etlplus 0.11.5__py3-none-any.whl → 0.12.1__py3-none-any.whl

etlplus/file/core.py

@@ -11,11 +11,21 @@ from dataclasses import dataclass
 from pathlib import Path
 
 from ..types import JSONData
-from ..types import StrPath
+from . import avro
 from . import csv
+from . import feather
+from . import gz
 from . import json
+from . import ndjson
+from . import orc
+from . import parquet
+from . import tsv
+from . import txt
+from . import xls
+from . import xlsx
 from . import xml
 from . import yaml
+from . import zip
 from .enums import FileFormat
 from .enums import infer_file_format_and_compression
 
@@ -43,7 +53,15 @@ class File:
         Path to the file on disk.
     file_format : FileFormat | None, optional
         Explicit format. If omitted, the format is inferred from the file
-        extension (``.csv``, ``.json``, or ``.xml``).
+        extension (``.csv``, ``.json``, etc.).
+
+    Parameters
+    ----------
+    path : StrPath
+        Path to the file on disk.
+    file_format : FileFormat | str | None, optional
+        Explicit format. If omitted, the format is inferred from the file
+        extension (``.csv``, ``.json``, etc.).
     """
 
     # -- Attributes -- #
@@ -62,16 +80,10 @@ class File:
         extension is unknown, the attribute is left as ``None`` and will be
         validated later by :meth:`_ensure_format`.
         """
-        # Normalize incoming path (allow str in constructor) to Path.
-        if isinstance(self.path, str):
-            self.path = Path(self.path)
-
+        self.path = Path(self.path)
+        self.file_format = self._coerce_format(self.file_format)
         if self.file_format is None:
-            try:
-                self.file_format = self._guess_format()
-            except ValueError:
-                # Leave as None; _ensure_format() will raise on use if needed.
-                pass
+            self.file_format = self._maybe_guess_format()
 
     # -- Internal Instance Methods -- #
 
@@ -84,6 +96,28 @@ class File:
         if not self.path.exists():
             raise FileNotFoundError(f'File not found: {self.path}')
 
+    def _coerce_format(
+        self,
+        file_format: FileFormat | str | None,
+    ) -> FileFormat | None:
+        """
+        Normalize the file format input.
+
+        Parameters
+        ----------
+        file_format : FileFormat | str | None
+            File format specifier. Strings are coerced into
+            :class:`FileFormat`.
+
+        Returns
+        -------
+        FileFormat | None
+            A normalized file format, or ``None`` when unspecified.
+        """
+        if file_format is None or isinstance(file_format, FileFormat):
+            return file_format
+        return FileFormat.coerce(file_format)
+
     def _ensure_format(self) -> FileFormat:
         """
         Resolve the active format, guessing from extension if needed.
@@ -125,7 +159,22 @@ class File:
             f'Cannot infer file format from extension {self.path.suffix!r}',
         )
 
-    # -- Instance Methods (Generic API) -- #
+    def _maybe_guess_format(self) -> FileFormat | None:
+        """
+        Try to infer the format, returning ``None`` if it cannot be inferred.
+
+        Returns
+        -------
+        FileFormat | None
+            The inferred format, or ``None`` if inference fails.
+        """
+        try:
+            return self._guess_format()
+        except ValueError:
+            # Leave as None; _ensure_format() will raise on use if needed.
+            return None
+
+    # -- Instance Methods -- #
 
     def read(self) -> JSONData:
         """
@@ -144,14 +193,36 @@ class File:
         self._assert_exists()
         fmt = self._ensure_format()
         match fmt:
+            case FileFormat.AVRO:
+                return avro.read(self.path)
             case FileFormat.CSV:
                 return csv.read(self.path)
+            case FileFormat.FEATHER:
+                return feather.read(self.path)
+            case FileFormat.GZ:
+                return gz.read(self.path)
             case FileFormat.JSON:
                 return json.read(self.path)
+            case FileFormat.NDJSON:
+                return ndjson.read(self.path)
+            case FileFormat.ORC:
+                return orc.read(self.path)
+            case FileFormat.PARQUET:
+                return parquet.read(self.path)
+            case FileFormat.TSV:
+                return tsv.read(self.path)
+            case FileFormat.TXT:
+                return txt.read(self.path)
+            case FileFormat.XLS:
+                return xls.read(self.path)
+            case FileFormat.XLSX:
+                return xlsx.read(self.path)
             case FileFormat.XML:
                 return xml.read(self.path)
             case FileFormat.YAML:
                 return yaml.read(self.path)
+            case FileFormat.ZIP:
+                return zip.read(self.path)
         raise ValueError(f'Unsupported format: {fmt}')
 
