVariable) else data == ""
+ count += np.sum(missing)
+ return count
 
  def get_nan_frequency_attribute(self):
  return self.__get_nan_frequency(self.X)
 
  def get_nan_frequency_class(self):
  return self.__get_nan_frequency(self.Y)
 
+ def get_nan_frequency_metas(self):
+ return self.get_nan_count_metas() / (np.prod(self.metas.shape) or 1)
+
  def checksum(self, include_metas=True):
  # TODO: zlib.adler32 does not work for numpy arrays with dtype object
  # (after pickling and unpickling such arrays, checksum changes)

Orange/data/tests/test_io.py

@@ -122,7 +122,7 @@ def setUp(self):
  self.domain,
  np.array([[1, 0.5], [2, np.nan], [np.nan, 1.0625]]),
  np.array([3, 1, 7]),
- np.array(["foo bar baz".split()]).T
+ np.array([["foo", "bar", np.nan]], dtype=object).T
  )
 
  def test_write_tab(self):
@@ -137,7 +137,7 @@ def test_write_tab(self):
 class\tmeta\t\t
 3\tfoo\ty\t0.500
 1\tbar\tz\t
-7\tbaz\t\t1.06250""".strip())
+7\t\t\t1.06250""".strip())
  finally:
  os.remove(fname)
 
@@ -149,7 +149,8 @@ def test_roundtrip_xlsx(self):
  data = ExcelReader(fname).read()
  np.testing.assert_equal(data.X, self.data.X)
  np.testing.assert_equal(data.Y, self.data.Y)
- np.testing.assert_equal(data.metas, self.data.metas)
+ np.testing.assert_equal(data.metas[:2], self.data.metas[:2])
+ self.assertEqual(data.metas[2, 0], "")
  np.testing.assert_equal(data.domain, self.data.domain)
  finally:
  os.remove(fname)

Orange/data/tests/test_variable.py

@@ -634,6 +634,8 @@ def test_val(self):
  self.assertEqual(a.str_val(""), "?")
  self.assertEqual(a.str_val(Value(a, "")), "?")
  self.assertEqual(a.repr_val(Value(a, "foo")), '"foo"')
+ self.assertEqual(a.str_val(np.nan), "?")
+ self.assertEqual(a.str_val(None), "?")
 
 
 @variabletest(TimeVariable)