Centralize 'from_module' checking and check the whole mro. (#557)

src/uproot/_util.py

@@ -90,6 +90,27 @@ def ensure_numpy(array, types=(numpy.bool_, numpy.integer, numpy.floating)):
 _regularize_filter_regex = re.compile("^/(.*)/([iLmsux]*)$")
 
 
+def from_module(obj, module_name):
+    """
+    Returns True if ``obj`` is an instance of a class from a module
+    given by name.
+
+    This is like ``isinstance`` (in that it searches the whole ``mro``),
+    except that the module providing the type to check against doesn't
+    have to be imported and doesn't get imported (as a side effect) by
+    this function.
+    """
+    try:
+        mro = type(obj).mro()
+    except TypeError:
+        return False
+
+    for t in mro:
+        if t.__module__ == module_name or t.__module__.startswith(module_name + "."):
+            return True
+    return False
+
+
 def _regularize_filter_regex_flags(flags):
     flagsbyte = 0
     for flag in flags:

src/uproot/interpretation/objects.py

@@ -205,9 +205,7 @@ def final_array(
 
             start = stop
 
-        if all(
-            type(x).__module__.startswith("awkward") for x in basket_arrays.values()
-        ):
+        if all(uproot._util.from_module(x, "awkward") for x in basket_arrays.values()):
             assert isinstance(library, uproot.interpretation.library.Awkward)
             awkward = library.imported
             output = awkward.concatenate(trimmed, mergebool=False, highlevel=False)

src/uproot/language/python.py

@@ -455,8 +455,7 @@ def getter(name):
                     file_path,
                     object_path,
                 )()
-                module_name = type(output[expression]).__module__
-                if module_name == "pandas" or module_name.startswith("pandas."):
+                if uproot._util.from_module(output[expression], "pandas"):
                     is_pandas = True
 
         cut = None
@@ -472,8 +471,7 @@ def getter(name):
                     file_path,
                     object_path,
                 )()
-                module_name = type(cut).__module__
-                if module_name == "pandas" or module_name.startswith("pandas."):
+                if uproot._util.from_module(cut, "pandas"):
                     is_pandas = True
                 break
 

src/uproot/source/chunk.py

@@ -28,6 +28,7 @@ class Resource:
     :doc:`uproot.source.futures.ResourceFuture`.
     """
 
+    @property
     def file_path(self):
         """
         A path to the file (or URL).

src/uproot/writing/_cascadetree.py

@@ -116,7 +116,7 @@ def __init__(
 
             else:
                 try:
-                    if type(branch_type).__module__.startswith("awkward."):
+                    if uproot._util.from_module(branch_type, "awkward"):
                         raise TypeError
                     if (
                         uproot._util.isstr(branch_type)
@@ -509,15 +509,14 @@ def extend(self, file, sink, data):
             sink.flush()
 
         provided = None
-        module_name = type(data).__module__
 
-        if module_name == "pandas" or module_name.startswith("pandas."):
+        if uproot._util.from_module(data, "pandas"):
             import pandas
 
             if isinstance(data, pandas.DataFrame) and data.index.is_numeric():
                 provided = dataframe_to_dict(data)
 
-        if module_name == "awkward" or module_name.startswith("awkward."):
+        if uproot._util.from_module(data, "awkward"):
             try:
                 awkward = uproot.extras.awkward()
             except ModuleNotFoundError as err:
@@ -550,12 +549,9 @@ def extend(self, file, sink, data):
 
             provided = {}
             for k, v in data.items():
-                module_name = type(v).__module__
-                if not (
-                    module_name == "pandas" or module_name.startswith("pandas.")
-                ) and not (
-                    module_name == "awkward" or module_name.startswith("awkward.")
-                ):
+                if not uproot._util.from_module(
+                    v, "pandas"
+                ) and not uproot._util.from_module(v, "awkward"):
                     if not hasattr(v, "dtype") and not isinstance(v, Mapping):
                         try:
                             with warnings.catch_warnings():
@@ -583,8 +579,7 @@ def extend(self, file, sink, data):
                             ) from err
                         v = awkward.from_iter(v)
 
-                module_name = type(v).__module__
-                if module_name == "awkward" or module_name.startswith("awkward."):
+                if uproot._util.from_module(v, "awkward"):
                     try:
                         awkward = uproot.extras.awkward()
                     except ModuleNotFoundError as err:
@@ -622,8 +617,7 @@ def extend(self, file, sink, data):
                 if datum["name"] in provided:
                     recordarray = provided.pop(datum["name"])
 
-                    module_name = type(recordarray).__module__
-                    if module_name == "pandas" or module_name.startswith("pandas."):
+                    if uproot._util.from_module(recordarray, "pandas"):
                         import pandas
 
                         if isinstance(recordarray, pandas.DataFrame):

src/uproot/writing/identify.py

@@ -48,15 +48,14 @@ def add_to_directory(obj, name, directory, streamers):
     Raises ``TypeError`` if ``obj`` is not recognized as writable data.
     """
     is_ttree = False
-    module_name = type(obj).__module__
 
-    if module_name == "pandas" or module_name.startswith("pandas."):
+    if uproot._util.from_module(obj, "pandas"):
         import pandas
 
         if isinstance(obj, pandas.DataFrame) and obj.index.is_numeric():
             obj = uproot.writing._cascadetree.dataframe_to_dict(obj)
 