     def write(
@@ -183,105 +254,34 @@ class File:
         """
         fmt = self._ensure_format()
         match fmt:
+            case FileFormat.AVRO:
+                return avro.write(self.path, data)
             case FileFormat.CSV:
                 return csv.write(self.path, data)
+            case FileFormat.FEATHER:
+                return feather.write(self.path, data)
+            case FileFormat.GZ:
+                return gz.write(self.path, data)
             case FileFormat.JSON:
                 return json.write(self.path, data)
+            case FileFormat.NDJSON:
+                return ndjson.write(self.path, data)
+            case FileFormat.ORC:
+                return orc.write(self.path, data)
+            case FileFormat.PARQUET:
+                return parquet.write(self.path, data)
+            case FileFormat.TSV:
+                return tsv.write(self.path, data)
+            case FileFormat.TXT:
+                return txt.write(self.path, data)
+            case FileFormat.XLS:
+                return xls.write(self.path, data)
+            case FileFormat.XLSX:
+                return xlsx.write(self.path, data)
             case FileFormat.XML:
                 return xml.write(self.path, data, root_tag=root_tag)
             case FileFormat.YAML:
                 return yaml.write(self.path, data)
+            case FileFormat.ZIP:
+                return zip.write(self.path, data)
         raise ValueError(f'Unsupported format: {fmt}')
-
-    # -- Class Methods -- #
-
-    @classmethod
-    def from_path(
-        cls,
-        path: StrPath,
-        *,
-        file_format: FileFormat | str | None = None,
-    ) -> File:
-        """
-        Create a :class:`File` from any path-like and optional format.
-
-        Parameters
-        ----------
-        path : StrPath
-            Path to the file on disk.
-        file_format : FileFormat | str | None, optional
-            Explicit format. If omitted, the format is inferred from the file
-            extension (``.csv``, ``.json``, or ``.xml``).
-
-        Returns
-        -------
-        File
-            The constructed :class:`File` instance.
-        """
-        resolved = Path(path)
-        ff: FileFormat | None
-        if isinstance(file_format, str):
-            ff = FileFormat.coerce(file_format)
-        else:
-            ff = file_format
-
-        return cls(resolved, ff)
-
-    @classmethod
-    def read_file(
-        cls,
-        path: StrPath,
-        file_format: FileFormat | str | None = None,
-    ) -> JSONData:
-        """
-        Read structured data.
-
-        Parameters
-        ----------
-        path : StrPath
-            Path to the file on disk.
-        file_format : FileFormat | str | None, optional
-            Explicit format. If omitted, the format is inferred from the file
-            extension (``.csv``, ``.json``, or ``.xml``).
-
-        Returns
-        -------
-        JSONData
-            The structured data read from the file.
-        """
-        return cls.from_path(path, file_format=file_format).read()
-
-    @classmethod
-    def write_file(
-        cls,
-        path: StrPath,
-        data: JSONData,
-        file_format: FileFormat | str | None = None,
-        *,
-        root_tag: str = xml.DEFAULT_XML_ROOT,
-    ) -> int:
-        """
-        Write structured data and count written records.
-
-        Parameters
-        ----------
-        path : StrPath
-            Path to the file on disk.
-        data : JSONData
-            Data to write to the file.
-        file_format : FileFormat | str | None, optional
-            Explicit format. If omitted, the format is inferred from the file
-            extension (``.csv``, ``.json``, or ``.xml``).
-        root_tag : str, optional
-            Root tag name to use when writing XML files. Defaults to
-            ``'root'``.
-
-        Returns
-        -------
-        int
-            The number of records written to the file.
-        """
-        return cls.from_path(path, file_format=file_format).write(
-            data,
-            root_tag=root_tag,
-        )

etlplus/file/csv.py

@@ -1,7 +1,7 @@
 """
 :mod:`etlplus.file.csv` module.
 
