seabird

This page provides functions for parsing data files in Seabird Scientific format. The Seabird Scientific format is commonly used for oceanographic data collection and is supported by Seabird Scientific.

`btl(file_path, encoding='UTF-8', xml_parsing_error_level='ERROR', save_orginal_header=False)` ¶

Parse Seabird BTL format

Parameters:

Name	Type	Description	Default
`file_path`	`str`	file path	required
`encoding`	`str`	Encoding to use. Defaults to "UTF-8".	`'UTF-8'`
`xml_parsing_error_level`	`str`	Error level for XML parsing. Defaults to "ERROR".	`'ERROR'`

Returns:

Type	Description
`Dataset`	xarray.Dataset: Dataset

Source code in ocean_data_parser/parsers/seabird.py

def btl(
    file_path: str,
    encoding: str = "UTF-8",
    xml_parsing_error_level="ERROR",
    save_orginal_header: bool = False,
) -> xarray.Dataset:
    """Parse Seabird BTL format

    Args:
        file_path (str): file path
        encoding (str, optional): Encoding to use. Defaults to "UTF-8".
        xml_parsing_error_level (str, optional): Error level for XML parsing. Defaults to "ERROR".

    Returns:
        xarray.Dataset: Dataset
    """

    with open(file_path, encoding=encoding) as f:
        header = _parse_seabird_file_header(
            f, xml_parsing_error_level=xml_parsing_error_level
        )

        # Retrieve variables from bottle header and lower the first letter of each variable
        variable_list = [
            var[0].lower() + var[1:] for var in header["bottle_columns"]
        ] + ["stats"]
        df = pd.read_fwf(
            f,
            widths=[10, 12] + [11] * (len(header["bottle_columns"]) - 1),
            names=variable_list,
            dtype={var: var_dtypes.get(var, float) for var in variable_list},
        )

    # Split statistical data info separate dateframes
    df["bottle"] = df["bottle"].ffill().astype(int)
    df["stats"] = df["stats"].str.extract(r"\((.*)\)")
    df = df.set_index("bottle")
    df_grouped = df.query("stats=='avg'")
    for stats in df.query("stats!='avg'")["stats"].drop_duplicates().to_list():
        df_grouped = df_grouped.join(df.query("stats==@stats").add_suffix(f"_{stats}"))
    df = df_grouped

    # Generate time variable
    df["time"] = pd.to_datetime(df.filter(like="date").apply(" ".join, axis="columns"))

    # Ignore extra variables
    drop_columns = [col for col in df if re.search("^date|^stats|^bottle_", col)]
    df = df.drop(columns=drop_columns)

    # Improve metadata

    n_scan_per_bottle = [
        int(step["scans_per_bottle"])
        for step in header["processing"]
        if step["module"] == "datcnv"
    ]
    if not n_scan_per_bottle:
        logger.warning("Failed to retrieve the number of scans per bottle")
        n_scan_per_bottle = "unknown"
    else:
        n_scan_per_bottle = n_scan_per_bottle[0]
    header = _generate_seabird_cf_history(header)

    # Retrieve vocabulary associated with each variables
    header["variables"] = {var: header["variables"].get(var, {}) for var in df.columns}
    header["variables"] = _add_seabird_vocabulary(header["variables"])

    # Convert to xarray
    ds = _convert_sbe_dataframe_to_dataset(df, header)

    # Add cell_method attribute
    for var in ds:
        if var.endswith("_sdev") and var[:-5] in ds:
            ds[var].attrs["cell_method"] = (
                f"scan: standard_deviation (previous {n_scan_per_bottle} scans)"
            )
            # TODO confirm that seabird uses the previous records from this timestamp
        elif var not in ["time", "bottle"]:
            ds[var].attrs["cell_method"] = (
                f"scan: mean (previous {n_scan_per_bottle} scans)"
            )

    if not save_orginal_header:
        ds.attrs.pop("seabird_header")
    return standardize_dataset(ds)

`cnv(file_path, encoding='UTF-8', encoding_errors='strict', xml_parsing_error_level='ERROR', generate_instrument_variables=False, save_orginal_header=False)` ¶

Parse Seabird CNV format