FEAT-#6832: Implement read_xml_glob, to_xml_glob

docs/flow/modin/experimental/pandas.rst

@@ -15,6 +15,8 @@ Experimental API Reference
 .. autofunction:: read_pickle_distributed
 .. autofunction:: read_parquet_glob
 .. autofunction:: read_json_glob
+.. autofunction:: read_xml_glob
 .. automethod:: modin.pandas.DataFrame.modin::to_pickle_distributed
 .. automethod:: modin.pandas.DataFrame.modin::to_parquet_glob
 .. automethod:: modin.pandas.DataFrame.modin::to_json_glob
+.. automethod:: modin.pandas.DataFrame.modin::to_xml_glob

docs/supported_apis/dataframe_supported.rst

@@ -414,6 +414,10 @@ default to pandas.
 |                            |                           |                        | Experimental implementation:                       |
 |                            |                           |                        | DataFrame.modin.to_json_glob                       |
 +----------------------------+---------------------------+------------------------+----------------------------------------------------+
+| ``to_xml``                 | `to_xml`_                 | D                      |                                                    |
+|                            |                           |                        | Experimental implementation:                       |
+|                            |                           |                        | DataFrame.modin.to_xml_glob                        |
++----------------------------+---------------------------+------------------------+----------------------------------------------------+
 | ``to_latex``               | `to_latex`_               | D                      |                                                    |
 +----------------------------+---------------------------+------------------------+----------------------------------------------------+
 | ``to_orc``                 | `to_orc`_                 | D                      |                                                    |
@@ -651,6 +655,7 @@ default to pandas.
 .. _`to_hdf`: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.to_hdf.html#pandas.DataFrame.to_hdf
 .. _`to_html`: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.to_html.html#pandas.DataFrame.to_html
 .. _`to_json`: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.to_json.html#pandas.DataFrame.to_json
+.. _`to_xml`: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.to_xml.html#pandas.DataFrame.to_xml
 .. _`to_latex`: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.to_latex.html#pandas.DataFrame.to_latex
 .. _`to_orc`: https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.to_orc.html#pandas.DataFrame.to_orc
 .. _`to_parquet`: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.to_parquet.html#pandas.DataFrame.to_parquet

docs/supported_apis/io_supported.rst

@@ -51,6 +51,8 @@ default to pandas.
 | `read_json`_      | P                               | Implemented for ``lines=True``                         |
 |                   |                                 | Experimental implementation: read_json_glob            |
 +-------------------+---------------------------------+--------------------------------------------------------+
+| `read_xml`        | D                               | Experimental implementation: read_xml_glob             |
++-------------------+---------------------------------+--------------------------------------------------------+
 | `read_html`_      | D                               |                                                        |
 +-------------------+---------------------------------+--------------------------------------------------------+
 | `read_clipboard`_ | D                               |                                                        |

docs/usage_guide/advanced_usage/index.rst

@@ -44,9 +44,11 @@ Modin also supports these experimental APIs on top of pandas that are under acti
 - :py:func:`~modin.experimental.pandas.read_pickle_distributed`  -- read multiple pickle files in a directory
 - :py:func:`~modin.experimental.pandas.read_parquet_glob`  -- read multiple parquet files in a directory
 - :py:func:`~modin.experimental.pandas.read_json_glob`  -- read multiple json files in a directory
+- :py:func:`~modin.experimental.pandas.read_xml_glob`  -- read multiple xml files in a directory
 - :py:meth:`~modin.pandas.DataFrame.modin.to_pickle_distributed` -- write to multiple pickle files in a directory
 - :py:meth:`~modin.pandas.DataFrame.modin.to_parquet_glob` -- write to multiple parquet files in a directory
 - :py:meth:`~modin.pandas.DataFrame.modin.to_json_glob` -- write to multiple json files in a directory
+- :py:meth:`~modin.pandas.DataFrame.modin.to_xml_glob` -- write to multiple xml files in a directory
 
 DataFrame partitioning API
 --------------------------

modin/core/execution/dask/implementations/pandas_on_dask/io/io.py

@@ -52,6 +52,7 @@
     ExperimentalPandasJsonParser,
     ExperimentalPandasParquetParser,
     ExperimentalPandasPickleParser,
+    ExperimentalPandasXmlParser,
 )
 
