Revert non-related changes and focus on the _GMT_DATASET class

Author: seisman

pygmt/helpers/__init__.py

@@ -20,6 +20,5 @@
     is_nonstr_iter,
     launch_external_viewer,
     non_ascii_to_octal,
-    return_table,
 )
 from pygmt.helpers.validators import validate_output_table_type

pygmt/helpers/utils.py

@@ -11,7 +11,6 @@
 import time
 import webbrowser
 from collections.abc import Iterable
-from typing import Literal
 
 import xarray as xr
 from pygmt.exceptions import GMTInvalidInput
@@ -556,48 +555,3 @@ def args_in_kwargs(args, kwargs):
     return any(
         kwargs.get(arg) is not None and kwargs.get(arg) is not False for arg in args
     )
-
-
-def return_table(
-    session,
-    output_type: Literal["pandas", "numpy", "file"],
-    vfile: str,
-    column_names: list[str] | None = None,
-):
-    """
-    Return an output table from a virtual file based on the output type.
-
-    Parameters
-    ----------
-    session : :class:`pygmt.clib.Session`
-        The current session.
-    output_type
-        The output type. Valid values are ``"pandas"``, ``"numpy"``, or ``"file"``.
-    vfile
-        The virtual file name.
-    column_names
-        The column names for the :class:`pandas.DataFrame` output.
-
-    Returns
-    -------
-    :class:`pandas.DataFrame` or :class:`numpy.ndarray` or None
-        The output table. If ``output_type`` is ``"file"``, returns ``None``.
-    """
-    if output_type == "file":  # Already written to file, so return None
-        return None
-    # Read the virtual file as a GMT dataset and convert to pandas.DataFrame
-    result = session.read_virtualfile(vfile, kind="dataset").contents.to_dataframe()
-    # assign column names
-    if column_names is not None:
-        result.columns = column_names
-    # convert text data from object dtype to string dtype
-    result = result.convert_dtypes(
-        convert_string=True,
-        convert_integer=False,
-        convert_floating=False,
-        convert_boolean=False,
-    )
-    if output_type == "pandas":
-        return result
-    # NumPy.ndarray output
-    return result.to_numpy()

pygmt/src/blockm.py

@@ -5,18 +5,17 @@
 import pandas as pd
 from pygmt.clib import Session
 from pygmt.helpers import (
+    GMTTempFile,
     build_arg_string,
     fmt_docstring,
     kwargs_to_strings,
-    return_table,
     use_alias,
-    validate_output_table_type,
 )
 
 __doctest_skip__ = ["blockmean", "blockmedian", "blockmode"]
 
 
-def _blockm(block_method, data, x, y, z, output_type, outfile, **kwargs):
+def _blockm(block_method, data, x, y, z, outfile, **kwargs):
     r"""
     Block average (x, y, z) data tables by mean, median, or mode estimation.
 
@@ -42,26 +41,31 @@ def _blockm(block_method, data, x, y, z, output_type, outfile, **kwargs):
         - None if ``outfile`` is set (filtered output will be stored in file
           set by ``outfile``)
     """
-    output_type = validate_output_table_type(output_type, outfile=outfile)
-
-    with Session() as lib:
-        with lib.virtualfile_from_data(
-            check_kind="vector", data=data, x=x, y=y, z=z, required_z=True
-        ) as vintbl, lib.virtualfile_to_data(kind="dataset", fname=outfile) as vouttbl:
-            lib.call_module(
-                module=block_method,
-                args=build_arg_string(kwargs, infile=vintbl, outfile=vouttbl),
+    with GMTTempFile(suffix=".csv") as tmpfile:
+        with Session() as lib:
+            table_context = lib.virtualfile_from_data(
+                check_kind="vector", data=data, x=x, y=y, z=z, required_z=True
             )
-        column_names = None
-        if isinstance(data, pd.DataFrame):
-            column_names = data.columns.to_list()
-
-        return return_table(
-            session=lib,
-            output_type=output_type,
-            vfile=vouttbl,
-            column_names=column_names,
-        )
+            # Run blockm* on data table
+            with table_context as infile:
+                if outfile is None:
+                    outfile = tmpfile.name
+                lib.call_module(
+                    module=block_method,
+                    args=build_arg_string(kwargs, infile=infile, outfile=outfile),
+                )
+
+        # Read temporary csv output to a pandas table
+        if outfile == tmpfile.name:  # if user did not set outfile, return pd.DataFrame
+            try:
+                column_names = data.columns.to_list()
+                result = pd.read_csv(tmpfile.name, sep="\t", names=column_names)
+            except AttributeError:  # 'str' object has no attribute 'columns'
+                result = pd.read_csv(tmpfile.name, sep="\t", header=None, comment=">")
+        elif outfile != tmpfile.name:  # return None if outfile set, output in outfile
+            result = None
+
+    return result
 
