Removed the colon-parsing and replaced it with dicts. (#81)

* [skip ci] Removed the colon-parsing and replaced it with dicts. Tests will fail because they still assume you can use colons.

* All tests should now pass.

* Added tests for the file/tree selection features.

* Implemented conditional colon-parsing and full error message as described here: #79 (comment)

* Fixed flake8 error.

tests/test_0016-interpretations.py

@@ -81,7 +81,7 @@ def test_recovery(mini):
     # flat array to recover:
     filename = skhep_testdata.data_path("uproot-issue21.root")
     with uproot4.open(
-        "file:" + filename + " : nllscan/mH", minimal_ttree_metadata=mini
+        {"file:" + filename: "nllscan/mH"}, minimal_ttree_metadata=mini
     ) as branch:
         basket = branch.basket(0)
         assert basket.data.view(">f8").tolist()[:10] == [
@@ -107,7 +107,7 @@ def test_recovery(mini):
     # uproot-from-geant4.root Details: numgood, TrackedRays: Event phi
     filename = skhep_testdata.data_path("uproot-issue327.root")
     with uproot4.open(
-        "file:" + filename + " : DstTree/fTracks.fCharge", minimal_ttree_metadata=mini
+        {"file:" + filename: "DstTree/fTracks.fCharge"}, minimal_ttree_metadata=mini
     ) as branch:
         basket = branch.basket(0)
         assert basket.data.view("i1")[:10].tolist() == [

tests/test_0043-iterate-function.py

@@ -159,14 +159,13 @@ def test_iterate_report_2():
 
 
 def test_function_iterate():
-    files = (
-        skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
-            "6.20.04", "*"
-        )
-        + ":sample"
+    files = skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
+        "6.20.04", "*"
     )
     expect = 0
-    for arrays, report in uproot4.iterate(files, "i8", report=True, library="np"):
+    for arrays, report in uproot4.iterate(
+        {files: "sample"}, "i8", report=True, library="np"
+    ):
         assert arrays["i8"][:5].tolist() == [-15, -14, -13, -12, -11]
         assert report.global_entry_start == expect
         assert report.global_entry_stop == expect + len(arrays["i8"])
@@ -175,14 +174,13 @@ def test_function_iterate():
 
 def test_function_iterate_pandas():
     pandas = pytest.importorskip("pandas")
-    files = (
-        skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
-            "6.20.04", "*"
-        )
-        + ":sample"
+    files = skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
+        "6.20.04", "*"
     )
     expect = 0
-    for arrays, report in uproot4.iterate(files, "i8", report=True, library="pd"):
+    for arrays, report in uproot4.iterate(
+        {files: "sample"}, "i8", report=True, library="pd"
+    ):
         assert arrays["i8"].values[:5].tolist() == [-15, -14, -13, -12, -11]
         assert arrays.index.values[0] == expect
         assert report.global_entry_start == expect
@@ -192,13 +190,12 @@ def test_function_iterate_pandas():
 
 def test_function_iterate_pandas_2():
     pandas = pytest.importorskip("pandas")
-    files = (
-        skhep_testdata.data_path("uproot-HZZ.root").replace(
-            "HZZ", "HZZ-{uncompressed,zlib,lz4}"
-        )
-        + ":events"
+    files = skhep_testdata.data_path("uproot-HZZ.root").replace(
+        "HZZ", "HZZ-{uncompressed,zlib,lz4}"
     )
     expect = 0
-    for arrays, report in uproot4.iterate(files, "Muon_Px", report=True, library="pd"):
+    for arrays, report in uproot4.iterate(
+        {files: "events"}, "Muon_Px", report=True, library="pd"
+    ):
         assert arrays["Muon_Px"].index.values[0] == (expect, 0)
         expect += report.tree.num_entries

tests/test_0044-concatenate-function.py

@@ -10,40 +10,31 @@
 
 
 def test_concatenate_numpy():
-    files = (
-        skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
-            "6.20.04", "*"
-        )
-        + ":sample"
+    files = skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
+        "6.20.04", "*"
     )
-    arrays = uproot4.concatenate(files, ["i8", "f8"], library="np")
+    arrays = uproot4.concatenate({files: "sample"}, ["i8", "f8"], library="np")
     assert len(arrays["i8"]) == 420
     assert len(arrays["f8"]) == 420
 
 
 def test_concatenate_awkward():
     awkward1 = pytest.importorskip("awkward1")
-    files = (
-        skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
-            "6.20.04", "*"
-        )
-        + ":sample"
+    files = skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
+        "6.20.04", "*"
     )
-    arrays = uproot4.concatenate(files, ["i8", "f8"], library="ak")
+    arrays = uproot4.concatenate({files: "sample"}, ["i8", "f8"], library="ak")
     assert isinstance(arrays, awkward1.Array)
     assert set(awkward1.keys(arrays)) == set(["i8", "f8"])
     assert len(arrays) == 420
 
