Skip to content

climate_ref.datasets.netcdf_utils #

Shared utilities for reading NetCDF metadata using netCDF4 directly.

These functions avoid the overhead of xarray for metadata-only reads, providing significant speedup per file by skipping Store construction, dask array creation, and full CF time decoding.

read_global_attrs(ds, keys) #

Read global attributes from a netCDF4 Dataset

Parameters:

Name Type Description Default
ds Dataset

Open netCDF4 Dataset

 required
keys list[str] | tuple[str, ...]

Attribute names to read

 required

Returns:

Type Description
dict[str, Any]

Dictionary mapping attribute names to values (None if missing)
Source code in packages/climate-ref/src/climate_ref/datasets/netcdf_utils.py 
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
def read_global_attrs(ds: netCDF4.Dataset, keys: list[str] | tuple[str, ...]) -> dict[str, Any]:
    """
    Read global attributes from a netCDF4 Dataset

    Parameters
    ----------
    ds
        Open netCDF4 Dataset
    keys
        Attribute names to read

    Returns
    -------
    :
        Dictionary mapping attribute names to values (None if missing)
    """
    return {key: getattr(ds, key, None) for key in keys}

read_mandatory_attr(ds, key) #

Read a mandatory global attribute from a netCDF4 Dataset, ensuring it is present and non-empty.

Parameters:

Name Type Description Default
ds Dataset

Open netCDF4 Dataset

 required
key str

Attribute name to read

 required

Returns:

Type Description
Any

Value of the attribute

Raises:

Type Description
ValueError

If the attribute is missing or empty
Source code in packages/climate-ref/src/climate_ref/datasets/netcdf_utils.py 
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
def read_mandatory_attr(ds: netCDF4.Dataset, key: str) -> Any:
    """
    Read a mandatory global attribute from a netCDF4 Dataset, ensuring it is present and non-empty.

    Parameters
    ----------
    ds
        Open netCDF4 Dataset
    key
        Attribute name to read

    Returns
    -------
    :
        Value of the attribute

    Raises
    ------
    ValueError
        If the attribute is missing or empty
    """
    if not hasattr(ds, key) or getattr(ds, key) in (None, ""):
        raise ValueError(f"Missing mandatory attribute: {key}")
    return getattr(ds, key)

read_time_bounds(ds) #

Read the first and last time values from a netCDF4 Dataset

Reads only two raw numeric values and decodes them with cftime.num2date, matching xarray's CF time decoding output exactly.

Parameters:

Name Type Description Default
ds Dataset

Open netCDF4 Dataset

 required

Returns:

Type Description
tuple[str | None, str | None]

Tuple of (start_time, end_time) as strings, or (None, None) if no time variable or time dimension is empty
Source code in packages/climate-ref/src/climate_ref/datasets/netcdf_utils.py 
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
def read_time_bounds(ds: netCDF4.Dataset) -> tuple[str | None, str | None]:
    """
    Read the first and last time values from a netCDF4 Dataset

    Reads only two raw numeric values and decodes them with ``cftime.num2date``,
    matching xarray's CF time decoding output exactly.

    Parameters
    ----------
    ds
        Open netCDF4 Dataset

    Returns
    -------
    :
        Tuple of (start_time, end_time) as strings, or (None, None) if no time variable
        or time dimension is empty
    """
    if "time" not in ds.variables:
        return None, None

    time_var = ds.variables["time"]
    n = len(time_var)
    if n == 0:
        return None, None

    units = getattr(time_var, "units", None)
    calendar = getattr(time_var, "calendar", "standard")

    if not units:
        return None, None

    times = cftime.num2date([time_var[0], time_var[-1]], units, calendar)
    return str(times[0]), str(times[1])

read_time_metadata(ds) #

Read time encoding metadata from a netCDF4 Dataset.

Parameters:

Name Type Description Default
ds Dataset

Open netCDF4 Dataset

 required

Returns:

Type Description
tuple[str | None, str | None]

Tuple of (time_units, calendar). Returns (None, None) if no time variable.
Source code in packages/climate-ref/src/climate_ref/datasets/netcdf_utils.py 
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
def read_time_metadata(ds: netCDF4.Dataset) -> tuple[str | None, str | None]:
    """
    Read time encoding metadata from a netCDF4 Dataset.

    Parameters
    ----------
    ds
        Open netCDF4 Dataset

    Returns
    -------
    :
        Tuple of (time_units, calendar). Returns (None, None) if no time variable.
    """
    if "time" not in ds.variables:
        return None, None

    time_var = ds.variables["time"]
    units = getattr(time_var, "units", None)
    calendar = getattr(time_var, "calendar", "standard")
    return units, calendar

read_variable_attrs(ds, variable_id, attr_names) #

Read attributes from a specific variable in a netCDF4 Dataset

Parameters:

Name Type Description Default
ds Dataset

Open netCDF4 Dataset

 required
variable_id str

Name of the variable to read attributes from

 required
attr_names list[str] | tuple[str, ...]

Attribute names to read

 required

Returns:

Type Description
dict[str, Any]

Dictionary mapping attribute names to values (None if missing or variable not found)
Source code in packages/climate-ref/src/climate_ref/datasets/netcdf_utils.py 
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
def read_variable_attrs(
    ds: netCDF4.Dataset,
    variable_id: str,
    attr_names: list[str] | tuple[str, ...],
) -> dict[str, Any]:
    """
    Read attributes from a specific variable in a netCDF4 Dataset

    Parameters
    ----------
    ds
        Open netCDF4 Dataset
    variable_id
        Name of the variable to read attributes from
    attr_names
        Attribute names to read

    Returns
    -------
    :
        Dictionary mapping attribute names to values (None if missing or variable not found)
    """
    if variable_id and variable_id in ds.variables:
        var = ds.variables[variable_id]
        return {attr: getattr(var, attr, None) for attr in attr_names}
    return {attr: None for attr in attr_names}

read_vertical_levels(ds) #

Count the number of vertical levels in a netCDF4 Dataset

Checks known vertical dimension names and returns the size of the first match.

Parameters:

Name Type Description Default
ds Dataset

Open netCDF4 Dataset

 required

Returns:

Type Description
int

Number of vertical levels (defaults to 1 if no vertical dimension found)
Source code in packages/climate-ref/src/climate_ref/datasets/netcdf_utils.py 
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
def read_vertical_levels(ds: netCDF4.Dataset) -> int:
    """
    Count the number of vertical levels in a netCDF4 Dataset

    Checks known vertical dimension names and returns the size of the first match.

    Parameters
    ----------
    ds
        Open netCDF4 Dataset

    Returns
    -------
    :
        Number of vertical levels (defaults to 1 if no vertical dimension found)
    """
    for dim_name in VERTICAL_DIM_NAMES:
        if dim_name in ds.dimensions:
            return len(ds.dimensions[dim_name])
    return 1