webview.py

"""
 @file
 @brief This file loads the interactive HTML timeline
 @author Noah Figg <eggmunkee@hotmail.com>
 @author Jonathan Thomas <jonathan@openshot.org>
 @author Olivier Girard <eolinwen@gmail.com>

 @section LICENSE

 Copyright (c) 2008-2018 OpenShot Studios, LLC
 (http://www.openshotstudios.com). This file is part of
 OpenShot Video Editor (http://www.openshot.org), an open-source project
 dedicated to delivering high quality video editing and animation solutions
 to the world.

 OpenShot Video Editor is free software: you can redistribute it and/or modify
 it under the terms of the GNU General Public License as published by
 the Free Software Foundation, either version 3 of the License, or
 (at your option) any later version.

 OpenShot Video Editor is distributed in the hope that it will be useful,
 but WITHOUT ANY WARRANTY; without even the implied warranty of
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 GNU General Public License for more details.

 You should have received a copy of the GNU General Public License
 along with OpenShot Library.  If not, see <http://www.gnu.org/licenses/>.
 """

import os
import time
from copy import deepcopy
from functools import partial
from random import uniform
from operator import itemgetter
import logging
import json
import openshot  # Python module for libopenshot (required video editing module installed separately)

from PyQt5.QtCore import QFileInfo, pyqtSlot, QUrl, Qt, QCoreApplication, QTimer
from PyQt5.QtGui import QCursor, QKeySequence, QColor
from PyQt5.QtWidgets import QMenu, QDialog

from classes import info, updates
from classes import settings
from classes.app import get_app
from classes.logger import log
from classes.query import File, Clip, Transition, Track
from classes.waveform import get_audio_data
from classes.conversion import zoomToSeconds, secondsToZoom
from classes.effect_init import effect_options

# Constants used by this file
JS_SCOPE_SELECTOR = "$('body').scope()"

MENU_FADE_NONE = 0
MENU_FADE_IN_FAST = 1
MENU_FADE_IN_SLOW = 2
MENU_FADE_OUT_FAST = 3
MENU_FADE_OUT_SLOW = 4
MENU_FADE_IN_OUT_FAST = 5
MENU_FADE_IN_OUT_SLOW = 6

MENU_ROTATE_NONE = 0
MENU_ROTATE_90_RIGHT = 1
MENU_ROTATE_90_LEFT = 2
MENU_ROTATE_180_FLIP = 3

MENU_LAYOUT_NONE = 0
MENU_LAYOUT_CENTER = 1
MENU_LAYOUT_TOP_LEFT = 2
MENU_LAYOUT_TOP_RIGHT = 3
MENU_LAYOUT_BOTTOM_LEFT = 4
MENU_LAYOUT_BOTTOM_RIGHT = 5
MENU_LAYOUT_ALL_WITH_ASPECT = 6
MENU_LAYOUT_ALL_WITHOUT_ASPECT = 7

MENU_ALIGN_LEFT = 0
MENU_ALIGN_RIGHT = 1

MENU_ANIMATE_NONE = 0
MENU_ANIMATE_IN_50_100 = 1
MENU_ANIMATE_IN_75_100 = 2
MENU_ANIMATE_IN_100_150 = 3
MENU_ANIMATE_OUT_100_75 = 4
MENU_ANIMATE_OUT_100_50 = 5
MENU_ANIMATE_OUT_150_100 = 6
MENU_ANIMATE_CENTER_TOP = 7
MENU_ANIMATE_CENTER_LEFT = 8
MENU_ANIMATE_CENTER_RIGHT = 9
MENU_ANIMATE_CENTER_BOTTOM = 10
MENU_ANIMATE_TOP_CENTER = 11
MENU_ANIMATE_LEFT_CENTER = 12
MENU_ANIMATE_RIGHT_CENTER = 13
MENU_ANIMATE_BOTTOM_CENTER = 14
MENU_ANIMATE_TOP_BOTTOM = 15
MENU_ANIMATE_LEFT_RIGHT = 16
MENU_ANIMATE_RIGHT_LEFT = 17
MENU_ANIMATE_BOTTOM_TOP = 18
MENU_ANIMATE_RANDOM = 19

MENU_VOLUME_NONE = 1
MENU_VOLUME_FADE_IN_FAST = 2
MENU_VOLUME_FADE_IN_SLOW = 3
MENU_VOLUME_FADE_OUT_FAST = 4
MENU_VOLUME_FADE_OUT_SLOW = 5
MENU_VOLUME_FADE_IN_OUT_FAST = 6
MENU_VOLUME_FADE_IN_OUT_SLOW = 7
MENU_VOLUME_LEVEL_100 = 100
MENU_VOLUME_LEVEL_90 = 90
MENU_VOLUME_LEVEL_80 = 80
MENU_VOLUME_LEVEL_70 = 70
MENU_VOLUME_LEVEL_60 = 60
MENU_VOLUME_LEVEL_50 = 50
MENU_VOLUME_LEVEL_40 = 40
MENU_VOLUME_LEVEL_30 = 30
MENU_VOLUME_LEVEL_20 = 20
MENU_VOLUME_LEVEL_10 = 10
MENU_VOLUME_LEVEL_0 = 0