 
 def test_concatenate_pandas():
     pandas = pytest.importorskip("pandas")
-    files = (
-        skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
-            "6.20.04", "*"
-        )
-        + ":sample"
+    files = skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
+        "6.20.04", "*"
     )
-    arrays = uproot4.concatenate(files, ["i8", "f8"], library="pd")
+    arrays = uproot4.concatenate({files: "sample"}, ["i8", "f8"], library="pd")
     assert isinstance(arrays, pandas.DataFrame)
     assert set(arrays.columns.tolist()) == set(["i8", "f8"])
     assert len(arrays) == 420

tests/test_0045-lazy-arrays-1.py

@@ -71,7 +71,7 @@ def test_branch_pluralization():
                 assert False
 
     for i, arrays in enumerate(
-        uproot4.iterate(skhep_testdata.data_path("uproot-Zmumu.root") + ":events/px1")
+        uproot4.iterate({skhep_testdata.data_path("uproot-Zmumu.root"): "events/px1"})
     ):
         if i == 0:
             assert arrays["px1"][:5].tolist() == [
@@ -103,14 +103,11 @@ def test_branch_pluralization():
 
 def test_awkward():
     awkward1 = pytest.importorskip("awkward1")
-    files = (
-        skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
-            "6.20.04", "*"
-        )
-        + ":sample"
+    files = skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
+        "6.20.04", "*"
     )
     cache = {}
-    array = uproot4.lazy(files, array_cache=cache)
+    array = uproot4.lazy({files: "sample"}, array_cache=cache)
     assert len(cache) == 0
 
     assert awkward1.to_list(array[:5, "i4"]) == [-15, -14, -13, -12, -11]
@@ -146,10 +143,8 @@ def test_awkward():
 
 def test_awkward_pluralization():
     awkward1 = pytest.importorskip("awkward1")
-    files = (
-        skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
-            "6.20.04", "*"
-        )
-        + ":sample/i4"
+    files = skhep_testdata.data_path("uproot-sample-6.20.04-uncompressed.root").replace(
+        "6.20.04", "*"
     )
-    assert awkward1.to_list(uproot4.lazy(files)[:5, "i4"]) == [-15, -14, -13, -12, -11]
+    array = uproot4.lazy({files: "sample"})
+    assert awkward1.to_list(array[:5, "i4"]) == [-15, -14, -13, -12, -11]

tests/test_0066-fix-http-fallback-freeze.py

@@ -10,6 +10,8 @@
 
 @pytest.mark.network
 def test():
-    with uproot4.open("http://scikit-hep.org/uproot/examples/HZZ.root:events") as t:
+    with uproot4.open(
+        {"http://scikit-hep.org/uproot/examples/HZZ.root": "events"}
+    ) as t:
         t["MET_px"].array()
         t["MET_py"].array()