-    if module_name == "awkward" or module_name.startswith("awkward."):
+    if uproot._util.from_module(obj, "awkward"):
         import awkward
 
         if isinstance(obj, awkward.Array):
@@ -70,9 +69,7 @@ def add_to_directory(obj, name, directory, streamers):
         metadata = {}
 
         for branch_name, branch_array in obj.items():
-            module_name = type(branch_array).__module__
-
-            if module_name == "pandas" or module_name.startswith("pandas."):
+            if uproot._util.from_module(branch_array, "pandas"):
                 import pandas
 
                 if isinstance(branch_array, pandas.DataFrame):
@@ -113,8 +110,7 @@ def add_to_directory(obj, name, directory, streamers):
                 metadata[branch_name] = metadatum
 
             else:
-                module_name = type(branch_array).__module__
-                if module_name == "awkward" or module_name.startswith("awkward."):
+                if uproot._util.from_module(branch_array, "awkward"):
                     data[branch_name] = branch_array
                     metadata[branch_name] = branch_array.type
 

tests/test_0001-source-class.py

@@ -94,6 +94,7 @@ def test_memmap_fail(tmpdir):
         uproot.source.file.MultithreadedFileSource(filename + "-does-not-exist")
 
 
+@pytest.mark.skip(reason="RECHECK: example.com is flaky, too")
 @pytest.mark.network
 def test_http():
     source = uproot.source.http.HTTPSource(
@@ -117,6 +118,7 @@ def test_http():
         assert [tobytes(x.raw_data) for x in chunks] == [one, two, three]
 
 
+@pytest.mark.skip(reason="RECHECK: example.com is flaky, too")
 def colons_and_ports():
     assert uproot._util.file_object_path_split("https://example.com:443") == (
         "https://example.com:443",
@@ -131,6 +133,7 @@ def colons_and_ports():
     ) == ("https://example.com:443/something", "else")
 
 
+@pytest.mark.skip(reason="RECHECK: example.com is flaky, too")
 @pytest.mark.network
 def test_http_port():
     source = uproot.source.http.HTTPSource(

tests/test_0006-notify-when-downloaded.py

@@ -63,6 +63,7 @@ def test_memmap(tmpdir):
             expected.pop((chunk.start, chunk.stop))
 
 
+@pytest.mark.skip(reason="RECHECK: example.com is flaky, too")
 @pytest.mark.network
 def test_http_multipart():
     notifications = queue.Queue()
@@ -78,6 +79,7 @@ def test_http_multipart():
             expected.pop((chunk.start, chunk.stop))
 
 
+@pytest.mark.skip(reason="RECHECK: example.com is flaky, too")
 @pytest.mark.network
 def test_http():
     notifications = queue.Queue()
@@ -93,6 +95,7 @@ def test_http():
             expected.pop((chunk.start, chunk.stop))
 
 
+@pytest.mark.skip(reason="RECHECK: example.com is flaky, too")
 @pytest.mark.network
 def test_http_workers():
     notifications = queue.Queue()

tests/test_0007-single-chunk-interface.py

@@ -74,6 +74,7 @@ def test_memmap(tmpdir):
         )
 
 
+@pytest.mark.skip(reason="RECHECK: example.com is flaky, too")
 @pytest.mark.network
 def test_http():
     for num_workers in [1, 2]:
@@ -84,15 +85,20 @@ def test_http():
                 chunk = source.chunk(start, stop)
                 assert len(tobytes(chunk.raw_data)) == stop - start
 
-            with pytest.raises(Exception):
-                with uproot.source.http.MultithreadedHTTPSource(
-                    "https://wonky.cern/does-not-exist",
-                    num_workers=num_workers,
-                    timeout=0.1,
-                ) as source:
-                    source.chunk(0, 100)
+
+@pytest.mark.network
+def test_http_fail():
+    for num_workers in [1, 2]:
+        with pytest.raises(Exception):
+            with uproot.source.http.MultithreadedHTTPSource(
+                "https://wonky.cern/does-not-exist",
+                num_workers=num_workers,
+                timeout=0.1,
+            ) as source:
+                source.chunk(0, 100)
 
 
+@pytest.mark.skip(reason="RECHECK: example.com is flaky, too")
 @pytest.mark.network
 def test_http_multipart():
     with uproot.source.http.HTTPSource(
@@ -102,11 +108,14 @@ def test_http_multipart():
             chunk = source.chunk(start, stop)
             assert len(tobytes(chunk.raw_data)) == stop - start
 
-        with pytest.raises(Exception):
-            with uproot.source.http.HTTPSource(
-                "https://wonky.cern/does-not-exist", timeout=0.1, num_fallback_workers=1
-            ) as source:
-                tobytes(source.chunk(0, 100).raw_data)
+
+@pytest.mark.network
+def test_http_multipart_fail():
+    with pytest.raises(Exception):
+        with uproot.source.http.HTTPSource(
+            "https://wonky.cern/does-not-exist", timeout=0.1, num_fallback_workers=1
+        ) as source:
+            tobytes(source.chunk(0, 100).raw_data)
 
 
 @pytest.mark.skip(