MENU_TRANSFORM = 0

MENU_TIME_NONE = 0
MENU_TIME_FORWARD = 1
MENU_TIME_BACKWARD = 2
MENU_TIME_FREEZE = 3
MENU_TIME_FREEZE_ZOOM = 4

MENU_COPY_ALL = -1
MENU_COPY_CLIP = 0
MENU_COPY_KEYFRAMES_ALL = 1
MENU_COPY_KEYFRAMES_ALPHA = 2
MENU_COPY_KEYFRAMES_SCALE = 3
MENU_COPY_KEYFRAMES_ROTATE = 4
MENU_COPY_KEYFRAMES_LOCATION = 5
MENU_COPY_KEYFRAMES_TIME = 6
MENU_COPY_KEYFRAMES_VOLUME = 7
MENU_COPY_EFFECTS = 8
MENU_PASTE = 9

MENU_COPY_TRANSITION = 10
MENU_COPY_KEYFRAMES_BRIGHTNESS = 11
MENU_COPY_KEYFRAMES_CONTRAST = 12

MENU_SLICE_KEEP_BOTH = 0
MENU_SLICE_KEEP_LEFT = 1
MENU_SLICE_KEEP_RIGHT = 2

MENU_SPLIT_AUDIO_SINGLE = 0
MENU_SPLIT_AUDIO_MULTIPLE = 1

# Import shenanigans
WEBVIEW_LOADED = None

if info.WEB_BACKEND and info.WEB_BACKEND == "webkit":
    from .webview_backend.webkit import TimelineWebKitView
    WebViewClass = TimelineWebKitView
    WEBVIEW_LOADED = True
elif info.WEB_BACKEND and info.WEB_BACKEND == "webengine":
    from .webview_backend.webengine import TimelineWebEngineView
    WebViewClass = TimelineWebEngineView
    WEBVIEW_LOADED = True
else:
    try:
        from .webview_backend.webengine import TimelineWebEngineView as WebViewClass
        WEBVIEW_LOADED = True
    except ImportError as ex:
        try:
            from .webview_backend.webkit import TimelineWebKitView as WebViewClass
            WEBVIEW_LOADED = True
        except ImportError:
            log.error("Import failure loading WebKit backend", exc_info=1)
        finally:
            if not WEBVIEW_LOADED:
                raise RuntimeError(
                    "Need PyQt5.QtWebEngine (or PyQt5.QtWebView on Win32)"
                    ) from ex


class TimelineWebView(updates.UpdateInterface, WebViewClass):
    """ A Web(Engine)View QWidget used to load the Timeline """

    # Path to html file
    html_path = os.path.join(info.PATH, 'timeline', 'index.html')

    @pyqtSlot()
    def page_ready(self):
        """Document.Ready event has fired, and is initialized"""
        self.document_is_ready = True

    @pyqtSlot(result=str)
    def get_thumb_address(self):
        """Return the thumbnail HTTP server address"""
        thumb_server_details = self.window.http_server_thread.server_address
        while not thumb_server_details:
            log.info('No HTTP thumbnail server found yet... keep waiting...')
            time.sleep(0.25)
            thumb_server_details = self.window.http_server_thread.server_address

        thumb_address = "http://%s:%s/thumbnails/" % (thumb_server_details[0], thumb_server_details[1])
        return thumb_address

    # This method is invoked by the UpdateManager each time a change happens (i.e UpdateInterface)
    def changed(self, action):
        # Remove unused action attribute (old_values)
        action = deepcopy(action)
        action.old_values = {}

        # Send a JSON version of the UpdateAction to the timeline webview method: applyJsonDiff()
        if action.type == "load":
            # Set thumbnail server
            self.run_js(JS_SCOPE_SELECTOR + ".setThumbAddress('" + self.get_thumb_address() + "');")

            _ = get_app()._tr
            # Initialize translated track name
            self.run_js(JS_SCOPE_SELECTOR + ".setTrackLabel('" + _("Track %s") + "');")

            # Load entire project data
            self.run_js(JS_SCOPE_SELECTOR + ".loadJson(" + action.json() + ");")

        elif action.key[0] != "files":
            # Apply diff to part of project data
            self.run_js(JS_SCOPE_SELECTOR + ".applyJsonDiff([" + action.json() + "]);")

        # Reset the scale when loading new JSON
        if action.type == "load":
            # Set the scale again (to project setting)
            initial_scale = get_app().project.get("scale") or 15
            self.window.sliderZoom.setValue(secondsToZoom(initial_scale))

            # The setValue() above doesn't trigger update_zoom when a project file is
            # loaded on the command line (too early?), so also call the JS directly
            self.run_js(JS_SCOPE_SELECTOR + ".setScale(" + str(initial_scale) + ", 0);")

    # Javascript callable function to update the project data when a clip changes
    @pyqtSlot(str, bool, bool, bool)
    def update_clip_data(self, clip_json, only_basic_props=True, ignore_reader=False, ignore_refresh=False):
        """ Create an updateAction and send it to the update manager """

        # read clip json
        try:
            if not isinstance(clip_json, dict):
                clip_data = json.loads(clip_json)
            else:
                clip_data = clip_json
        except Exception:
            # Failed to parse json, do nothing
            log.warning('Failed to parse clip JSON data', exc_info=1)

        # Search for matching clip in project data (if any)
        existing_clip = Clip.get(id=clip_data["id"])
        if not existing_clip:
            # Create a new clip (if not exists)
            existing_clip = Clip()