 
@@ -105,6 +106,11 @@ def __make_write(*classes, build_args=build_args):
         ExperimentalGlobDispatcher,
         build_args={**build_args, "base_write": BaseIO.to_json},
     )
+    read_xml_glob = __make_read(ExperimentalPandasXmlParser, ExperimentalGlobDispatcher)
+    to_xml_glob = __make_write(
+        ExperimentalGlobDispatcher,
+        build_args={**build_args, "base_write": BaseIO.to_xml},
+    )
     read_pickle_distributed = __make_read(
         ExperimentalPandasPickleParser, ExperimentalGlobDispatcher
     )

modin/core/execution/dispatching/factories/dispatcher.py

@@ -316,6 +316,16 @@
     def to_json_glob(cls, *args, **kwargs):
         return cls.get_factory()._to_json_glob(*args, **kwargs)
 
+    @classmethod
+    @_inherit_docstrings(factories.PandasOnRayFactory._read_xml_glob)
+    def read_xml_glob(cls, *args, **kwargs):
+        return cls.get_factory()._read_xml_glob(*args, **kwargs)
+
+    @classmethod
+    @_inherit_docstrings(factories.PandasOnRayFactory._to_xml_glob)
+    def to_xml_glob(cls, *args, **kwargs):
+        return cls.get_factory()._to_xml_glob(*args, **kwargs)
+
     @classmethod
     @_inherit_docstrings(factories.PandasOnRayFactory._read_custom_text)
     def read_custom_text(cls, **kwargs):
@@ -331,6 +341,11 @@
     def to_json(cls, *args, **kwargs):
         return cls.get_factory()._to_json(*args, **kwargs)
 
+    @classmethod
+    @_inherit_docstrings(factories.BaseFactory._to_xml)
+    def to_xml(cls, *args, **kwargs):
+        return cls.get_factory()._to_xml(*args, **kwargs)
+
     @classmethod
     @_inherit_docstrings(factories.BaseFactory._to_parquet)
     def to_parquet(cls, *args, **kwargs):

modin/core/execution/dispatching/factories/factories.py

@@ -427,6 +427,20 @@
         """
         return cls.io_cls.to_json(*args, **kwargs)
 
+    @classmethod
+    def _to_xml(cls, *args, **kwargs):
+        """
+        Write query compiler content to a XML file.
+
+        Parameters
+        ----------
+        *args : args
+            Arguments to pass to the writer method.
+        **kwargs : kwargs
+            Arguments to pass to the writer method.
+        """
+        return cls.io_cls.to_xml(*args, **kwargs)
+
     @classmethod
     def _to_parquet(cls, *args, **kwargs):
         """
@@ -596,6 +610,39 @@
             )
         return cls.io_cls.to_json_glob(*args, **kwargs)
 
+    @classmethod
+    @doc(
+        _doc_io_method_raw_template,
+        source="XML files",
+        params=_doc_io_method_kwargs_params,
+    )
+    def _read_xml_glob(cls, **kwargs):
+        current_execution = get_current_execution()
+        if current_execution not in supported_executions:
+            raise NotImplementedError(
+                f"`_read_xml_glob()` is not implemented for {current_execution} execution."
+            )
+        return cls.io_cls.read_xml_glob(**kwargs)
+
+    @classmethod
+    def _to_xml_glob(cls, *args, **kwargs):
+        """
+        Write query compiler content to several XML files.
+
+        Parameters
+        ----------
+        *args : args
+            Arguments to pass to the writer method.
+        **kwargs : kwargs
+            Arguments to pass to the writer method.
+        """
+        current_execution = get_current_execution()
+        if current_execution not in supported_executions:
+            raise NotImplementedError(
+                f"`_to_xml_glob()` is not implemented for {current_execution} execution."
+            )
+        return cls.io_cls.to_xml_glob(*args, **kwargs)
+
 
