Run doctests via pytest and fix broken doctests.

changes/4000.misc.md

@@ -0,0 +1 @@
+Run all doctests via pytest and fix all broken doctests.

pyproject.toml

@@ -414,7 +414,7 @@ ignore_errors = true
 
 [tool.pytest.ini_options]
 minversion = "7"
-testpaths = ["tests", "docs/user-guide"]
+testpaths = ["src", "tests", "docs/user-guide"]
 log_cli_level = "INFO"
 log_level = "INFO"
 xfail_strict = true
@@ -430,6 +430,9 @@ addopts = [
     "--benchmark-disable", # benchmark routines run as tests without benchmarking instrumentation
     "--durations", "10",
     "-ra", "--strict-config", "--strict-markers",
+    "--doctest-modules",
+    "--ignore=tests/test_regression/scripts",
+    "--ignore=src/zarr/_cli",
 ]
 filterwarnings = [
     "error",

src/zarr/abc/store.py

@@ -256,12 +256,20 @@ async def _get_bytes(
 
         Examples
         --------
-        >>> store = await MemoryStore.open()
-        >>> await store.set("data", Buffer.from_bytes(b"hello world"))
-        >>> data = await store.get_bytes("data", prototype=default_buffer_prototype())
-        >>> print(data)
+        >>> async def example():
+        ...     from zarr.core.buffer.cpu import Buffer
+        ...     from zarr.storage import MemoryStore
+        ...
+        ...     store = await MemoryStore.open()
+        ...     await store.set("data", Buffer.from_bytes(b"hello world"))
+        ...     # No need to specify prototype for MemoryStore
+        ...     return await store._get_bytes("data")
+
+        >>> import asyncio
+        >>> asyncio.run(example())
         b'hello world'
         """
+
         buffer = await self.get(key, prototype, byte_range)
         if buffer is None:
             raise FileNotFoundError(key)
@@ -309,10 +317,11 @@ def _get_bytes_sync(
 
         Examples
         --------
+        >>> from zarr.core.buffer.cpu import Buffer
+        >>> from zarr.storage import MemoryStore
         >>> store = MemoryStore()
-        >>> await store.set("data", Buffer.from_bytes(b"hello world"))
-        >>> data = store.get_bytes_sync("data", prototype=default_buffer_prototype())
-        >>> print(data)
+        >>> store.set_sync("data", Buffer.from_bytes(b"hello world"))
+        >>> store._get_bytes_sync("data")  # No need to specify prototype for MemoryStore
         b'hello world'
         """
 
@@ -358,11 +367,18 @@ async def _get_json(
 
         Examples
         --------
-        >>> store = await MemoryStore.open()
-        >>> metadata = {"zarr_format": 3, "node_type": "array"}
-        >>> await store.set("zarr.json", Buffer.from_bytes(json.dumps(metadata).encode()))
-        >>> data = await store.get_json("zarr.json", prototype=default_buffer_prototype())
-        >>> print(data)
+        >>> async def example():
+        ...     from zarr.core.buffer.cpu import Buffer
+        ...     from zarr.storage import MemoryStore
+        ...
+        ...     store = await MemoryStore.open()
+        ...     metadata = {"zarr_format": 3, "node_type": "array"}
+        ...     await store.set("zarr.json", Buffer.from_bytes(json.dumps(metadata).encode()))
+        ...     # No need to specify prototype for MemoryStore
+        ...     return await store._get_json("zarr.json")
+
+        >>> import asyncio
+        >>> asyncio.run(example())
         {'zarr_format': 3, 'node_type': 'array'}
         """
 
@@ -414,11 +430,12 @@ def _get_json_sync(
 
         Examples
         --------
+        >>> from zarr.core.buffer.cpu import Buffer
+        >>> from zarr.storage import MemoryStore
         >>> store = MemoryStore()
         >>> metadata = {"zarr_format": 3, "node_type": "array"}
-        >>> store.set("zarr.json", Buffer.from_bytes(json.dumps(metadata).encode()))
-        >>> data = store.get_json_sync("zarr.json", prototype=default_buffer_prototype())
-        >>> print(data)
+        >>> store.set_sync("zarr.json", Buffer.from_bytes(json.dumps(metadata).encode()))
+        >>> store._get_json_sync("zarr.json")  # No need to specify prototype for MemoryStore
         {'zarr_format': 3, 'node_type': 'array'}
         """
 