        # Update clip data
        existing_clip.data = clip_data

        # Remove unneeded properties (since they don't change here... this is a performance boost)
        if only_basic_props:
            existing_clip.data = {}
            existing_clip.data["id"] = clip_data["id"]
            existing_clip.data["layer"] = clip_data["layer"]
            existing_clip.data["position"] = clip_data["position"]
            existing_clip.data["start"] = clip_data["start"]
            existing_clip.data["end"] = clip_data["end"]

        # Always remove the Reader attribute (since nothing updates it,
        # and we are wrapping clips in FrameMappers anyway)
        if ignore_reader and "reader" in existing_clip.data:
            existing_clip.data.pop("reader")

        # Save clip
        existing_clip.save()

        # Update the preview and reselect current frame in properties
        if not ignore_refresh:
            self.window.refreshFrameSignal.emit()
            self.window.propertyTableView.select_frame(self.window.preview_thread.player.Position())

    # Add missing transition
    @pyqtSlot(str)
    def add_missing_transition(self, transition_json):

        transition_details = json.loads(transition_json)

        # Get FPS from project
        fps = get_app().project.get("fps")
        fps_float = float(fps["num"]) / float(fps["den"])

        # Open up QtImageReader for transition Image
        transition_reader = openshot.QtImageReader(
            os.path.join(info.PATH, "transitions", "common", "fade.svg"))

        # Generate transition object
        transition_object = openshot.Mask()

        # Set brightness and contrast, to correctly transition for overlapping clips
        brightness = transition_object.brightness
        brightness.AddPoint(1, 1.0, openshot.BEZIER)
        brightness.AddPoint(round(transition_details["end"] * fps_float) + 1, -1.0, openshot.BEZIER)
        contrast = openshot.Keyframe(3.0)

        # Create transition dictionary
        transitions_data = {
            "id": get_app().project.generate_id(),
            "layer": transition_details["layer"],
            "title": "Transition",
            "type": "Mask",
            "position": transition_details["position"],
            "start": transition_details["start"],
            "end": transition_details["end"],
            "brightness": json.loads(brightness.Json()),
            "contrast": json.loads(contrast.Json()),
            "reader": json.loads(transition_reader.Json()),
            "replace_image": False
        }

        # Send to update manager
        self.update_transition_data(transitions_data, only_basic_props=False)

    # Javascript callable function to update the project data when a transition changes
    @pyqtSlot(str, bool, bool)
    def update_transition_data(self, transition_json, only_basic_props=True, ignore_refresh=False):
        """ Create an updateAction and send it to the update manager """

        # read clip json
        if not isinstance(transition_json, dict):
            transition_data = json.loads(transition_json)
        else:
            transition_data = transition_json

        # Search for matching clip in project data (if any)
        existing_item = Transition.get(id=transition_data["id"])
        needs_resize = True
        if not existing_item:
            # Create a new clip (if not exists)
            existing_item = Transition()
            needs_resize = False
        existing_item.data = transition_data

        # Get FPS from project
        fps = get_app().project.get("fps")
        fps_float = float(fps["num"]) / float(fps["den"])
        duration = existing_item.data["end"] - existing_item.data["start"]

        # Update the brightness and contrast keyframes to match the duration of the transition
        # This is a hack until I can think of something better
        brightness = None
        contrast = None
        if needs_resize:
            # Adjust transition's brightness keyframes to match the size of the transition
            brightness = existing_item.data["brightness"]
            if len(brightness["Points"]) > 1:
                # If multiple points, move the final one to the 'new' end
                brightness["Points"][-1]["co"]["X"] = round(duration * fps_float) + 1

            # Adjust transition's contrast keyframes to match the size of the transition
            contrast = existing_item.data["contrast"]
            if len(contrast["Points"]) > 1:
                # If multiple points, move the final one to the 'new' end
                contrast["Points"][-1]["co"]["X"] = round(duration * fps_float) + 1
        else:
            # Create new brightness and contrast Keyframes
            b = openshot.Keyframe()
            b.AddPoint(1, 1.0, openshot.BEZIER)
            b.AddPoint(round(duration * fps_float) + 1, -1.0, openshot.BEZIER)
            brightness = json.loads(b.Json())

        # Only include the basic properties (performance boost)
        if only_basic_props:
            existing_item.data = {}
            existing_item.data["id"] = transition_data["id"]
            existing_item.data["layer"] = transition_data["layer"]
            existing_item.data["position"] = transition_data["position"]
            existing_item.data["start"] = transition_data["start"]
            existing_item.data["end"] = transition_data["end"]

            log.debug('transition start: %s' % transition_data["start"])
            log.debug('transition end: %s' % transition_data["end"])

            if brightness:
                existing_item.data["brightness"] = brightness
            if contrast:
                existing_item.data["contrast"] = contrast

        # Save transition
        existing_item.save()

        # Update the preview and reselct current frame in properties
        if not ignore_refresh:
            self.window.refreshFrameSignal.emit()
            self.window.propertyTableView.select_frame(self.window.preview_thread.player.Position())

    # Prevent default context menu, and ignore, so that javascript can intercept
    def contextMenuEvent(self, event):
        event.ignore()

    # Javascript callable function to show clip or transition content menus, passing in type to show
    @pyqtSlot(float)
    def ShowPlayheadMenu(self, position=None):
        log.debug('ShowPlayheadMenu: %s' % position)

