scopeinpython.fluspect
======================

FLUSPECT-B-Cx, the SCOPE variant (``getFluspect.Cx.SCOPE``) that SCOPE's
own leaf-optics pipeline actually calls -- **not the same function** as
``toolsrtm.fluspect_cx`` despite very similar code. Direct port of
``SCOPEinR/R/fluspect_Cx_forSCOPE.R``.

Key differences from ``toolsrtm.fluspect_cx``:

- A caller-configurable ``step`` (nm) controls the ``Mb``/``Mf``
  excitation-emission matrix resolution. SCOPE's own default ``step=5``
  gives 53x71 matrices; ``step=1`` gives 211x351 (like the non-SCOPE
  version, but see below -- still not numerically identical).
- The SIF response is scaled by ``step`` itself, not a fixed constant
  (the non-SCOPE version always uses ``int=5`` regardless of its own
  fixed 1 nm grid).
- A single combined ``Mb``/``Mf`` pair (one ``phi`` spectrum), same as
  the non-SCOPE Cx version.
- Uses ``match()`` (correct) for locating excitation/emission wavelengths
  within the full grid -- does **not** have the ``intersect()`` indexing
  bug documented in :func:`toolsrtm.fluspect.fluspect_cx`.
- Uses ``SCOPEinR::optipar2021.Pro.CX`` as its real default optical
  parameter table (``optipar2017.ProspectD`` is missing ``Kp``/``Kcbc``
  data needed for the PROSPECT-PRO/Cx ``Kall`` formula).

.. automodule:: scopeinpython.fluspect
   :members:
   :undoc-members:
   :show-inheritance: