New CLI Options for Copy and Extract Info

.gitignore

@@ -25,7 +25,8 @@ share/python-wheels/
 .installed.cfg
 *.egg
 MANIFEST
-
+pip-*
+tmp*
 # PyInstaller
 #  Usually these files are written by a python script from a template
 #  before PyInstaller builds the exe, so as to inject date/other infos into it.
@@ -150,4 +151,5 @@ cython_debug/
 mdio1/*
 */mdio1/*
 pytest-of-*
-tmp/
+tmp
+debugging/*

noxfile.py

@@ -132,6 +132,7 @@ def precommit(session: Session) -> None:
         "pre-commit",
         "pre-commit-hooks",
         "pyupgrade",
+        "pytest-dependency",
     )
     session.run("pre-commit", *args)
     if args and args[0] == "install":

src/mdio/__main__.py

@@ -10,6 +10,8 @@
 import click_params
 
 import mdio
+from mdio.commands.utility import copy
+from mdio.commands.utility import info
 
 
 plugin_folder = os.path.join(os.path.dirname(__file__), "commands")
@@ -40,11 +42,18 @@ class MyCLI(click.MultiCommand):
         http://lybniz2.sourceforge.net/safeeval.html
     """
 
+    _command_mapping = {
+        "copy": copy,
+        "info": info,
+    }
+
+    protected_files = ["__init__.py", "utility.py"]
+
     def list_commands(self, ctx: click.Context) -> list[str]:
         """List commands available under `commands` module."""
-        rv = []
+        rv = list(self._command_mapping.keys())
         for filename in os.listdir(plugin_folder):
-            if filename.endswith(".py") and filename != "__init__.py":
+            if filename.endswith(".py") and filename not in self.protected_files:
                 rv.append(filename[:-3])
         rv.sort()
 
@@ -61,6 +70,8 @@ def get_command(self, ctx: click.Context, name: str) -> dict[Callable]:
         }
         local_ns = {}
 
+        if name in self._command_mapping.keys():
+            return self._command_mapping[name]
         fn = os.path.join(plugin_folder, name + ".py")
         with open(fn) as f:
             code = compile(f.read(), fn, "exec")