        # Get translation method
        _ = get_app()._tr

        # Get list of intercepting clips with position (if any)
        intersecting_clips = Clip.filter(intersect=position)
        intersecting_trans = Transition.filter(intersect=position)

        menu = QMenu(self)
        if intersecting_clips or intersecting_trans:
            # Get list of clip ids
            clip_ids = [c.id for c in intersecting_clips]
            trans_ids = [t.id for t in intersecting_trans]

            # Add split clip menu
            Slice_Menu = QMenu(_("Slice All"), self)
            Slice_Keep_Both = Slice_Menu.addAction(_("Keep Both Sides"))
            Slice_Keep_Both.setShortcut(QKeySequence(self.window.getShortcutByName("sliceAllKeepBothSides")))
            Slice_Keep_Both.triggered.connect(partial(
                self.Slice_Triggered, MENU_SLICE_KEEP_BOTH, clip_ids, trans_ids, position))
            Slice_Keep_Left = Slice_Menu.addAction(_("Keep Left Side"))
            Slice_Keep_Left.setShortcut(QKeySequence(self.window.getShortcutByName("sliceAllKeepLeftSide")))
            Slice_Keep_Left.triggered.connect(partial(
                self.Slice_Triggered, MENU_SLICE_KEEP_LEFT, clip_ids, trans_ids, position))
            Slice_Keep_Right = Slice_Menu.addAction(_("Keep Right Side"))
            Slice_Keep_Right.setShortcut(QKeySequence(self.window.getShortcutByName("sliceAllKeepRightSide")))
            Slice_Keep_Right.triggered.connect(partial(
                self.Slice_Triggered, MENU_SLICE_KEEP_RIGHT, clip_ids, trans_ids, position))
            menu.addMenu(Slice_Menu)
            return menu.exec_(QCursor.pos())

    @pyqtSlot(str)
    def ShowEffectMenu(self, effect_id=None):
        log.debug('ShowEffectMenu: %s' % effect_id)

        # Set the selected clip (if needed)
        self.window.addSelection(effect_id, 'effect', True)

        menu = QMenu(self)
        # Properties
        menu.addAction(self.window.actionProperties)

        # Remove Effect Menu
        menu.addSeparator()
        menu.addAction(self.window.actionRemoveEffect)
        return menu.exec_(QCursor.pos())

    @pyqtSlot(float, int)
    def ShowTimelineMenu(self, position, layer_id):
        log.debug('ShowTimelineMenu: position: %s, layer: %s' % (position, layer_id))

        # Get translation method
        _ = get_app()._tr

        # Get list of clipboard items (that are complete clips or transitions)
        # i.e. ignore partial clipboard items (keyframes / effects / etc...)
        clipboard_clip_ids = [k for k, v in self.copy_clipboard.items() if v.get('id')]
        clipboard_tran_ids = [k for k, v in self.copy_transition_clipboard.items() if v.get('id')]

        # Paste Menu (if entire clips or transitions are copied)
        have_clipboard = (
            (self.copy_clipboard or self.copy_transition_clipboard)
            and (len(clipboard_clip_ids) + len(clipboard_tran_ids) > 0)
        )

        if not have_clipboard:
            return

        menu = QMenu(self)
        Paste_Clip = menu.addAction(_("Paste"))
        Paste_Clip.setShortcut(QKeySequence(self.window.getShortcutByName("pasteAll")))
        Paste_Clip.triggered.connect(
            partial(self.Paste_Triggered, MENU_PASTE, float(position), int(layer_id), [], [])
        )

        return menu.exec_(QCursor.pos())

    @pyqtSlot(str)
    def ShowClipMenu(self, clip_id=None):
        log.debug('ShowClipMenu: %s' % clip_id)

        # Get translation method
        _ = get_app()._tr

        # Get existing clip object
        clip = Clip.get(id=clip_id)
        if not clip:
            # Not a valid clip id
            return

        # Set the selected clip (if needed)
        if clip_id not in self.window.selected_clips:
            self.window.addSelection(clip_id, 'clip')
        # Get list of selected clips
        clip_ids = self.window.selected_clips
        tran_ids = self.window.selected_transitions

        # Get framerate
        fps = get_app().project.get("fps")
        fps_float = float(fps["num"]) / float(fps["den"])

        # Get playhead position
        playhead_position = float(self.window.preview_thread.current_frame) / fps_float

        # Create blank context menu
        menu = QMenu(self)

        # Copy Menu
        if len(tran_ids) + len(clip_ids) > 1:
            # Show Copy All menu (clips and transitions are selected)
            Copy_All = menu.addAction(_("Copy"))
            Copy_All.setShortcut(QKeySequence(self.window.getShortcutByName("copyAll")))
            Copy_All.triggered.connect(partial(self.Copy_Triggered, MENU_COPY_ALL, clip_ids, tran_ids))
        else:
            # Only a single clip is selected (Show normal copy menus)
            Copy_Menu = QMenu(_("Copy"), self)
            Copy_Clip = Copy_Menu.addAction(_("Clip"))
            Copy_Clip.setShortcut(QKeySequence(self.window.getShortcutByName("copyAll")))
            Copy_Clip.triggered.connect(partial(self.Copy_Triggered, MENU_COPY_CLIP, [clip_id], []))