 
 @fmt_docstring
@@ -82,9 +86,7 @@ def _blockm(block_method, data, x, y, z, output_type, outfile, **kwargs):
     w="wrap",
 )
 @kwargs_to_strings(I="sequence", R="sequence", i="sequence_comma", o="sequence_comma")
-def blockmean(
-    data=None, x=None, y=None, z=None, output_type="pandas", outfile=None, **kwargs
-):
+def blockmean(data=None, x=None, y=None, z=None, outfile=None, **kwargs):
     r"""
     Block average (x, y, z) data tables by mean estimation.
 
@@ -157,14 +159,7 @@ def blockmean(
     >>> data_bmean = pygmt.blockmean(data=data, region=[245, 255, 20, 30], spacing="5m")
     """
     return _blockm(
-        block_method="blockmean",
-        data=data,
-        x=x,
-        y=y,
-        z=z,
-        output_type=output_type,
-        outfile=outfile,
-        **kwargs,
+        block_method="blockmean", data=data, x=x, y=y, z=z, outfile=outfile, **kwargs
     )
 
 
@@ -185,9 +180,7 @@ def blockmean(
     w="wrap",
 )
 @kwargs_to_strings(I="sequence", R="sequence", i="sequence_comma", o="sequence_comma")
-def blockmedian(
-    data=None, x=None, y=None, z=None, output_type="pandas", outfile=None, **kwargs
-):
+def blockmedian(data=None, x=None, y=None, z=None, outfile=None, **kwargs):
     r"""
     Block average (x, y, z) data tables by median estimation.
 
@@ -253,14 +246,7 @@ def blockmedian(
     ... )
     """
     return _blockm(
-        block_method="blockmedian",
-        data=data,
-        x=x,
-        y=y,
-        z=z,
-        output_type=output_type,
-        outfile=outfile,
-        **kwargs,
+        block_method="blockmedian", data=data, x=x, y=y, z=z, outfile=outfile, **kwargs
     )
 
 
@@ -281,9 +267,7 @@ def blockmedian(
     w="wrap",
 )
 @kwargs_to_strings(I="sequence", R="sequence", i="sequence_comma", o="sequence_comma")
-def blockmode(
-    data=None, x=None, y=None, z=None, output_type="pandas", outfile=None, **kwargs
-):
+def blockmode(data=None, x=None, y=None, z=None, outfile=None, **kwargs):
     r"""
     Block average (x, y, z) data tables by mode estimation.
 
@@ -347,12 +331,5 @@ def blockmode(
     >>> data_bmode = pygmt.blockmode(data=data, region=[245, 255, 20, 30], spacing="5m")
     """
     return _blockm(
-        block_method="blockmode",
-        data=data,
-        x=x,
-        y=y,
-        z=z,
-        output_type=output_type,
-        outfile=outfile,
-        **kwargs,
+        block_method="blockmode", data=data, x=x, y=y, z=z, outfile=outfile, **kwargs
     )

pygmt/src/filter1d.py

@@ -2,12 +2,13 @@
 filter1d - Time domain filtering of 1-D data tables
 """
 
