Add read_nwb_method for local paths in both hdf5 and zarr (#1994)

* add read_nwb_method * add tests * docstring and tutorial editions * more docstring corrections * CHANGELOG.md * steph suggestions * codespell * Apply suggestions from code review Co-authored-by: Steph Prince <[email protected]> * Update src/pynwb/__init__.py Co-authored-by: Steph Prince <[email protected]> * fix test after Steph commit * add io integration tests to test.py --------- Co-authored-by: Ryan Ly <[email protected]> Co-authored-by: Steph Prince <[email protected]>
NeurodataWithoutBorders · Jan 23, 2025 · 739ee54 · 739ee54
1 parent 3d04646
commit 739ee54
Show file tree

Hide file tree

Showing 8 changed files with 156 additions and 7 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -3,6 +3,7 @@
 ## PyNWB 3.0.0 (Upcoming)
 
 ### Enhancements and minor changes
+- Added `pynwb.read_nwb` convenience method to simplify reading an NWBFile written with any backend @h-mayorquin [#1994](https://github.com/NeurodataWithoutBorders/pynwb/pull/1994)
 - Added support for NWB schema 2.8.0. @rly [#2001](https://github.com/NeurodataWithoutBorders/pynwb/pull/2001)
   - Removed `SpatialSeries.bounds` field that was not functional. This will be fixed in a future release. @rly [#1907](https://github.com/NeurodataWithoutBorders/pynwb/pull/1907), [#1996](https://github.com/NeurodataWithoutBorders/pynwb/pull/1996)
   - Added support for `NWBFile.was_generated_by` field. @stephprince [#1924](https://github.com/NeurodataWithoutBorders/pynwb/pull/1924)

diff --git a/docs/gallery/general/plot_read_basics.py b/docs/gallery/general/plot_read_basics.py
@@ -30,7 +30,6 @@
 import matplotlib.pyplot as plt
 import numpy as np
 
-from pynwb import NWBHDF5IO
 
 ####################
 # We will access NWB data on the `DANDI Archive <https://gui.dandiarchive.org/>`_,
@@ -103,14 +102,17 @@
 # read the data into a :py:class:`~pynwb.file.NWBFile` object.
 
 filepath = "sub-P11HMH_ses-20061101_ecephys+image.nwb"
-# Open the file in read mode "r",
-io = NWBHDF5IO(filepath, mode="r")
-nwbfile = io.read()
+from pynwb import read_nwb
+
+nwbfile = read_nwb(filepath)
 nwbfile
 
 #######################################
-# :py:class:`~pynwb.NWBHDF5IO` can also be used as a context manager:
+# For more advanced use cases, the :py:class:~pynwb.NWBHDF5IO class provides additional functionality. 
+# Below, we demonstrate how :py:class:~pynwb.NWBHDF5IO can be used as a context manager 
+# to read data from an NWB file in a more controlled manner:
 
+from pynwb import NWBHDF5IO
 with NWBHDF5IO(filepath, mode="r") as io2:
     nwbfile2 = io2.read()
 
@@ -291,4 +293,4 @@
 # -----------------------
 # It is good practice, especially on Windows, to close any files that you have opened.
 
-io.close()
+nwbfile.get_read_io().close()
diff --git a/requirements-opt.txt b/requirements-opt.txt
@@ -6,3 +6,6 @@ oaklib==0.5.32
 fsspec==2024.10.0
 requests==2.32.3
 aiohttp==3.10.11
+
+# For read_nwb tests
+hdmf-zarr
diff --git a/src/pynwb/__init__.py b/src/pynwb/__init__.py
@@ -531,6 +531,71 @@ def read_nwb(**kwargs):
 
         return nwbfile
 
+@docval({'name': 'path', 'type': (str, Path), 
+         'doc': 'Path to the NWB file. Can be either a local filesystem path to '
+                'an HDF5 (.nwb) or Zarr (.zarr) file.'}, 
+        is_method=False)
+def read_nwb(**kwargs):
+    """Read an NWB file from a local path.
+
+    High-level interface for reading NWB files. Automatically handles both HDF5 
+    and Zarr formats. For advanced use cases (parallel I/O, custom namespaces), 
+    use NWBHDF5IO or NWBZarrIO.
+
+    See also 
+        * :py:class:`~pynwb.NWBHDF5IO`: Core I/O class for HDF5 files with advanced options.
+        * :py:class:`~hdmf_zarr.nwb.NWBZarrIO`: Core I/O class for Zarr files with advanced options.
+
+    Notes
+        This function uses the following defaults:
+            * Always opens in read-only mode
+            * Automatically loads namespaces
+            * Reads any backend (e.g. HDF5 or Zarr) if there is an IO class available.
+
+        Advanced features requiring direct use of IO classes (e.g. NWBHDF5IO NWBZarrIO) include:
+            * Streaming data from s3
+            * Custom namespace extensions
+            * Parallel I/O with MPI
+            * Custom build managers
+            * Write or append modes
+            * Pre-opened HDF5 file objects or Zarr stores
+            * Remote file access configuration
+ 
+    Example usage reading a local NWB file:
+
+    .. code-block:: python
+
+        from pynwb import read_nwb
+        nwbfile = read_nwb("path/to/file.nwb")    
+
+    :Returns: pynwb.NWBFile The loaded NWB file object.
+    """
+
+    path = popargs('path', kwargs)
+    # HDF5 is always available so we try that first
+    backend_is_hdf5 = NWBHDF5IO.can_read(path=path)
+    if backend_is_hdf5:
+        return NWBHDF5IO.read_nwb(path=path)
+    else:
+        # If hdmf5 zarr is available we try that next
+        try:
+            from hdmf_zarr import NWBZarrIO
+            backend_is_zarr = NWBZarrIO.can_read(path=path)
+            if backend_is_zarr:
+                return NWBZarrIO.read_nwb(path=path) 
+            else:
+                raise ValueError(
+                    f"Unable to read file: '{path}'. The file is not recognized as "
+                    "either a valid HDF5 or Zarr NWB file. Please ensure the file exists and contains valid NWB data."
+                )     
+        except ImportError:
+            raise ValueError(
+                f"Unable to read file: '{path}'. The file is not recognized as an HDF5 NWB file. "
+                "If you are trying to read a Zarr file, please install hdmf-zarr using: pip install hdmf-zarr"
+            )
+
+
+
 from . import io as __io  # noqa: F401,E402
 from .core import NWBContainer, NWBData  # noqa: F401,E402
 from .base import TimeSeries, ProcessingModule  # noqa: F401,E402