            Keyframe_Menu = QMenu(_("Keyframes"), self)
            Copy_Keyframes_All = Keyframe_Menu.addAction(_("All"))
            Copy_Keyframes_All.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_ALL, [clip_id], []))
            Keyframe_Menu.addSeparator()
            Copy_Keyframes_Alpha = Keyframe_Menu.addAction(_("Alpha"))
            Copy_Keyframes_Alpha.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_ALPHA, [clip_id], []))
            Copy_Keyframes_Scale = Keyframe_Menu.addAction(_("Scale"))
            Copy_Keyframes_Scale.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_SCALE, [clip_id], []))
            Copy_Keyframes_Rotate = Keyframe_Menu.addAction(_("Rotation"))
            Copy_Keyframes_Rotate.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_ROTATE, [clip_id], []))
            Copy_Keyframes_Locate = Keyframe_Menu.addAction(_("Location"))
            Copy_Keyframes_Locate.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_LOCATION, [clip_id], []))
            Copy_Keyframes_Time = Keyframe_Menu.addAction(_("Time"))
            Copy_Keyframes_Time.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_TIME, [clip_id], []))
            Copy_Keyframes_Volume = Keyframe_Menu.addAction(_("Volume"))
            Copy_Keyframes_Volume.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_VOLUME, [clip_id], []))

            # Only add copy->effects and copy->keyframes if 1 clip is selected
            Copy_Effects = Copy_Menu.addAction(_("Effects"))
            Copy_Effects.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_EFFECTS, [clip_id], []))
            Copy_Menu.addMenu(Keyframe_Menu)
            menu.addMenu(Copy_Menu)

        # Get list of clipboard items (that are complete clips or transitions)
        # i.e. ignore partial clipboard items (keyframes / effects / etc...)
        clipboard_clip_ids = [k for k, v in self.copy_clipboard.items() if v.get('id')]
        clipboard_tran_ids = [k for k, v in self.copy_transition_clipboard.items() if v.get('id')]
        # Determine if the paste menu should be shown
        if self.copy_clipboard and len(clipboard_clip_ids) + len(clipboard_tran_ids) == 0:
            # Paste Menu (Only show if partial clipboard available)
            Paste_Clip = menu.addAction(_("Paste"))
            Paste_Clip.triggered.connect(partial(self.Paste_Triggered, MENU_PASTE, 0.0, 0, clip_ids, []))

        menu.addSeparator()

        # Alignment Menu (if multiple selections)
        if len(clip_ids) > 1:
            Alignment_Menu = QMenu(_("Align"), self)
            Align_Left = Alignment_Menu.addAction(_("Left"))
            Align_Left.triggered.connect(partial(self.Align_Triggered, MENU_ALIGN_LEFT, clip_ids, tran_ids))
            Align_Right = Alignment_Menu.addAction(_("Right"))
            Align_Right.triggered.connect(partial(self.Align_Triggered, MENU_ALIGN_RIGHT, clip_ids, tran_ids))

            # Add menu to parent
            menu.addMenu(Alignment_Menu)

        # Fade In Menu
        Fade_Menu = QMenu(_("Fade"), self)
        Fade_None = Fade_Menu.addAction(_("No Fade"))
        Fade_None.triggered.connect(partial(self.Fade_Triggered, MENU_FADE_NONE, clip_ids))
        Fade_Menu.addSeparator()
        for position, position_label in [
            ("Start of Clip", _("Start of Clip")),
            ("End of Clip", _("End of Clip")),
            ("Entire Clip", _("Entire Clip"))
        ]:
            Position_Menu = QMenu(position_label, self)

            if position == "Start of Clip":
                Fade_In_Fast = Position_Menu.addAction(_("Fade In (Fast)"))
                Fade_In_Fast.triggered.connect(partial(
                    self.Fade_Triggered, MENU_FADE_IN_FAST, clip_ids, position))
                Fade_In_Slow = Position_Menu.addAction(_("Fade In (Slow)"))
                Fade_In_Slow.triggered.connect(partial(
                    self.Fade_Triggered, MENU_FADE_IN_SLOW, clip_ids, position))

            elif position == "End of Clip":
                Fade_Out_Fast = Position_Menu.addAction(_("Fade Out (Fast)"))
                Fade_Out_Fast.triggered.connect(partial(
                    self.Fade_Triggered, MENU_FADE_OUT_FAST, clip_ids, position))
                Fade_Out_Slow = Position_Menu.addAction(_("Fade Out (Slow)"))
                Fade_Out_Slow.triggered.connect(partial(
                    self.Fade_Triggered, MENU_FADE_OUT_SLOW, clip_ids, position))

            else:
                Fade_In_Out_Fast = Position_Menu.addAction(_("Fade In and Out (Fast)"))
                Fade_In_Out_Fast.triggered.connect(partial(
                    self.Fade_Triggered, MENU_FADE_IN_OUT_FAST, clip_ids, position))
                Fade_In_Out_Slow = Position_Menu.addAction(_("Fade In and Out (Slow)"))
                Fade_In_Out_Slow.triggered.connect(partial(
                    self.Fade_Triggered, MENU_FADE_IN_OUT_SLOW, clip_ids, position))
                Position_Menu.addSeparator()
                Fade_In_Slow = Position_Menu.addAction(_("Fade In (Entire Clip)"))
                Fade_In_Slow.triggered.connect(partial(
                    self.Fade_Triggered, MENU_FADE_IN_SLOW, clip_ids, position))
                Fade_Out_Slow = Position_Menu.addAction(_("Fade Out (Entire Clip)"))
                Fade_Out_Slow.triggered.connect(partial(
                    self.Fade_Triggered, MENU_FADE_OUT_SLOW, clip_ids, position))