+import pandas as pd
 from pygmt.clib import Session
 from pygmt.exceptions import GMTInvalidInput
 from pygmt.helpers import (
+    GMTTempFile,
     build_arg_string,
     fmt_docstring,
-    return_table,
     use_alias,
     validate_output_table_type,
 )
@@ -116,13 +117,23 @@ def filter1d(data, output_type="pandas", outfile=None, **kwargs):
 
     output_type = validate_output_table_type(output_type, outfile=outfile)
 
-    with Session() as lib:
-        with lib.virtualfile_from_data(
-            check_kind="vector", data=data
-        ) as vintbl, lib.virtualfile_to_data(kind="dataset", fname=outfile) as vouttbl:
-            lib.call_module(
-                module="filter1d",
-                args=build_arg_string(kwargs, infile=vintbl, outfile=vouttbl),
-            )
-
-        return return_table(session=lib, output_type=output_type, vfile=vouttbl)
+    with GMTTempFile() as tmpfile:
+        with Session() as lib:
+            file_context = lib.virtualfile_from_data(check_kind="vector", data=data)
+            with file_context as infile:
+                if outfile is None:
+                    outfile = tmpfile.name
+                lib.call_module(
+                    module="filter1d",
+                    args=build_arg_string(kwargs, infile=infile, outfile=outfile),
+                )
+
+        # Read temporary csv output to a pandas table
+        if outfile == tmpfile.name:  # if user did not set outfile, return pd.DataFrame
+            result = pd.read_csv(tmpfile.name, sep="\t", header=None, comment=">")
+        elif outfile != tmpfile.name:  # return None if outfile set, output in outfile
+            result = None
+
+        if output_type == "numpy":
+            result = result.to_numpy()
+    return result

pygmt/src/grd2xyz.py

@@ -2,14 +2,15 @@
 grd2xyz - Convert grid to data table
 """
 
+import pandas as pd
 import xarray as xr
 from pygmt.clib import Session
 from pygmt.exceptions import GMTInvalidInput
 from pygmt.helpers import (
+    GMTTempFile,
     build_arg_string,
     fmt_docstring,
     kwargs_to_strings,
-    return_table,
     use_alias,
     validate_output_table_type,
 )
@@ -149,24 +150,31 @@ def grd2xyz(grid, output_type="pandas", outfile=None, **kwargs):
         )
 
     # Set the default column names for the pandas dataframe header
-    column_names = ["x", "y", "z"]
+    dataframe_header = ["x", "y", "z"]
     # Let output pandas column names match input DataArray dimension names
-    if isinstance(grid, xr.DataArray):
+    if isinstance(grid, xr.DataArray) and output_type == "pandas":
         # Reverse the dims because it is rows, columns ordered.
-        column_names = [grid.dims[1], grid.dims[0], grid.name]
-
-    with Session() as lib:
-        with lib.virtualfile_from_data(
-            check_kind="raster", data=grid
-        ) as vingrd, lib.virtualfile_to_data(kind="dataset", fname=outfile) as vouttbl:
-            lib.call_module(
-                module="grd2xyz",
-                args=build_arg_string(kwargs, infile=vingrd, outfile=vouttbl),
+        dataframe_header = [grid.dims[1], grid.dims[0], grid.name]
+
+    with GMTTempFile() as tmpfile:
+        with Session() as lib:
+            file_context = lib.virtualfile_from_data(check_kind="raster", data=grid)
+            with file_context as infile:
+                if outfile is None:
+                    outfile = tmpfile.name
+                lib.call_module(
+                    module="grd2xyz",
+                    args=build_arg_string(kwargs, infile=infile, outfile=outfile),
+                )
+
+        # Read temporary csv output to a pandas table
+        if outfile == tmpfile.name:  # if user did not set outfile, return pd.DataFrame
+            result = pd.read_csv(
+                tmpfile.name, sep="\t", names=dataframe_header, comment=">"
             )
+        elif outfile != tmpfile.name:  # return None if outfile set, output in outfile
+            result = None
 
-        return return_table(
-            session=lib,
-            output_type=output_type,
-            vfile=vouttbl,
-            column_names=column_names,
-        )
+        if output_type == "numpy":
+            result = result.to_numpy()
+    return result