src/zarr/api/synchronous.py

@@ -1126,63 +1126,45 @@ def from_array(
     --------
     Create an array from an existing Array:
 
-    ```python
-    import zarr
-    store = zarr.storage.MemoryStore()
-    store2 = zarr.storage.LocalStore('example_from_array.zarr')
-    arr = zarr.create_array(
-        store=store,
-        shape=(100,100),
-        chunks=(10,10),
-        dtype='int32',
-        fill_value=0)
-    arr2 = zarr.from_array(store2, data=arr, overwrite=True)
-    # <Array file://example_from_array.zarr shape=(100, 100) dtype=int32>
-    ```
+    >>> import asyncio
+    >>> import zarr
+    >>> store = zarr.storage.LocalStore("example_from_array.zarr")
+    >>> arr = zarr.create_array(
+    ...     store={},
+    ...     shape=(100,100),
+    ...     chunks=(10,10),
+    ...     dtype="int32",
+    ...     fill_value=0
+    ... )
+    >>> arr2 = zarr.from_array(store, data=arr, overwrite=True)
+    >>> arr2
+    <Array file://example_from_array.zarr shape=(100, 100) dtype=int32>
+    >>> asyncio.run(store.clear())  # Remove files generated by test
 
     Create an array from an existing NumPy array:
 
-    ```python
-    import zarr
-    import numpy as np
-    arr3 = zarr.from_array(
-            zarr.storage.MemoryStore(),
-        data=np.arange(10000, dtype='i4').reshape(100, 100),
-    )
-    # <Array memory://... shape=(100, 100) dtype=int32>
-    ```
+    >>> import numpy as np
+    >>> zarr.from_array({}, data=np.arange(10000, dtype="i4").reshape(100, 100))
+    <Array memory://... shape=(100, 100) dtype=int32>
 
     Create an array from any array-like object:
 
-    ```python
-    import zarr
-    arr4 = zarr.from_array(
-        zarr.storage.MemoryStore(),
-        data=[[1, 2], [3, 4]],
-    )
-    # <Array memory://... shape=(2, 2) dtype=int64>
-    arr4[...]
-    # array([[1, 2],[3, 4]])
-    ```
+    >>> arr3 = zarr.from_array({}, data=[[1, 2], [3, 4]])
+    >>> arr3
+    <Array memory://... shape=(2, 2) dtype=int64>
+    >>> arr3[...]
+    array([[1, 2], [3, 4]])
 
     Create an array from an existing Array without copying the data:
 
-    ```python
-    import zarr
-    arr4 = zarr.from_array(
-        zarr.storage.MemoryStore(),
-        data=[[1, 2], [3, 4]],
-    )
-    arr5 = zarr.from_array(
-        zarr.storage.MemoryStore(),
-        data=arr4,
-        write_data=False,
-    )
-    # <Array memory://... shape=(2, 2) dtype=int64>
-    arr5[...]
-    # array([[0, 0],[0, 0]])
-    ```
+    >>> arr4 = zarr.from_array({}, data=[[1, 2], [3, 4]])
+    >>> arr5 = zarr.from_array({}, data=arr4, write_data=False)
+    >>> arr5
+    <Array memory://... shape=(2, 2) dtype=int64>
+    >>> arr5[...]
+    array([[0, 0], [0, 0]])
     """
+
     return Array(
         sync(
             zarr.core.array.from_array(

src/zarr/codecs/numcodecs/_codecs.py

@@ -8,14 +8,16 @@
 import zarr
 import zarr.codecs.numcodecs as numcodecs
 
+store = zarr.storage.MemoryStore()
 array = zarr.create_array(
-  store="data_numcodecs.zarr",
-  shape=(1024, 1024),
-  chunks=(64, 64),
-  dtype="uint32",
-  filters=[numcodecs.Delta(dtype="uint32")],
-  compressors=[numcodecs.BZ2(level=5)],
-  overwrite=True)
+    store=store,
+    shape=(1024, 1024),
+    chunks=(64, 64),
+    dtype="uint32",
+    filters=[numcodecs.Delta(dtype="uint32")],
+    compressors=[numcodecs.BZ2(level=5)],
+    overwrite=True
+)
 array[:] = np.arange(np.prod(array.shape), dtype=array.dtype).reshape(*array.shape)
 ```