Skip to content

Commit

Permalink
Merge pull request pytroll#2915 from sfinkens/seviri-doc-raw-metadata
Browse files Browse the repository at this point in the history
Improve SEVIRI metadata documentation
  • Loading branch information
mraspaud authored Sep 30, 2024
2 parents 5d2b1fd + 6a4f2af commit 201dc41
Show file tree
Hide file tree
Showing 2 changed files with 18 additions and 6 deletions.
2 changes: 1 addition & 1 deletion .github/workflows/ci.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -41,7 +41,7 @@ jobs:
miniforge-version: latest
python-version: ${{ matrix.python-version }}
activate-environment: test-environment
mamba-version: "*"
mamba-version: "1.5.10"
channels: conda-forge

- name: Set cache environment variables
Expand Down
22 changes: 17 additions & 5 deletions satpy/readers/seviri_base.py
Original file line number Diff line number Diff line change
Expand Up @@ -153,9 +153,21 @@
scn['IR_108']['y'] = mi
scn['IR_108'].sel(time=np.datetime64('2019-03-01T12:06:13.052000000'))
* Raw metadata from the file header can be included by setting the reader
argument ``include_raw_metadata=True`` (HRIT and Native format only). Note
that this comes with a performance penalty of up to 10% if raw metadata from
* HRIT and Native readers can add raw metadata from the file header, such
as calibration coefficients, to dataset attributes. Use the reader keyword
argument ``include_raw_metadata``. Here's an example for extracting
calibration coefficients from Native files.
.. code-block:: python
scene = satpy.Scene(filenames,
reader='seviri_l1b_native',
reader_kwargs={'include_raw_metadata': True})
scene.load(["IR_108"])
mda = scene["IR_108"].attrs["raw_metadata"]
coefs = mda["15_DATA_HEADER"]["RadiometricProcessing"]["Level15ImageCalibration"]
Note that this comes with a performance penalty of up to 10% if raw metadata from
multiple segments or scans need to be combined. By default, arrays with more
than 100 elements are excluded to limit the performance penalty. This
threshold can be adjusted using the ``mda_max_array_size`` reader keyword
Expand All @@ -164,8 +176,8 @@
.. code-block:: python
scene = satpy.Scene(filenames,
reader='seviri_l1b_hrit/native',
reader_kwargs={'include_raw_metadata': True,
reader='seviri_l1b_native',
reader_kwargs={'include_raw_metadata': True,
'mda_max_array_size': 1000})
References:
Expand Down

0 comments on commit 201dc41

Please sign in to comment.