etlplus 0.10.5__py3-none-any.whl → 0.11.1__py3-none-any.whl

etlplus/cli/commands.py

@@ -36,7 +36,7 @@ from typing import cast
 import typer
 
 from .. import __version__
-from ..enums import FileFormat
+from ..file import FileFormat
 from . import handlers
 from .constants import CLI_DESCRIPTION
 from .constants import CLI_EPILOG

etlplus/cli/constants.py

@@ -9,7 +9,7 @@ from __future__ import annotations
 from typing import Final
 
 from ..enums import DataConnectorType
-from ..enums import FileFormat
+from ..file import FileFormat
 
 # SECTION: EXPORTS ========================================================== #
 

etlplus/cli/io.py

@@ -15,8 +15,8 @@ from pathlib import Path
 from typing import Any
 from typing import cast
 
-from ..enums import FileFormat
 from ..file import File
+from ..file import FileFormat
 from ..types import JSONData
 from ..utils import print_json
 
@@ -331,6 +331,6 @@ def write_json_output(
     """
     if not output_path or output_path == '-':
         return False
-    File(Path(output_path), FileFormat.JSON).write_json(data)
+    File(Path(output_path), FileFormat.JSON).write(data)
     print(f'{success_message} {output_path}')
     return True

etlplus/config/pipeline.py

@@ -24,8 +24,8 @@ from typing import Any
 from typing import Self
 
 from ..api import ApiConfig
-from ..enums import FileFormat
 from ..file import File
+from ..file import FileFormat
 from ..types import StrAnyMap
 from ..utils import coerce_dict
 from ..utils import maybe_mapping
@@ -246,7 +246,7 @@ class PipelineConfig:
         TypeError
             If the YAML root is not a mapping/object.
         """
-        raw = File(Path(path), FileFormat.YAML).read_yaml()
+        raw = File(Path(path), FileFormat.YAML).read()
         if not isinstance(raw, dict):
             raise TypeError('Pipeline YAML must have a mapping/object root')
 

etlplus/enums.py

@@ -8,7 +8,6 @@ from __future__ import annotations
 
 import enum
 import operator as _op
-from pathlib import PurePath
 from statistics import fmean
 from typing import Self
 
@@ -23,18 +22,13 @@ __all__ = [
     # Enums
     'AggregateName',
     'CoercibleStrEnum',
-    'CompressionFormat',
     'DataConnectorType',
-    'FileFormat',
     'HttpMethod',
     'OperatorName',
     'PipelineStep',
     # Functions
-    'coerce_compression_format',
     'coerce_data_connector_type',
-    'coerce_file_format',
     'coerce_http_method',
-    'infer_file_format_and_compression',
 ]
 
 
@@ -178,39 +172,6 @@ class AggregateName(CoercibleStrEnum):
         return lambda xs, n: (fmean(xs) if xs else 0.0)
 
 
-class CompressionFormat(CoercibleStrEnum):
-    """Supported compression formats for data files."""
-
-    # -- Constants -- #
-
-    GZ = 'gz'
-    ZIP = 'zip'
-
-    # -- Class Methods -- #
-
-    @classmethod
-    def aliases(cls) -> StrStrMap:
-        """
-        Return a mapping of common aliases for each enum member.
-
-        Returns
-        -------
-        StrStrMap
-            A mapping of alias names to their corresponding enum member names.
-        """
-        return {
-            # File extensions
-            '.gz': 'gz',
-            '.gzip': 'gz',
-            '.zip': 'zip',
-            # MIME types
-            'application/gzip': 'gz',
-            'application/x-gzip': 'gz',
-            'application/zip': 'zip',
-            'application/x-zip-compressed': 'zip',
-        }
-
-
 class DataConnectorType(CoercibleStrEnum):
     """Supported data connector types."""
 