tests/test_0081-dont-parse-colons.py

@@ -0,0 +1,140 @@
+# BSD 3-Clause License; see https://github.com/scikit-hep/uproot4/blob/master/LICENSE
+
+from __future__ import absolute_import
+
+import pytest
+import skhep_testdata
+
+import uproot4
+
+
+def test_open():
+    assert isinstance(
+        uproot4.open(skhep_testdata.data_path("uproot-issue63.root")),
+        uproot4.reading.ReadOnlyDirectory,
+    )
+    assert isinstance(
+        uproot4.open(
+            {skhep_testdata.data_path("uproot-issue63.root"): "WtLoop_nominal"}
+        ),
+        uproot4.behaviors.TTree.TTree,
+    )
+
+    with pytest.raises(ValueError):
+        uproot4.open([skhep_testdata.data_path("uproot-issue63.root")])
+
+
+def test_lazy():
+    with pytest.raises(ValueError):
+        uproot4.lazy(skhep_testdata.data_path("uproot-issue63.root"))
+
+    with pytest.raises(ValueError):
+        uproot4.lazy(
+            {skhep_testdata.data_path("uproot-issue63.root"): "blah"},
+            allow_missing=True,
+        )
+
+    uproot4.lazy({skhep_testdata.data_path("uproot-issue63.root"): "WtLoop_nominal"})
+    uproot4.lazy(
+        {
+            skhep_testdata.data_path("uproot-issue63.root"): "WtLoop_nominal",
+            skhep_testdata.data_path("uproot-issue63.root"): "WtLoop_Fake_nominal",
+        }
+    )
+
+    uproot4.lazy([{skhep_testdata.data_path("uproot-issue63.root"): "WtLoop_nominal"}])
+    uproot4.lazy(
+        {skhep_testdata.data_path("uproot-issue63.root") + "*": "WtLoop_nominal"}
+    )
+    uproot4.lazy(
+        [{skhep_testdata.data_path("uproot-issue63.root") + "*": "WtLoop_nominal"}]
+    )
+
+
+def test_concatenate():
+    with pytest.raises(ValueError):
+        uproot4.concatenate(skhep_testdata.data_path("uproot-issue63.root"))
+
+    assert (
+        len(
+            uproot4.concatenate(
+                {skhep_testdata.data_path("uproot-issue63.root"): "blah"},
+                allow_missing=True,
+            )
+        )
+        == 0
+    )
+
+    files = skhep_testdata.data_path("uproot-sample-6.16.00-uncompressed.root").replace(
+        "6.16.00", "*"
+    )
+
+    uproot4.concatenate(files, "Ai8")
+    uproot4.concatenate({files: "sample"}, "Ai8")
+    uproot4.concatenate([files], "Ai8")
+    uproot4.concatenate([{files: "sample"}], "Ai8")
+
+
+def test_iterate():
+    with pytest.raises(ValueError):
+        for arrays in uproot4.iterate(skhep_testdata.data_path("uproot-issue63.root")):
+            pass
+
+    assert (
+        len(
+            list(
+                uproot4.iterate(
+                    {skhep_testdata.data_path("uproot-issue63.root"): "blah"},
+                    allow_missing=True,
+                )
+            )
+        )
+        == 0
+    )
+
+    files = skhep_testdata.data_path("uproot-sample-6.16.00-uncompressed.root").replace(
+        "6.16.00", "*"
+    )
+
+    for arrays in uproot4.iterate(files, "Ai8"):
+        pass
+    for arrays in uproot4.iterate({files: "sample"}, "Ai8"):
+        pass
+    for arrays in uproot4.iterate([files], "Ai8"):
+        pass
+    for arrays in uproot4.iterate([{files: "sample"}], "Ai8"):
+        pass
+
+
+pathlib = pytest.importorskip("pathlib")
+
+
+def test_open_colon():
+    assert isinstance(
+        uproot4.open(
+            skhep_testdata.data_path("uproot-issue63.root") + ":WtLoop_nominal"
+        ),
+        uproot4.behaviors.TTree.TTree,
+    )
+
+    with pytest.raises(FileNotFoundError):
+        uproot4.open(
+            pathlib.Path(
+                skhep_testdata.data_path("uproot-issue63.root") + ":WtLoop_nominal"
+            )
+        )
+
+    with pytest.raises(FileNotFoundError):
+        uproot4.open(
+            {skhep_testdata.data_path("uproot-issue63.root") + ":WtLoop_nominal": None}
+        )
+
+
+def test_lazy_colon():
+    uproot4.lazy(skhep_testdata.data_path("uproot-issue63.root") + ":WtLoop_nominal")
+    uproot4.lazy(
+        [
+            skhep_testdata.data_path("uproot-issue63.root") + ":WtLoop_nominal",
+            skhep_testdata.data_path("uproot-issue63.root") + ":WtLoop_Fake_nominal",
+        ]
+    )

uproot4/_util.py

@@ -226,6 +226,52 @@ def file_path_to_source_class(file_path, options):
         raise ValueError("URI scheme not recognized: {0}".format(file_path))
 
 
+if isinstance(__builtins__, dict):
+    if "FileNotFoundError" in __builtins__:
+        _FileNotFoundError = __builtins__["FileNotFoundError"]
+    else:
+        _FileNotFoundError = __builtins__["IOError"]
+else:
+    if hasattr(__builtins__, "FileNotFoundError"):
+        _FileNotFoundError = __builtins__.FileNotFoundError
+    else:
+        _FileNotFoundError = __builtins__.IOError
+
+
+def _file_not_found(files, message=None):
+    if message is None:
+        message = ""
+    else:
+        message = " (" + message + ")"
+
+    return _FileNotFoundError(
+        """file not found{0}
+
+    {1}
+
+Files may be specified as:
+   * str/bytes: relative or absolute filesystem path or URL, without any colons
+         other than Windows drive letter or URL schema.
+         Examples: "rel/file.root", "C:\\abs\\file.root", "http://where/what.root"
+   * str/bytes: same with an object-within-ROOT path, separated by a colon.
+         Example: "rel/file.root:tdirectory/ttree"
+   * pathlib.Path: always interpreted as a filesystem path or URL only (no
+         object-within-ROOT path), regardless of whether there are any colons.
+         Examples: Path("rel:/file.root"), Path("/abs/path:stuff.root")
+
+Functions that accept many files (uproot4.iterate, etc.) also allow:
+   * glob syntax in str/bytes and pathlib.Path.
+         Examples: Path("rel/*.root"), "/abs/*.root:tdirectory/ttree"
+   * dict: keys are filesystem paths, values are objects-within-ROOT paths.
+         Example: {{"/data_v1/*.root": "ttree_v1", "/data_v2/*.root": "ttree_v2"}}
+   * already-open TTree objects.
+   * iterables of the above.
+""".format(
+            message, repr(files)
+        )
+    )
+
+
 def memory_size(data, error_message=None):
     """
     Regularizes strings like '## kB' and plain integer number of bytes to