-CSV read/write helpers.
+Helpers for reading/writing CSV files.
 """
 
 from __future__ import annotations
@@ -14,6 +14,15 @@ from ..types import JSONData
 from ..types import JSONDict
 from ..types import JSONList
 
+# SECTION: EXPORTS ========================================================== #
+
+
+__all__ = [
+    'read',
+    'write',
+]
+
+
 # SECTION: FUNCTIONS ======================================================== #
 
 
@@ -21,7 +30,7 @@ def read(
     path: Path,
 ) -> JSONList:
     """
-    Load CSV content as a list of dictionaries.
+    Read CSV content from ``path``.
 
     Parameters
     ----------
@@ -48,7 +57,7 @@ def write(
     data: JSONData,
 ) -> int:
     """
-    Write CSV rows to ``path`` and return the number of rows.
+    Write ``data`` to CSV at ``path`` and return record count.
 
     Parameters
     ----------

etlplus/file/feather.py

@@ -0,0 +1,144 @@
+"""
+:mod:`etlplus.file.feather` module.
+
+Helpers for reading/writing Feather files.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Any
+from typing import cast
+
+from ..types import JSONData
+from ..types import JSONDict
+from ..types import JSONList
+
+# SECTION: EXPORTS ========================================================== #
+
+
+__all__ = [
+    'read',
+    'write',
+]
+
+
+# SECTION: INTERNAL CONSTANTS =============================================== #
+
+
+_PANDAS_CACHE: dict[str, Any] = {}
+
+
+# SECTION: INTERNAL FUNCTIONS =============================================== #
+
+
+def _get_pandas() -> Any:
+    """
+    Return the pandas module, importing it on first use.
+
+    Raises an informative ImportError if the optional dependency is missing.
+    """
+    mod = _PANDAS_CACHE.get('mod')
+    if mod is not None:  # pragma: no cover - tiny branch
+        return mod
+    try:
+        _pd = __import__('pandas')  # type: ignore[assignment]
+    except ImportError as e:  # pragma: no cover
+        raise ImportError(
+            'Feather support requires optional dependency "pandas".\n'
+            'Install with: pip install pandas',
+        ) from e
+    _PANDAS_CACHE['mod'] = _pd
+
+    return _pd
+
+
+def _normalize_records(data: JSONData) -> JSONList:
+    """
+    Normalize JSON payloads into a list of dictionaries.
+
+    Raises TypeError when payloads contain non-dict items.
+    """
+    if isinstance(data, list):
+        if not all(isinstance(item, dict) for item in data):
+            raise TypeError(
+                'Feather payloads must contain only objects (dicts)',
+            )
+        return cast(JSONList, data)
+    return [cast(JSONDict, data)]
+
+
+# SECTION: FUNCTIONS ======================================================== #
+
+
+def read(
+    path: Path,
+) -> JSONList:
+    """
+    Read Feather content from ``path``.
+
+    Parameters
+    ----------
+    path : Path
+        Path to the Feather file on disk.
+
+    Returns
+    -------
+    JSONList
+        The list of dictionaries read from the Feather file.
+
+    Raises
+    ------
+    ImportError
+        When optional dependency "pyarrow" is missing.
+    """
+    pandas = _get_pandas()
+    try:
+        frame = pandas.read_feather(path)
+    except ImportError as e:  # pragma: no cover
+        raise ImportError(
+            'Feather support requires optional dependency "pyarrow".\n'
+            'Install with: pip install pyarrow',
+        ) from e
+    return cast(JSONList, frame.to_dict(orient='records'))
+
+
+def write(
+    path: Path,
+    data: JSONData,
+) -> int:
+    """
+    Write ``data`` to Feather at ``path`` and return record count.
+
+    Parameters
+    ----------
+    path : Path
+        Path to the Feather file on disk.
+    data : JSONData
+        Data to write.
+
+    Returns
+    -------
+    int
+        Number of records written.
+
+    Raises
+    ------
+    ImportError
+        When optional dependency "pyarrow" is missing.
+    """
+    records = _normalize_records(data)
+    if not records:
+        return 0
+
+    pandas = _get_pandas()
+    path.parent.mkdir(parents=True, exist_ok=True)
+    frame = pandas.DataFrame.from_records(records)
+    try:
+        frame.to_feather(path)
+    except ImportError as e:  # pragma: no cover
+        raise ImportError(
+            'Feather support requires optional dependency "pyarrow".\n'
+            'Install with: pip install pyarrow',
+        ) from e
+    return len(records)