@@ -242,99 +203,6 @@ class DataConnectorType(CoercibleStrEnum):
         }
 
 
-class FileFormat(CoercibleStrEnum):
-    """Supported file formats for extraction."""
-
-    # -- Constants -- #
-
-    AVRO = 'avro'
-    CSV = 'csv'
-    FEATHER = 'feather'
-    GZ = 'gz'
-    JSON = 'json'
-    NDJSON = 'ndjson'
-    ORC = 'orc'
-    PARQUET = 'parquet'
-    TSV = 'tsv'
-    TXT = 'txt'
-    XLS = 'xls'
-    XLSX = 'xlsx'
-    ZIP = 'zip'
-    XML = 'xml'
-    YAML = 'yaml'
-
-    # -- Class Methods -- #
-
-    @classmethod
-    def aliases(cls) -> StrStrMap:
-        """
-        Return a mapping of common aliases for each enum member.
-
-        Returns
-        -------
-        StrStrMap
-            A mapping of alias names to their corresponding enum member names.
-        """
-        return {
-            # Common shorthand
-            'parq': 'parquet',
-            'yml': 'yaml',
-            # File extensions
-            '.avro': 'avro',
-            '.csv': 'csv',
-            '.feather': 'feather',
-            '.gz': 'gz',
-            '.json': 'json',
-            '.jsonl': 'ndjson',
-            '.ndjson': 'ndjson',
-            '.orc': 'orc',
-            '.parquet': 'parquet',
-            '.pq': 'parquet',
-            '.tsv': 'tsv',
-            '.txt': 'txt',
-            '.xls': 'xls',
-            '.xlsx': 'xlsx',
-            '.zip': 'zip',
-            '.xml': 'xml',
-            '.yaml': 'yaml',
-            '.yml': 'yaml',
-            # MIME types
-            'application/avro': 'avro',
-            'application/csv': 'csv',
-            'application/feather': 'feather',
-            'application/gzip': 'gz',
-            'application/json': 'json',
-            'application/jsonlines': 'ndjson',
-            'application/ndjson': 'ndjson',
-            'application/orc': 'orc',
-            'application/parquet': 'parquet',
-            'application/vnd.apache.avro': 'avro',
-            'application/vnd.apache.parquet': 'parquet',
-            'application/vnd.apache.arrow.file': 'feather',
-            'application/vnd.apache.orc': 'orc',
-            'application/vnd.ms-excel': 'xls',
-            (
-                'application/vnd.openxmlformats-'
-                'officedocument.spreadsheetml.sheet'
-            ): 'xlsx',
-            'application/x-avro': 'avro',
-            'application/x-csv': 'csv',
-            'application/x-feather': 'feather',
-            'application/x-orc': 'orc',
-            'application/x-ndjson': 'ndjson',
-            'application/x-parquet': 'parquet',
-            'application/x-yaml': 'yaml',
-            'application/xml': 'xml',
-            'application/zip': 'zip',
-            'text/csv': 'csv',
-            'text/plain': 'txt',
-            'text/tab-separated-values': 'tsv',
-            'text/tsv': 'tsv',
-            'text/xml': 'xml',
-            'text/yaml': 'yaml',
-        }
-
-
 class HttpMethod(CoercibleStrEnum):
     """Supported HTTP verbs that accept JSON payloads."""
 
@@ -360,8 +228,8 @@ class HttpMethod(CoercibleStrEnum):
         Notes
         -----
         - RFCs do not strictly forbid bodies on some other methods (e.g.,
-          ``DELETE``), but many servers/clients do not expect them. We mark
-          ``POST``, ``PUT``, and ``PATCH`` as True.
+            ``DELETE``), but many servers/clients do not expect them. We mark
+            ``POST``, ``PUT``, and ``PATCH`` as True.
         """
         return self in {HttpMethod.POST, HttpMethod.PUT, HttpMethod.PATCH}
 
@@ -465,13 +333,6 @@ class PipelineStep(CoercibleStrEnum):
 # SECTION: INTERNAL CONSTANTS ============================================== #
 
 
-# Compression formats that are also file formats.
-_COMPRESSION_FILE_FORMATS: set[FileFormat] = {
-    FileFormat.GZ,
-    FileFormat.ZIP,
-}
-
-
 # Precomputed order index for PipelineStep; avoids recomputing on each access.
 _PIPELINE_ORDER_INDEX: dict[PipelineStep, int] = {
     PipelineStep.FILTER: 0,
@@ -497,30 +358,6 @@ def coerce_data_connector_type(
     return DataConnectorType.coerce(connector)
 