 @doc(_doc_factory_class, execution_name="PandasOnRay")
 class PandasOnRayFactory(BaseFactory):

modin/core/execution/ray/implementations/pandas_on_ray/io/io.py

@@ -51,6 +51,7 @@
     ExperimentalPandasJsonParser,
     ExperimentalPandasParquetParser,
     ExperimentalPandasPickleParser,
+    ExperimentalPandasXmlParser,
 )
 
 from ..dataframe import PandasOnRayDataframe
@@ -107,6 +108,11 @@ def __make_write(*classes, build_args=build_args):
         ExperimentalGlobDispatcher,
         build_args={**build_args, "base_write": RayIO.to_json},
     )
+    read_xml_glob = __make_read(ExperimentalPandasXmlParser, ExperimentalGlobDispatcher)
+    to_xml_glob = __make_write(
+        ExperimentalGlobDispatcher,
+        build_args={**build_args, "base_write": RayIO.to_xml},
+    )
     read_pickle_distributed = __make_read(
         ExperimentalPandasPickleParser, ExperimentalGlobDispatcher
     )

modin/core/execution/unidist/implementations/pandas_on_unidist/io/io.py

@@ -51,6 +51,7 @@
     ExperimentalPandasJsonParser,
     ExperimentalPandasParquetParser,
     ExperimentalPandasPickleParser,
+    ExperimentalPandasXmlParser,
 )
 
 from ..dataframe import PandasOnUnidistDataframe
@@ -107,6 +108,11 @@ def __make_write(*classes, build_args=build_args):
         ExperimentalGlobDispatcher,
         build_args={**build_args, "base_write": UnidistIO.to_json},
     )
+    read_xml_glob = __make_read(ExperimentalPandasXmlParser, ExperimentalGlobDispatcher)
+    to_xml_glob = __make_write(
+        ExperimentalGlobDispatcher,
+        build_args={**build_args, "base_write": UnidistIO.to_xml},
+    )
     read_pickle_distributed = __make_read(
         ExperimentalPandasPickleParser, ExperimentalGlobDispatcher
     )

modin/core/io/io.py

@@ -665,6 +665,20 @@ def to_json(cls, obj, path, **kwargs):  # noqa: PR01
 
         return obj.to_json(path, **kwargs)
 
+    @classmethod
+    @_inherit_docstrings(pandas.DataFrame.to_xml, apilink="pandas.DataFrame.to_xml")
+    def to_xml(cls, obj, path, **kwargs):  # noqa: PR01
+        """
+        Convert the object to a XML string.
+
+        For parameters description please refer to pandas API.
+        """
+        ErrorMessage.default_to_pandas("`to_xml`")
+        if isinstance(obj, BaseQueryCompiler):
+            obj = obj.to_pandas()
+
+        return obj.to_xml(path, **kwargs)
+
     @classmethod
     @_inherit_docstrings(
         pandas.DataFrame.to_parquet, apilink="pandas.DataFrame.to_parquet"

modin/experimental/core/io/glob/glob_dispatcher.py

@@ -54,6 +54,8 @@ def _read(cls, **kwargs):
             path_key = "path"
         elif "path_or_buf" in kwargs:
             path_key = "path_or_buf"
+        elif "path_or_buffer" in kwargs:
+            path_key = "path_or_buffer"
         filepath_or_buffer = kwargs.pop(path_key)
         filepath_or_buffer = stringify_path(filepath_or_buffer)
         if not (isinstance(filepath_or_buffer, str) and "*" in filepath_or_buffer):
@@ -123,6 +125,8 @@ def write(cls, qc, **kwargs):
             path_key = "path"
         elif "path_or_buf" in kwargs:
             path_key = "path_or_buf"
+        elif "path_or_buffer" in kwargs:
+            path_key = "path_or_buffer"
         filepath_or_buffer = kwargs.pop(path_key)
         filepath_or_buffer = stringify_path(filepath_or_buffer)
         if not (

modin/experimental/core/storage_formats/pandas/parsers.py

@@ -150,6 +150,24 @@ def parse(fname, **kwargs):
         return _split_result_for_readers(1, num_splits, df) + [length, width]
 
