.. _model_views:

Model Videos
============

Videos from a model can be extracted after the simulation has been completed. This functionality internally uses the
orcaflex `AVI video extraction <https://www.orcina.com/webhelp/OrcFxAPI/Content/html/C_AVIFileInitialise.htm>`_ functionality
and can be used to extract multiple videos with different model views and video parameters for the same model

Defining video parameters
-------------------------

Video specification parameters can be defined using a sequence of ``qalx_orcaflexdata_models.VideoSpecification`` instances. Only ``start_time`` and ```end_time``
are required parameters and default values are used for any unspecified specification parameters.

.. note::

    The ``model_views`` parameter accepts an optional list of ``ModelView`` instances, and a video will be extracted for
    each one of these views. In the case that the ``model_views`` parameter is not specified, only a single video with
    the default model view will be extracted for that specific instance of VideoSpecification

.. code-block:: python

    from qalx_orcaflex import data_models as dm

    view_1 = dm.ModelView(ViewName="Front view", ViewAzimuth=270.0)
    view_2 = dm.ModelView(ViewName="Side view", ViewAzimuth=315.0)

    video_parameters = dm.VideoSpecification(
        start_time=-8.0,
        end_time=16.0,
        width=600,
        height=600,
        codec="MRLE",
        model_views=[view_1, view_2]
    )

Usage
-----

The example below shows how to extract two videos per model in a batch, using a different azimuth

.. code-block:: python

    import OrcFxAPI as ofx

    import qalx_orcaflex.data_models as dm
    from qalx_orcaflex.core import ModelSource, OrcaFlexBatch, QalxOrcaFlex

    qfx = QalxOrcaFlex()
    batch_options = dm.BatchOptions(
        batch_queue="example-batch-queue",
        sim_queue="example-sim-queue",
    )

    with OrcaFlexBatch(name=f"My Hang-off Batch", session=qfx,
                       batch_options=batch_options) as batch:
        m = ofx.Model()
        line = m.CreateObject(ofx.otLine, "My Line")
        for end_b_x in range(40, 80, 10):
            line.EndBX = end_b_x
            hang_off = 250 - end_b_x
            hang_off_info = dm.RawInfo(
                key="Hang-off point",
                value=hang_off
            )
            view_1 = dm.ModelView(
                ViewName=f"Front of hang-off @ {hang_off}m",
                ViewSize=50.0,
                ViewCentre=[end_b_x, 0.0, -21.5],
                ViewAzimuth=270.0,
                ViewGamma=0.0,
                ViewElevation=0.0,
                Width=1500,
                Height=800,
                GraphicsMode="WireFrame",
            )
            view_2 = dm.ModelView(
                ViewName=f"Side of hang-off @ {hang_off}m",
                ViewSize=50.0,
                ViewCentre=[end_b_x, 0.0, -21.5],
                ViewAzimuth=315.0,
                ViewGamma=0.0,
                ViewElevation=0.0,
                Width=1500,
                Height=800,
                GraphicsMode="WireFrame",
            )
            current_speed = dm.ModelInfo(
                object_name="Environment",
                data_name="RefCurrentSpeed",
                alias="Current Speed"
            )
            load_case_info = dm.LoadCaseInfo(
                model_info=[current_speed],
                raw_info=[hang_off_info])
            video_specification = [
                dm.VideoSpecification(
                    start_time=-8.0,
                    end_time=16.0,
                    model_views=[view_1, view_2]
                )
            ]
            batch.add(
                ModelSource(m, f"HangOff={hang_off}"),
                load_case_info=load_case_info,
                model_videos=[video_specification]
            )

.. note::
    The ``model_videos`` parameter passed to the ``add`` method needs to be a sequence, e.g. a ``list`` even in the case
    that only one instance of ``ViewSpecification`` is passed

In the case that no ``model_views`` are specified one video will be extracted with the default model view. In the following example, 
one such instance of ``ViewSpecification`` is created, along with a second instance:

.. code-block:: python

    import OrcFxAPI as ofx

    import qalx_orcaflex.data_models as dm
    from qalx_orcaflex.core import ModelSource, OrcaFlexBatch, QalxOrcaFlex

    qfx = QalxOrcaFlex()
    batch_options = dm.BatchOptions(
        batch_queue="example-batch-queue",
        sim_queue="example-sim-queue",
    )

    with OrcaFlexBatch(name=f"My Hang-off Batch", session=qfx,
                       batch_options=batch_options) as batch:
        m = ofx.Model()
        line = m.CreateObject(ofx.otLine, "My Line")
        for end_b_x in range(40, 80, 10):
            line.EndBX = end_b_x
            hang_off = 250 - end_b_x
            hang_off_info = dm.RawInfo(
                key="Hang-off point",
                value=hang_off
            )
            current_speed = dm.ModelInfo(
                object_name="Environment",
                data_name="RefCurrentSpeed",
                alias="Current Speed"
            )
            load_case_info = dm.LoadCaseInfo(
                model_info=[current_speed],
                raw_info=[hang_off_info])
            video_spec_1 = dm.VideoSpecification(
                start_time=4.0,
                end_time=10.0,
            )
            view_1 = dm.ModelView(
                ViewName=f"Front of hang-off @ {hang_off}m",
                ViewSize=50.0,
                ViewCentre=[end_b_x, 0.0, -21.5],
            )
            video_spec_2 = dm.VideoSpecification(
                start_time=-8.0,
                end_time=16.0,
                model_views=[view_1]
            )
            batch.add(
                ModelSource(m, f"HangOff={hang_off}"),
                load_case_info=load_case_info,
                model_videos=[video_spec_1, video_spec_2]
            )

In this example, for each model in the ``for`` loop, two videos will be extracted. One video from 4.0 sec to 10.0 sec and the
default model view and one video from -8.0 sec to 16.0 sec with the custom ``view_1`` view

Download videos from a batch
----------------------------

Videos specified on a batch can be then downloaded as shown below

.. code-block:: python

    from qalx_orcaflex.core import QalxOrcaFlex

    qfx = QalxOrcaFlex()

    qfx.save_batch_videos(
        batch_name="My Hang-off Batch",
        save_dir=r"C:/Users/AnneAlysis/OFX/Hang-offBatch/Videos"
    )

Extract videos from a sim file
------------------------------

In the case that you are interested in extracting a video from an existing simulation file(``.sim``), you can do this using
the functionality of the :ref:`video_bot`. A job can be send to the bot as shown below

.. code-block:: python

    from qalx_orcaflex.core import QalxOrcaFlex
    from qalx_orcaflex.data_models import OrcaFlexJob, ModelView, VideoSpecification

    qfx = QalxOrcaFlex()
    # Add the required videos specification
    model_view_1 = ModelView("View 1")
    model_view_2 = ModelView("View 2", ViewAzimuth=100.0, ViewElevation=1.0)
    model_view_3 = ModelView("View 3", ViewAzimuth=200.0, ViewElevation=-1.0)
    model_videos = [
        VideoSpecification(
            start_time=-8.0,
            end_time=15.9,
            width=200,
            height=200,
            model_views=[model_view_1, model_view_2],
        ),
        VideoSpecification(
            start_time=-8.0,
            end_time=15.9,
            width=200,
            height=200,
            codec="XVID",
            model_views=[model_view_3],
        ),
    ]
    # Add the sim file item
    qalx_sim_item = qfx.item.add(source="/path/to/sim_file/simple_model.sim")
    videos_job = OrcaFlexJob(model_videos=model_videos, sim_file=qalx_sim_item)
    qfx.save_batch_videos(
        batch_name="My Hang-off Batch",
        save_dir=r"C:/Users/AnneAlysis/OFX/Hang-offBatch/Videos"
    )
    # Send the job to the queue named "video-queue-example"
    video_set_guid = qfx.send_to_video_sim_queue(
        queue_name="video-queue-example", ofx_job=videos_job
    )
    # Once the video extraction process is complete, we can download the videos
    qfx.save_job_videos(video_set_guid, "/path/to/save/location")