 
-def coerce_file_format(
-    file_format: FileFormat | str,
-) -> FileFormat:
-    """
-    Normalize textual file format values to :class:`FileFormat`.
-
-    This thin wrapper is kept for backward compatibility; prefer
-    :meth:`FileFormat.coerce` going forward.
-    """
-    return FileFormat.coerce(file_format)
-
-
-def coerce_compression_format(
-    compression_format: CompressionFormat | str,
-) -> CompressionFormat:
-    """
-    Normalize textual compression format values to :class:`CompressionFormat`.
-
-    This thin wrapper is kept for backward compatibility; prefer
-    :meth:`CompressionFormat.coerce` going forward.
-    """
-    return CompressionFormat.coerce(compression_format)
-
-
 def coerce_http_method(
     http_method: HttpMethod | str,
 ) -> HttpMethod:
@@ -531,78 +368,3 @@ def coerce_http_method(
     :meth:`HttpMethod.coerce` going forward.
     """
     return HttpMethod.coerce(http_method)
-
-
-def infer_file_format_and_compression(
-    value: object,
-    filename: object | None = None,
-) -> tuple[FileFormat | None, CompressionFormat | None]:
-    """
-    Infer data format and compression from a filename, extension, or MIME type.
-
-    Parameters
-    ----------
-    value : object
-        A filename, extension, MIME type, or existing enum member.
-    filename : object | None, optional
-        A filename to consult for extension-based inference (e.g. when
-        ``value`` is ``application/octet-stream``).
-
-    Returns
-    -------
-    tuple[FileFormat | None, CompressionFormat | None]
-        The inferred data format and compression, if any.
-    """
-    if isinstance(value, FileFormat):
-        if value in _COMPRESSION_FILE_FORMATS:
-            return None, CompressionFormat.coerce(value.value)
-        return value, None
-    if isinstance(value, CompressionFormat):
-        return None, value
-
-    text = str(value).strip()
-    if not text:
-        return None, None
-
-    normalized = text.casefold()
-    mime = normalized.split(';', 1)[0].strip()
-
-    is_octet_stream = mime == 'application/octet-stream'
-    compression = CompressionFormat.try_coerce(mime)
-    fmt = None if is_octet_stream else FileFormat.try_coerce(mime)
-
-    is_mime = mime.startswith(
-        (
-            'application/',
-            'text/',
-            'audio/',
-            'image/',
-            'video/',
-            'multipart/',
-        ),
-    )
-    suffix_source: object | None = filename if filename is not None else text
-    if is_mime and filename is None:
-        suffix_source = None
-
-    suffixes = (
-        PurePath(str(suffix_source)).suffixes
-        if suffix_source is not None
-        else []
-    )
-    if suffixes:
-        normalized_suffixes = [suffix.casefold() for suffix in suffixes]
-        compression = (
-            CompressionFormat.try_coerce(normalized_suffixes[-1])
-            or compression
-        )
-        if compression is not None:
-            normalized_suffixes = normalized_suffixes[:-1]
-        if normalized_suffixes:
-            fmt = FileFormat.try_coerce(normalized_suffixes[-1]) or fmt
-
-    if fmt in _COMPRESSION_FILE_FORMATS:
-        compression = compression or CompressionFormat.coerce(fmt.value)
-        fmt = None
-
-    return fmt, compression

etlplus/extract.py

@@ -13,11 +13,11 @@ from typing import cast
 import requests  # type: ignore[import]
 
 from .enums import DataConnectorType
-from .enums import FileFormat
 from .enums import HttpMethod
 from .enums import coerce_data_connector_type
-from .enums import coerce_file_format
 from .file import File
+from .file import FileFormat
+from .file import coerce_file_format
 from .types import JSONData
 from .types import JSONDict
 from .types import JSONList

etlplus/file/__init__.py

@@ -0,0 +1,27 @@
+"""
+:mod:`etlplus.file` package.
+
+Public file IO helpers.
+"""
+
+from __future__ import annotations
+
+from .core import File
+from .enums import CompressionFormat
+from .enums import FileFormat
+from .enums import coerce_file_format
+from .enums import infer_file_format_and_compression
+
+# SECTION: EXPORTS ========================================================== #
+
+
+__all__ = [
+    # Class
+    'File',
+    # Enums
+    'CompressionFormat',
+    'FileFormat',
+    # Functions
+    'coerce_file_format',
+    'infer_file_format_and_compression',
+]