 
+@doc(_doc_pandas_parser_class, data_type="XML files")
+class ExperimentalPandasXmlParser(PandasParser):
+    @staticmethod
+    @doc(_doc_parse_func, parameters=_doc_parse_parameters_common)
+    def parse(fname, **kwargs):
+        warnings.filterwarnings("ignore")
+        num_splits = 1
+        single_worker_read = kwargs.pop("single_worker_read", None)
+        df = pandas.read_xml(fname, **kwargs)
+        if single_worker_read:
+            return df
+
+        length = len(df)
+        width = len(df.columns)
+
+        return _split_result_for_readers(1, num_splits, df) + [length, width]
+
+
 @doc(_doc_pandas_parser_class, data_type="custom text")
 class ExperimentalCustomTextParser(PandasParser):
     @staticmethod

modin/experimental/pandas/__init__.py

@@ -44,6 +44,7 @@
     read_parquet_glob,
     read_pickle_distributed,
     read_sql,
+    read_xml_glob,
     to_pickle_distributed,
 )
 

modin/experimental/pandas/io.py

@@ -600,3 +600,117 @@ def to_json_glob(
         storage_options=storage_options,
         mode=mode,
     )
+
+
+@expanduser_path_arg("path_or_buffer")
+def read_xml_glob(
+    path_or_buffer,
+    *,
+    xpath="./*",
+    namespaces=None,
+    elems_only=False,
+    attrs_only=False,
+    names=None,
+    dtype=None,
+    converters=None,
+    parse_dates=None,
+    encoding="utf-8",
+    parser="lxml",
+    stylesheet=None,
+    iterparse=None,
+    compression="infer",
+    storage_options: StorageOptions = None,
+    dtype_backend=lib.no_default,
+) -> DataFrame:  # noqa: PR01
+    """
+    Read XML document into a DataFrame object.
+
+    This experimental feature provides parallel reading from multiple XML files which are
+    defined by glob pattern. The files must contain parts of one dataframe, which can be
+    obtained, for example, by `DataFrame.modin.to_xml_glob` function.
+
+    Returns
+    -------
+    DataFrame
+
+    Notes
+    -----
+    * Only string type supported for `path_or_buffer` argument.
+    * The rest of the arguments are the same as for `pandas.read_xml`.
+    """
+    from modin.core.execution.dispatching.factories.dispatcher import FactoryDispatcher
+
+    return DataFrame(
+        query_compiler=FactoryDispatcher.read_xml_glob(
+            path_or_buffer=path_or_buffer,
+            xpath=xpath,
+            namespaces=namespaces,
+            elems_only=elems_only,
+            attrs_only=attrs_only,
+            names=names,
+            dtype=dtype,
+            converters=converters,
+            parse_dates=parse_dates,
+            encoding=encoding,
+            parser=parser,
+            stylesheet=stylesheet,
+            iterparse=iterparse,
+            compression=compression,
+            storage_options=storage_options,
+            dtype_backend=dtype_backend,
+        )
+    )
+
+
+@expanduser_path_arg("path_or_buffer")
+def to_xml_glob(
+    self,
+    path_or_buffer=None,
+    index=True,
+    root_name="data",
+    row_name="row",
+    na_rep=None,
+    attr_cols=None,
+    elem_cols=None,
+    namespaces=None,
+    prefix=None,
+    encoding="utf-8",
+    xml_declaration=True,
+    pretty_print=True,
+    parser="lxml",
+    stylesheet=None,
+    compression="infer",
+    storage_options=None,
+) -> None:  # noqa: PR01
+    """
+    Render a DataFrame to an XML document.
+
+    Notes
+    -----
+    * Only string type supported for `path_or_buffer` argument.
+    * The rest of the arguments are the same as for `pandas.to_xml`.
+    """
+    obj = self
+    from modin.core.execution.dispatching.factories.dispatcher import FactoryDispatcher
+
+    if isinstance(self, DataFrame):
+        obj = self._query_compiler
+    FactoryDispatcher.to_xml_glob(
+        obj,
+        path_or_buffer=path_or_buffer,
+        index=index,
+        root_name=root_name,
+        row_name=row_name,
+        na_rep=na_rep,
+        attr_cols=attr_cols,
+        elem_cols=elem_cols,
+        namespaces=namespaces,
+        prefix=prefix,
+        encoding=encoding,
+        xml_declaration=xml_declaration,
+        pretty_print=pretty_print,
+        parser=parser,
+        stylesheet=stylesheet,
+        compression=compression,
+        storage_options=storage_options,
+    )