Merge tag '3.34.0' into stable

.readthedocs.yaml

@@ -0,0 +1,14 @@
+version: 2
+
+build:
+   os: ubuntu-20.04
+   tools:
+      python: "3.10"
+
+sphinx:
+   # Path to the shared conf.py file.
+   configuration: doc/conf.py
+
+python:
+   install:
+   - requirements: requirements-readthedocs.txt

CHANGELOG.md

@@ -4,6 +4,26 @@ Change Log
 [next] - TBA
 ------------
 
+[3.34.0] - 2022-12-05
+--------------------
+##### Enhancements
+* Faster normalization ([#6202](../../pull/6202))
+* io.UrlReader: Add support for google drive share urls ([#6201](../../pull/6201))
+* Table: Add methods get_column and set_column ([#6058](../../pull/6058))
+* owfeatureconstructor: raise settings version
+* owfeaturecontructor: move meta to the end of namedtuple
+* Aggregate Columns: Add additional options for selection ([#6056](../../pull/6056))
+
+##### Bugfixes
+* conda-recipe: Remove explicit host numpy pinning ([#6235](../../pull/6235))
+* Data Sampler: Fix crash when requesting an empty sample ([#6208](../../pull/6208))
+* Remove pyqt5 install magic ([#6153](../../pull/6153))
+* stats: Handle empty array ([#6221](../../pull/6221))
+* Preprocess: Fix reporting for remove sparse and impute ([#6212](../../pull/6212))
+* Weighted mean computation in Orange.statistics.util.stats ([#6204](../../pull/6204))
+* owfeatureconstructor: raise settings version
+* owfeaturecontructor: move meta to the end of namedtuple
+
 
 [3.33.0] - 2022-09-30
 --------------------
@@ -1708,8 +1728,9 @@ Change Log
 * Initial version based on Python 1.5.2 and Qt 2.3
 
 
-[next]: https://github.com/biolab/orange3/compare/3.33.0...HEAD
-[3.32.0]: https://github.com/biolab/orange3/compare/3.32.0...3.33.0
+[next]: https://github.com/biolab/orange3/compare/3.34.0...HEAD
+[3.34.0]: https://github.com/biolab/orange3/compare/3.33.0...3.34.0
+[3.33.0]: https://github.com/biolab/orange3/compare/3.32.0...3.33.0
 [3.32.0]: https://github.com/biolab/orange3/compare/3.31.1...3.32.0
 [3.31.1]: https://github.com/biolab/orange3/compare/3.31.0...3.31.1
 [3.31.0]: https://github.com/biolab/orange3/compare/3.30.2...3.31.0

Orange/base.py

@@ -110,23 +110,9 @@ def fit_storage(self, data):
         return self.fit(X, Y, W)
 
     def __call__(self, data, progress_callback=None):
-
-        for cls in type(self).mro():
-            if 'incompatibility_reason' in cls.__dict__:
-                incompatibility_reason = \
-                    self.incompatibility_reason(data.domain)  # pylint: disable=assignment-from-none
-                if incompatibility_reason is not None:
-                    raise ValueError(incompatibility_reason)
-                break
-            if 'check_learner_adequacy' in cls.__dict__:
-                warnings.warn(
-                    "check_learner_adequacy is deprecated and will be removed "
-                    "in upcoming releases. Learners should instead implement "
-                    "the incompatibility_reason method.",
-                    OrangeDeprecationWarning)
-                if not self.check_learner_adequacy(data.domain):
-                    raise ValueError(self.learner_adequacy_err_msg)
-                break
+        reason = self.incompatibility_reason(data.domain)
+        if reason is not None:
+            raise ValueError(reason)
 
         origdomain = data.domain
 
@@ -192,9 +178,6 @@ def active_preprocessors(self):
                 self.preprocessors is not type(self).preprocessors):
             yield from type(self).preprocessors
 
-    def check_learner_adequacy(self, _):
-        return True
-
     # pylint: disable=no-self-use
     def incompatibility_reason(self, _: Domain) -> Optional[str]:
         """Return None if a learner can fit domain or string explaining why it can not."""

Orange/data/domain.py

@@ -69,7 +69,7 @@ def match(var):
                 sourceindex = source.index(sourcevar)
                 if var.is_discrete and var is not sourcevar:
                     mapping = var.get_mapper_from(sourcevar)
-                    return lambda table: mapping(table.get_column_view(sourceindex)[0])
+                    return lambda table: mapping(table.get_column(sourceindex))
                 return source.index(var)
             return var.compute_value  # , which may also be None
 

Orange/data/filter.py

@@ -31,6 +31,12 @@ def __init__(self, negate=False):
     def __call__(self, data):
         return
 
+    def __eq__(self, other):
+        return type(self) is type(other) and self.negate == other.negate
+
+    def __hash__(self):
+        return hash(self.negate)
+
 
 class IsDefined(Filter):
     """
@@ -53,7 +59,7 @@ class IsDefined(Filter):
 
     def __init__(self, columns=None, negate=False):
         super().__init__(negate)
-        self.columns = columns
+        self.columns = tuple(columns) if columns is not None else None
 
     def __call__(self, data):
         if isinstance(data, Instance):
@@ -70,6 +76,12 @@ def __call__(self, data):
             r = np.logical_not(r)
         return data[r]
 
+    def __eq__(self, other):
+        return super().__eq__(other) and self.columns == other.columns
+
+    def __hash__(self):
+        return hash((super().__hash__(), hash(self.columns)))
+
 
 class HasClass(Filter):
     """

Orange/data/io.py

@@ -459,6 +459,7 @@ def _resolve_redirects(self, url):
     def _trim(cls, url):
         URL_TRIMMERS = (
             cls._trim_googlesheet,
+            cls._trim_googledrive,
             cls._trim_dropbox,
         )
         for trim in URL_TRIMMERS:
@@ -488,6 +489,18 @@ def _trim_googlesheet(url):
             url += '&gid=' + sheet
         return url
 
+    @staticmethod
+    def _trim_googledrive(url):
+        parts = urlsplit(url)
+        if not parts.netloc.endswith("drive.google.com"):
+            raise ValueError
+        match = re.match(r'/file/d/(?P<id>[^/]+).*', parts.path)
+        if not match:
+            raise ValueError
+        id_ = match.group("id")
+        parts = parts._replace(path=f"uc?export=download&id={id_}", query=None)
+        return urlunsplit(parts)
+
     @staticmethod
     def _trim_dropbox(url):
         parts = urlsplit(url)

Orange/data/sql/filter.py

@@ -2,6 +2,8 @@
 
 
 class IsDefinedSql(filter.IsDefined):
+    InheritEq = True
+
     def to_sql(self):
         sql = " AND ".join([
             '%s IS NOT NULL' % column

Orange/data/sql/table.py

@@ -347,7 +347,7 @@ def has_weights(self):
         return False
 
     def _compute_basic_stats(self, columns=None,
-                             include_metas=False, compute_var=False):
+                             include_metas=False, compute_variance=False):
         if self.approx_len() > LARGE_TABLE:
             self = self.sample_time(DEFAULT_SAMPLE_TIME)