Skip to content

Load mixin

:class:sciwork.fs.load.Load expose :meth:any_data_loader, a facade over the specialized loaders available in sciwork.fs.loaders_base and related parsers. It combines :class:PathOpsBase, the :class:Classify <sciwork.fs.classify.Classify> mixin and the concrete loader implementations.

any_data_loader

Open a structured data file and return either a :class:pandas.DataFrame or a Python object, depending on the detected format. The method accepts a large set of keyword arguments; loaders ignore options that do not apply to their format.

Supported types include:

  • Excel (.xlsx, .xlsm, .xls) → DataFrame via openpyxl.
  • CSV/TSV → DataFrame with optional delimiter detection.
  • Plain text → DataFrame or parsed rows.
  • JSON → dict/list.
  • XML → DataFrame build from element dictionaries.
  • SIF (Andor scientific Image Format) → DataFrame via _load_sif.
  • UV/VIS SPC → format-specific parser returning a DataFrame.

Key parameters:

  • sheet_name — An Excel sheet to load. "choice" prompts the user to pick a sheet via :class:sciwork.console.Prompter when available.
  • encoding — override encoding for text formats (auto-detected otherwise; focused primarily on english/czech common language encodings).
  • delimiter — enforce a specific delimiter for CSV/TXT.
  • header — row index of the header when building DataFrames.
  • dtype — pass custom dtypes to the pandas reader.
  • include_hidden_rows — for Excel; include rows marked as hidden.
  • force_type — bypass classification and load using a specific type label.
df = fs.any_data_loader(
    "data/measurements.xlsx",
    sheet_name="Sheet1",
    include_hidden_rows=False,
)

If force_type is omitted the helper relies on :meth:Classify.classify_path <sciwork.fs.classify.Classify.classify_path> to determine which loader to call. Unknown extensions raise :class:ValueError.

Paths are resolved absolutely or relative to base_dir and missing file triggers a :class:FileNotFoundError before any loader is attempted.