            Fade_Menu.addMenu(Position_Menu)
        menu.addMenu(Fade_Menu)

        # Animate Menu
        Animate_Menu = QMenu(_("Animate"), self)
        Animate_None = Animate_Menu.addAction(_("No Animation"))
        Animate_None.triggered.connect(partial(self.Animate_Triggered, MENU_ANIMATE_NONE, clip_ids))
        Animate_Menu.addSeparator()
        for position, position_label in [
            ("Start of Clip", _("Start of Clip")),
            ("End of Clip", _("End of Clip")),
            ("Entire Clip", _("Entire Clip"))
        ]:
            Position_Menu = QMenu(position_label, self)

            # Scale
            Scale_Menu = QMenu(_("Zoom"), self)
            Animate_In_50_100 = Scale_Menu.addAction(_("Zoom In (50% to 100%)"))
            Animate_In_50_100.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_IN_50_100, clip_ids, position))
            Animate_In_75_100 = Scale_Menu.addAction(_("Zoom In (75% to 100%)"))
            Animate_In_75_100.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_IN_75_100, clip_ids, position))
            Animate_In_100_150 = Scale_Menu.addAction(_("Zoom In (100% to 150%)"))
            Animate_In_100_150.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_IN_100_150, clip_ids, position))
            Animate_Out_100_75 = Scale_Menu.addAction(_("Zoom Out (100% to 75%)"))
            Animate_Out_100_75.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_OUT_100_75, clip_ids, position))
            Animate_Out_100_50 = Scale_Menu.addAction(_("Zoom Out (100% to 50%)"))
            Animate_Out_100_50.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_OUT_100_50, clip_ids, position))
            Animate_Out_150_100 = Scale_Menu.addAction(_("Zoom Out (150% to 100%)"))
            Animate_Out_150_100.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_OUT_150_100, clip_ids, position))
            Position_Menu.addMenu(Scale_Menu)

            # Center to Edge
            Center_Edge_Menu = QMenu(_("Center to Edge"), self)
            Animate_Center_Top = Center_Edge_Menu.addAction(_("Center to Top"))
            Animate_Center_Top.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_CENTER_TOP, clip_ids, position))
            Animate_Center_Left = Center_Edge_Menu.addAction(_("Center to Left"))
            Animate_Center_Left.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_CENTER_LEFT, clip_ids, position))
            Animate_Center_Right = Center_Edge_Menu.addAction(_("Center to Right"))
            Animate_Center_Right.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_CENTER_RIGHT, clip_ids, position))
            Animate_Center_Bottom = Center_Edge_Menu.addAction(_("Center to Bottom"))
            Animate_Center_Bottom.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_CENTER_BOTTOM, clip_ids, position))
            Position_Menu.addMenu(Center_Edge_Menu)

            # Edge to Center
            Edge_Center_Menu = QMenu(_("Edge to Center"), self)
            Animate_Top_Center = Edge_Center_Menu.addAction(_("Top to Center"))
            Animate_Top_Center.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_TOP_CENTER, clip_ids, position))
            Animate_Left_Center = Edge_Center_Menu.addAction(_("Left to Center"))
            Animate_Left_Center.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_LEFT_CENTER, clip_ids, position))
            Animate_Right_Center = Edge_Center_Menu.addAction(_("Right to Center"))
            Animate_Right_Center.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_RIGHT_CENTER, clip_ids, position))
            Animate_Bottom_Center = Edge_Center_Menu.addAction(_("Bottom to Center"))
            Animate_Bottom_Center.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_BOTTOM_CENTER, clip_ids, position))
            Position_Menu.addMenu(Edge_Center_Menu)

            # Edge to Edge
            Edge_Edge_Menu = QMenu(_("Edge to Edge"), self)
            Animate_Top_Bottom = Edge_Edge_Menu.addAction(_("Top to Bottom"))
            Animate_Top_Bottom.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_TOP_BOTTOM, clip_ids, position))
            Animate_Left_Right = Edge_Edge_Menu.addAction(_("Left to Right"))
            Animate_Left_Right.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_LEFT_RIGHT, clip_ids, position))
            Animate_Right_Left = Edge_Edge_Menu.addAction(_("Right to Left"))
            Animate_Right_Left.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_RIGHT_LEFT, clip_ids, position))
            Animate_Bottom_Top = Edge_Edge_Menu.addAction(_("Bottom to Top"))
            Animate_Bottom_Top.triggered.connect(partial(
                self.Animate_Triggered, MENU_ANIMATE_BOTTOM_TOP, clip_ids, position))
            Position_Menu.addMenu(Edge_Edge_Menu)

            # Random Animation
            Position_Menu.addSeparator()
            Random = Position_Menu.addAction(_("Random"))
            Random.triggered.connect(partial(self.Animate_Triggered, MENU_ANIMATE_RANDOM, clip_ids, position))