diff --git a/test.py b/test.py
@@ -264,6 +264,7 @@ def run_integration_tests(verbose=True):
         logging.info('all classes have integration tests')
 
     run_test_suite("tests/integration/utils", "integration utils tests", verbose=verbose)
+    run_test_suite("tests/integration/io", "integration io tests", verbose=verbose)
 
 
 def clean_up_tests():

diff --git a/tests/integration/hdf5/test_io.py b/tests/integration/hdf5/test_io.py
@@ -613,7 +613,7 @@ def test_read_nwb_method_file(self):
             io.write(self.nwbfile)
 
         import h5py
-        
+
         file = h5py.File(self.path, 'r')
 
         read_nwbfile = NWBHDF5IO.read_nwb(file=file)

diff --git a/tests/integration/io/__init__.py b/tests/integration/io/__init__.py
diff --git a/tests/integration/io/test_read.py b/tests/integration/io/test_read.py
@@ -0,0 +1,77 @@
+from pathlib import Path
+import tempfile
+
+from pynwb import read_nwb
+from pynwb.testing.mock.file import mock_NWBFile
+from pynwb.testing import TestCase
+
+import unittest
+try:
+    from hdmf_zarr import NWBZarrIO  # noqa f401
+    HAVE_NWBZarrIO = True 
+except ImportError:
+    HAVE_NWBZarrIO = False
+
+
+class TestReadNWBMethod(TestCase):
+    """Test suite for the read_nwb function."""
+
+    def setUp(self):
+        self.nwbfile = mock_NWBFile()
+
+    def test_read_nwb_hdf5(self):
+        """Test reading a valid HDF5 NWB file."""
+        from pynwb import NWBHDF5IO
+
+        with tempfile.TemporaryDirectory() as temp_dir:
+            path = Path(temp_dir) / "test.nwb"
+            with NWBHDF5IO(path, 'w') as io:
+                io.write(self.nwbfile)
+
+            read_nwbfile = read_nwb(path=path)
+            self.assertContainerEqual(read_nwbfile, self.nwbfile)
+            read_nwbfile.get_read_io().close()
+
+    @unittest.skipIf(not HAVE_NWBZarrIO, "NWBZarrIO library not available")
+    def test_read_zarr(self):
+        """Test reading a valid Zarr NWB file."""
+        with tempfile.TemporaryDirectory() as temp_dir:
+            path = Path(temp_dir) / "test.zarr"
+            with NWBZarrIO(path, 'w') as io:
+                io.write(self.nwbfile)
+
+            read_nwbfile = read_nwb(path=path)
+            self.assertContainerEqual(read_nwbfile, self.nwbfile)
+            read_nwbfile.get_read_io().close()
+
+    def test_read_zarr_without_hdmf_zarr(self):
+        """Test attempting to read a Zarr file without hdmf_zarr installed."""
+        if HAVE_NWBZarrIO:
+            self.skipTest("hdmf_zarr is installed")
+
+        with tempfile.TemporaryDirectory() as temp_dir:
+            path = Path(temp_dir) / "test.zarr"
+            path.mkdir()  # Create empty directory to simulate Zarr store
+
+            expected_message = (
+                f"Unable to read file: '{path}'. The file is not recognized as an HDF5 NWB file. "
+                "If you are trying to read a Zarr file, please install hdmf-zarr using: pip install hdmf-zarr"
+            )
+
+            with self.assertRaisesWith(ValueError, expected_message):
+                read_nwb(path=path)
+
+    @unittest.skipIf(not HAVE_NWBZarrIO, "NWBZarrIO library not available. Need for correct error message.")
+    def test_read_invalid_file(self):
+        """Test attempting to read a file that exists but is neither HDF5 nor Zarr."""
+        with tempfile.TemporaryDirectory() as temp_dir:
+            path = Path(temp_dir) / "test.txt"
+            path.write_text("Not an NWB file")
+
+            expected_message = (
+                f"Unable to read file: '{path}'. The file is not recognized as either a valid HDF5 or Zarr NWB file. "
+                "Please ensure the file exists and contains valid NWB data."
+            )
+
+            with self.assertRaisesWith(ValueError, expected_message):
+                read_nwb(path=path)