etlplus/file/gz.py

@@ -0,0 +1,123 @@
+"""
+:mod:`etlplus.file.gz` module.
+
+Helpers for reading/writing GZ files.
+"""
+
+from __future__ import annotations
+
+import gzip
+import tempfile
+from pathlib import Path
+
+from ..types import JSONData
+from .enums import CompressionFormat
+from .enums import FileFormat
+from .enums import infer_file_format_and_compression
+
+# SECTION: EXPORTS ========================================================== #
+
+
+__all__ = [
+    'read',
+    'write',
+]
+
+
+# SECTION: INTERNAL FUNCTIONS =============================================== #
+
+
+def _resolve_format(
+    path: Path,
+) -> FileFormat:
+    """
+    Resolve the inner file format from a .gz filename.
+
+    Parameters
+    ----------
+    path : Path
+        Path to the GZ file on disk.
+
+    Returns
+    -------
+    FileFormat
+        The inferred inner file format.
+
+    Raises
+    ------
+    ValueError
+        If the file format cannot be inferred from the filename.
+    """
+    fmt, compression = infer_file_format_and_compression(path)
+    if compression is not CompressionFormat.GZ:
+        raise ValueError(f'Not a gzip file: {path}')
+    if fmt is None:
+        raise ValueError(
+            f'Cannot infer file format from compressed file {path!r}',
+        )
+    return fmt
+
+
+# SECTION: FUNCTIONS ======================================================== #
+
+
+def read(
+    path: Path,
+) -> JSONData:
+    """
+    Read GZ content from ``path`` and parse the inner payload.
+
+    Parameters
+    ----------
+    path : Path
+        Path to the GZ file on disk.
+
+    Returns
+    -------
+    JSONData
+        Parsed payload.
+    """
+    fmt = _resolve_format(path)
+    with gzip.open(path, 'rb') as handle:
+        payload = handle.read()
+
+    with tempfile.TemporaryDirectory() as tmpdir:
+        tmp_path = Path(tmpdir) / f'payload.{fmt.value}'
+        tmp_path.write_bytes(payload)
+        from .core import File
+
+        return File(tmp_path, fmt).read()
+
+
+def write(
+    path: Path,
+    data: JSONData,
+) -> int:
+    """
+    Write ``data`` to GZ at ``path`` and return record count.
+
+    Parameters
+    ----------
+    path : Path
+        Path to the GZ file on disk.
+    data : JSONData
+        Data to write.
+
+    Returns
+    -------
+    int
+        Number of records written.
+    """
+    fmt = _resolve_format(path)
+    with tempfile.TemporaryDirectory() as tmpdir:
+        tmp_path = Path(tmpdir) / f'payload.{fmt.value}'
+        from .core import File
+
+        count = File(tmp_path, fmt).write(data)
+        payload = tmp_path.read_bytes()
+
+    path.parent.mkdir(parents=True, exist_ok=True)
+    with gzip.open(path, 'wb') as handle:
+        handle.write(payload)
+
+    return count

etlplus/file/json.py

@@ -1,7 +1,7 @@
 """
 :mod:`etlplus.file.json` module.
 
-JSON read/write helpers.
+Helpers for reading/writing JSON files.
 """
 
 from __future__ import annotations
@@ -15,6 +15,15 @@ from ..types import JSONDict
 from ..types import JSONList
 from ..utils import count_records
 
+# SECTION: EXPORTS ========================================================== #
+
+
+__all__ = [
+    'read',
+    'write',
+]
+
+
 # SECTION: FUNCTIONS ======================================================== #
 
 
@@ -22,7 +31,9 @@ def read(
     path: Path,
 ) -> JSONData:
     """
-    Load and validate JSON payloads from ``path``.
+    Read JSON content from ``path``.
+
+    Validates that the JSON root is a dict or a list of dicts.
 
     Parameters
     ----------