            # Add Sub-Menu's to Position menu
            Animate_Menu.addMenu(Position_Menu)

        # Add Each position menu
        menu.addMenu(Animate_Menu)

        # Rotate Menu
        Rotation_Menu = QMenu(_("Rotate"), self)
        Rotation_None = Rotation_Menu.addAction(_("No Rotation"))
        Rotation_None.triggered.connect(partial(
            self.Rotate_Triggered, MENU_ROTATE_NONE, clip_ids))
        Rotation_Menu.addSeparator()
        Rotation_90_Right = Rotation_Menu.addAction(_("Rotate 90 (Right)"))
        Rotation_90_Right.triggered.connect(partial(
            self.Rotate_Triggered, MENU_ROTATE_90_RIGHT, clip_ids))
        Rotation_90_Left = Rotation_Menu.addAction(_("Rotate 90 (Left)"))
        Rotation_90_Left.triggered.connect(partial(
            self.Rotate_Triggered, MENU_ROTATE_90_LEFT, clip_ids))
        Rotation_180_Flip = Rotation_Menu.addAction(_("Rotate 180 (Flip)"))
        Rotation_180_Flip.triggered.connect(partial(
            self.Rotate_Triggered, MENU_ROTATE_180_FLIP, clip_ids))
        menu.addMenu(Rotation_Menu)

        # Layout Menu
        Layout_Menu = QMenu(_("Layout"), self)
        Layout_None = Layout_Menu.addAction(_("Reset Layout"))
        Layout_None.triggered.connect(partial(
            self.Layout_Triggered, MENU_LAYOUT_NONE, clip_ids))
        Layout_Menu.addSeparator()
        Layout_Center = Layout_Menu.addAction(_("1/4 Size - Center"))
        Layout_Center.triggered.connect(partial(
            self.Layout_Triggered, MENU_LAYOUT_CENTER, clip_ids))
        Layout_Top_Left = Layout_Menu.addAction(_("1/4 Size - Top Left"))
        Layout_Top_Left.triggered.connect(partial(
            self.Layout_Triggered, MENU_LAYOUT_TOP_LEFT, clip_ids))
        Layout_Top_Right = Layout_Menu.addAction(_("1/4 Size - Top Right"))
        Layout_Top_Right.triggered.connect(partial(
            self.Layout_Triggered, MENU_LAYOUT_TOP_RIGHT, clip_ids))
        Layout_Bottom_Left = Layout_Menu.addAction(_("1/4 Size - Bottom Left"))
        Layout_Bottom_Left.triggered.connect(partial(
            self.Layout_Triggered, MENU_LAYOUT_BOTTOM_LEFT, clip_ids))
        Layout_Bottom_Right = Layout_Menu.addAction(_("1/4 Size - Bottom Right"))
        Layout_Bottom_Right.triggered.connect(partial(
            self.Layout_Triggered, MENU_LAYOUT_BOTTOM_RIGHT, clip_ids))
        Layout_Menu.addSeparator()
        Layout_Bottom_All_With_Aspect = Layout_Menu.addAction(_("Show All (Maintain Ratio)"))
        Layout_Bottom_All_With_Aspect.triggered.connect(partial(
            self.Layout_Triggered, MENU_LAYOUT_ALL_WITH_ASPECT, clip_ids))
        Layout_Bottom_All_Without_Aspect = Layout_Menu.addAction(_("Show All (Distort)"))
        Layout_Bottom_All_Without_Aspect.triggered.connect(partial(
            self.Layout_Triggered, MENU_LAYOUT_ALL_WITHOUT_ASPECT, clip_ids))
        menu.addMenu(Layout_Menu)

        # Time Menu
        Time_Menu = QMenu(_("Time"), self)
        Time_None = Time_Menu.addAction(_("Reset Time"))
        Time_None.triggered.connect(partial(self.Time_Triggered, MENU_TIME_NONE, clip_ids, '1X'))
        Time_Menu.addSeparator()
        for speed, speed_values in [
            (_("Normal"), ['1X']),
            (_("Fast"), ['2X', '4X', '8X', '16X']),
            (_("Slow"), ['1/2X', '1/4X', '1/8X', '1/16X'])
        ]:
            Speed_Menu = QMenu(speed, self)

            for direction, direction_value in [
                (_("Forward"), MENU_TIME_FORWARD),
                (_("Backward"), MENU_TIME_BACKWARD)
            ]:
                Direction_Menu = QMenu(direction, self)

                for actual_speed in speed_values:
                    # Add menu option
                    Time_Option = Direction_Menu.addAction(_(actual_speed))
                    Time_Option.triggered.connect(
                        partial(self.Time_Triggered, direction_value, clip_ids, actual_speed))

                # Add menu to parent
                Speed_Menu.addMenu(Direction_Menu)
            # Add menu to parent
            Time_Menu.addMenu(Speed_Menu)

        # Add Freeze menu options
        Time_Menu.addSeparator()
        for freeze_type, trigger_type in [
            (_("Freeze"), MENU_TIME_FREEZE),
            (_("Freeze && Zoom"), MENU_TIME_FREEZE_ZOOM)
        ]:
            Freeze_Menu = QMenu(freeze_type, self)

            for freeze_seconds in [2, 4, 6, 8, 10, 20, 30]:
                # Add menu option
                Time_Option = Freeze_Menu.addAction(_('{} seconds').format(freeze_seconds))
                Time_Option.triggered.connect(
                    partial(self.Time_Triggered, trigger_type, clip_ids, freeze_seconds, playhead_position))