etlplus/file/core.py

@@ -0,0 +1,287 @@
+"""
+:mod:`etlplus.file.core` module.
+
+Shared helpers for reading and writing structured and semi-structured data
+files.
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from pathlib import Path
+
+from ..types import JSONData
+from ..types import StrPath
+from . import csv
+from . import json
+from . import xml
+from . import yaml
+from .enums import FileFormat
+from .enums import infer_file_format_and_compression
+
+# SECTION: EXPORTS ========================================================== #
+
+
+__all__ = ['File']
+
+
+# SECTION: CLASSES ========================================================== #
+
+
+@dataclass(slots=True)
+class File:
+    """
+    Convenience wrapper around structured file IO.
+
+    This class encapsulates the one-off helpers in this module as convenient
+    instance methods while retaining the original function API for
+    backward compatibility (those functions delegate to this class).
+
+    Attributes
+    ----------
+    path : Path
+        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``).
+    """
+
+    # -- Attributes -- #
+
+    path: Path
+    file_format: FileFormat | None = None
+
+    # -- Magic Methods (Object Lifecycle) -- #
+
+    def __post_init__(self) -> None:
+        """
+        Auto-detect and set the file format on initialization.
+
+        If no explicit ``file_format`` is provided, attempt to infer it from
+        the file path's extension and update :attr:`file_format`. If the
+        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)
+
+        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
+
+    # -- Internal Instance Methods -- #
+
+    def _assert_exists(self) -> None:
+        """
+        Raise FileNotFoundError if :attr:`path` does not exist.
+
+        This centralizes existence checks across multiple read methods.
+        """
+        if not self.path.exists():
+            raise FileNotFoundError(f'File not found: {self.path}')
+
+    def _ensure_format(self) -> FileFormat:
+        """
+        Resolve the active format, guessing from extension if needed.
+
+        Returns
+        -------
+        FileFormat
+            The resolved file format.
+        """
+        return (
+            self.file_format
+            if self.file_format is not None
+            else self._guess_format()
+        )
+
+    def _guess_format(self) -> FileFormat:
+        """
+        Infer the file format from the filename extension.
+
+        Returns
+        -------
+        FileFormat
+            The inferred file format based on the file extension.
+
+        Raises
+        ------
+        ValueError
+            If the extension is unknown or unsupported.
+        """
+        fmt, compression = infer_file_format_and_compression(self.path)
+        if fmt is not None:
+            return fmt
+        if compression is not None:
+            raise ValueError(
+                'Cannot infer file format from compressed file '
+                f'{self.path!r} with compression {compression.value!r}',
+            )
+        raise ValueError(
+            f'Cannot infer file format from extension {self.path.suffix!r}',
+        )
+
+    # -- Instance Methods (Generic API) -- #
+
+    def read(self) -> JSONData:
+        """
+        Read structured data from :attr:`path` using :attr:`file_format`.
+
+        Returns
+        -------
+        JSONData
+            The structured data read from the file.
+
+        Raises
+        ------
+        ValueError
+            If the resolved file format is unsupported.
+        """
+        self._assert_exists()
+        fmt = self._ensure_format()
+        match fmt:
+            case FileFormat.CSV:
+                return csv.read(self.path)
+            case FileFormat.JSON:
+                return json.read(self.path)
+            case FileFormat.XML:
+                return xml.read(self.path)
+            case FileFormat.YAML:
+                return yaml.read(self.path)
+        raise ValueError(f'Unsupported format: {fmt}')
+
+    def write(
+        self,
+        data: JSONData,
+        *,
+        root_tag: str = xml.DEFAULT_XML_ROOT,
+    ) -> int:
+        """
+        Write ``data`` to :attr:`path` using :attr:`file_format`.
+
+        Parameters
+        ----------
+        data : JSONData
+            Data to write to the file.
+        root_tag : str, optional
+            Root tag name to use when writing XML files. Defaults to
+            ``'root'``.
+
+        Returns
+        -------
+        int
+            The number of records written.
+
+        Raises
+        ------
+        ValueError
+            If the resolved file format is unsupported.
+        """
+        fmt = self._ensure_format()
+        match fmt:
+            case FileFormat.CSV:
+                return csv.write(self.path, data)
+            case FileFormat.JSON:
+                return json.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)
+        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,
+        )