src/mdio/commands/utility.py

@@ -0,0 +1,177 @@
+"""Dataset CLI Plugin."""
+
+try:
+    from typing import Optional
+
+    import click
+    import click_params
+
+    import mdio
+except SystemError:
+    pass
+
+
+@click.command(name="copy")
+@click.option(
+    "-i",
+    "--input-mdio-path",
+    required=True,
+    help="Input mdio path.",
+    type=click.Path(exists=True),
+)
+@click.option(
+    "-o",
+    "--output-mdio-path",
+    required=True,
+    help="Output path or URL to write the mdio dataset.",
+    type=click.STRING,
+)
+@click.option(
+    "-access",
+    "--access-pattern",
+    required=False,
+    default="012",
+    help="Access pattern of the file",
+    type=click.STRING,
+    show_default=True,
+)
+@click.option(
+    "-exc",
+    "--excludes",
+    required=False,
+    default="",
+    help="""Data to exclude during copy. i.e. chunked_012. The raw data won’t be
+    copied, but it will create an empty array to be filled. If left blank, it will
+    copy everything.""",
+    type=click.STRING,
+)
+@click.option(
+    "-inc",
+    "--includes",
+    required=False,
+    default="",
+    help="""Data to include during copy. i.e. trace_headers. If this is not
+    specified, and certain data is excluded, it will not copy headers. To
+    preserve headers, specify trace_headers. If left blank, it will copy
+    everything except specified in excludes parameter.""",
+    type=click.STRING,
+)
+@click.option(
+    "-storage",
+    "--storage-options",
+    required=False,
+    help="Custom storage options for cloud backends",
+    type=click_params.JSON,
+)
+@click.option(
+    "-overwrite",
+    "--overwrite",
+    required=False,
+    default=False,
+    help="Flag to overwrite if mdio file if it exists",
+    type=click.BOOL,
+    show_default=True,
+)
+def copy(
+    input_mdio_path: str,
+    output_mdio_path: str,
+    access_pattern: str = "012",
+    includes: str = "",
+    excludes: str = "",
+    storage_options: Optional[dict] = None,
+    overwrite: bool = False,
+):
+    """Copy a MDIO dataset to anpther MDIO dataset.
+
+    Can also copy with empty data to be filled later. See `excludes`
+    and `includes` parameters.
+
+    More documentation about `excludes` and `includes` can be found
+    in Zarr's documentation in `zarr.convenience.copy_store`.
+    """
+    reader = mdio.MDIOReader(
+        input_mdio_path, access_pattern=access_pattern, return_metadata=True
+    )
+    mdio.copy_mdio(
+        source=reader,
+        dest_path_or_buffer=output_mdio_path,
+        excludes=excludes,
+        includes=includes,
+        storage_options=storage_options,
+        overwrite=overwrite,
+    )
+
+
+@click.command(name="info")
+@click.option(
+    "-i",
+    "--input-mdio-file",
+    required=True,
+    help="Input path of the mdio file",
+    type=click.STRING,
+)
+@click.option(
+    "-access",
+    "--access-pattern",
+    required=False,
+    default="012",
+    help="Access pattern of the file",
+    type=click.STRING,
+    show_default=True,
+)
+@click.option(
+    "-format",
+    "--output-format",
+    required=False,
+    default="plain",
+    help="""Output format, plain is human readable.  JSON will output in json
+    format for easier passing. """,
+    type=click.Choice(["plain", "json"]),
+    show_default=True,
+    show_choices=True,
+)
+def info(
+    input_mdio_file,
+    output_format,
+    access_pattern,
+):
+    """Provide information on a MDIO dataset.
+
+    By default this returns human readable information about the grid and stats for
+    the dataset. If output-format is set to json then a json is returned to
+    facilitate parsing.
+    """
+    reader = mdio.MDIOReader(
+        input_mdio_file, access_pattern=access_pattern, return_metadata=True
+    )
+    mdio_dict = {}
+    mdio_dict["grid"] = {}
+    for axis in reader.grid.dim_names:
+        dim = reader.grid.select_dim(axis)
+        min = dim.coords[0]
+        max = dim.coords[-1]
+        size = dim.coords.shape[0]
+        axis_dict = {"name": axis, "min": min, "max": max, "size": size}
+        mdio_dict["grid"][axis] = axis_dict
+
+    if output_format == "plain":
+        click.echo("{:<10} {:<10} {:<10} {:<10}".format("NAME", "MIN", "MAX", "SIZE"))
+        click.echo("=" * 40)
+
+        for _, axis_dict in mdio_dict["grid"].items():
+            click.echo(
+                "{:<10} {:<10} {:<10} {:<10}".format(
+                    axis_dict["name"],
+                    axis_dict["min"],
+                    axis_dict["max"],
+                    axis_dict["size"],
+                )
+            )
+
+        click.echo("\n\n{:<10} {:<10}".format("STAT", "VALUE"))
+        click.echo("=" * 20)
+        for name, stat in reader.stats.items():
+            click.echo(f"{name:<10} {stat:<10}")
+    if output_format == "json":
+        mdio_dict["stats"] = reader.stats
+        click.echo(mdio_dict)

tests/conftest.py

@@ -32,6 +32,13 @@ def zarr_tmp(tmp_path_factory):
     return tmp_file
 
 
+@pytest.fixture(scope="session")
+def zarr_tmp2(tmp_path_factory):
+    """Make a temp file for the output MDIO."""
+    tmp_file = tmp_path_factory.mktemp(r"mdio2")
+    return tmp_file
+
+
 @pytest.fixture(scope="session")
 def segy_export_ibm_tmp(tmp_path_factory):
     """Make a temp file for the round-trip IBM SEG-Y."""

tests/test_main.py

@@ -15,6 +15,7 @@ def runner() -> CliRunner:
     return CliRunner()
 
 
+@pytest.mark.dependency()
 def test_main_succeeds(runner: CliRunner, segy_input: str, zarr_tmp: Path) -> None:
     """It exits with a status code of zero."""
     cli_args = ["segy", "import"]
@@ -27,6 +28,27 @@ def test_main_succeeds(runner: CliRunner, segy_input: str, zarr_tmp: Path) -> No
     assert result.exit_code == 0
 
 
+@pytest.mark.dependency(depends=["test_main_succeeds"])
+def test_main_info_succeeds(runner: CliRunner, zarr_tmp: Path) -> None:
+    """It exits with a status code of zero."""
+    cli_args = ["info"]
+    cli_args.extend(["-i", str(zarr_tmp)])
+
+    result = runner.invoke(__main__.main, args=cli_args)
+    assert result.exit_code == 0
+
+
+@pytest.mark.dependency(depends=["test_main_succeeds"])
+def test_main_copy_succeeds(runner: CliRunner, zarr_tmp: Path, zarr_tmp2: Path) -> None:
+    """It exits with a status code of zero."""
+    cli_args = ["copy"]
+    cli_args.extend(["-i", str(zarr_tmp)])
+    cli_args.extend(["-o", str(zarr_tmp2)])
+
+    result = runner.invoke(__main__.main, args=cli_args)
+    assert result.exit_code == 0
+
+
 def test_cli_version(runner: CliRunner) -> None:
     """Check if version prints without error."""
     cli_args = ["--version"]