            # Add menu to parent
            Time_Menu.addMenu(Freeze_Menu)

        # Add menu to parent
        menu.addMenu(Time_Menu)

        # Volume Menu
        Volume_Menu = QMenu(_("Volume"), self)
        Volume_None = Volume_Menu.addAction(_("Reset Volume"))
        Volume_None.triggered.connect(partial(self.Volume_Triggered, MENU_VOLUME_NONE, clip_ids))
        Volume_Menu.addSeparator()
        for position, position_label in [
            ("Start of Clip", _("Start of Clip")),
            ("End of Clip", _("End of Clip")),
            ("Entire Clip", _("Entire Clip"))
        ]:
            Position_Menu = QMenu(position_label, self)

            if position == "Start of Clip":
                Fade_In_Fast = Position_Menu.addAction(_("Fade In (Fast)"))
                Fade_In_Fast.triggered.connect(partial(
                    self.Volume_Triggered, MENU_VOLUME_FADE_IN_FAST, clip_ids, position))
                Fade_In_Slow = Position_Menu.addAction(_("Fade In (Slow)"))
                Fade_In_Slow.triggered.connect(partial(
                    self.Volume_Triggered, MENU_VOLUME_FADE_IN_SLOW, clip_ids, position))

            elif position == "End of Clip":
                Fade_Out_Fast = Position_Menu.addAction(_("Fade Out (Fast)"))
                Fade_Out_Fast.triggered.connect(partial(
                    self.Volume_Triggered, MENU_VOLUME_FADE_OUT_FAST, clip_ids, position))
                Fade_Out_Slow = Position_Menu.addAction(_("Fade Out (Slow)"))
                Fade_Out_Slow.triggered.connect(partial(
                    self.Volume_Triggered, MENU_VOLUME_FADE_OUT_SLOW, clip_ids, position))

            else:
                Fade_In_Out_Fast = Position_Menu.addAction(_("Fade In and Out (Fast)"))
                Fade_In_Out_Fast.triggered.connect(partial(
                    self.Volume_Triggered, MENU_VOLUME_FADE_IN_OUT_FAST, clip_ids, position))
                Fade_In_Out_Slow = Position_Menu.addAction(_("Fade In and Out (Slow)"))
                Fade_In_Out_Slow.triggered.connect(partial(
                    self.Volume_Triggered, MENU_VOLUME_FADE_IN_OUT_SLOW, clip_ids, position))
                Position_Menu.addSeparator()
                Fade_In_Slow = Position_Menu.addAction(_("Fade In (Entire Clip)"))
                Fade_In_Slow.triggered.connect(partial(
                    self.Volume_Triggered, MENU_VOLUME_FADE_IN_SLOW, clip_ids, position))
                Fade_Out_Slow = Position_Menu.addAction(_("Fade Out (Entire Clip)"))
                Fade_Out_Slow.triggered.connect(partial(
                    self.Volume_Triggered, MENU_VOLUME_FADE_OUT_SLOW, clip_ids, position))

            # Add levels (100% to 0%)
            Position_Menu.addSeparator()
            Volume_100 = Position_Menu.addAction(_("Level 100%"))
            Volume_100.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_100, clip_ids, position))
            Volume_90 = Position_Menu.addAction(_("Level 90%"))
            Volume_90.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_90, clip_ids, position))
            Volume_80 = Position_Menu.addAction(_("Level 80%"))
            Volume_80.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_80, clip_ids, position))
            Volume_70 = Position_Menu.addAction(_("Level 70%"))
            Volume_70.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_70, clip_ids, position))
            Volume_60 = Position_Menu.addAction(_("Level 60%"))
            Volume_60.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_60, clip_ids, position))
            Volume_50 = Position_Menu.addAction(_("Level 50%"))
            Volume_50.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_50, clip_ids, position))
            Volume_40 = Position_Menu.addAction(_("Level 40%"))
            Volume_40.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_40, clip_ids, position))
            Volume_30 = Position_Menu.addAction(_("Level 30%"))
            Volume_30.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_30, clip_ids, position))
            Volume_20 = Position_Menu.addAction(_("Level 20%"))
            Volume_20.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_20, clip_ids, position))
            Volume_10 = Position_Menu.addAction(_("Level 10%"))
            Volume_10.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_10, clip_ids, position))
            Volume_0 = Position_Menu.addAction(_("Level 0%"))
            Volume_0.triggered.connect(partial(
                self.Volume_Triggered, MENU_VOLUME_LEVEL_0, clip_ids, position))

            Volume_Menu.addMenu(Position_Menu)
        menu.addMenu(Volume_Menu)

        # Add separate audio menu
        Split_Audio_Channels_Menu = QMenu(_("Separate Audio"), self)
        Split_Single_Clip = Split_Audio_Channels_Menu.addAction(_("Single Clip (all channels)"))
        Split_Single_Clip.triggered.connect(partial(
            self.Split_Audio_Triggered, MENU_SPLIT_AUDIO_SINGLE, clip_ids))
        Split_Multiple_Clips = Split_Audio_Channels_Menu.addAction(_("Multiple Clips (each channel)"))
        Split_Multiple_Clips.triggered.connect(partial(
            self.Split_Audio_Triggered, MENU_SPLIT_AUDIO_MULTIPLE, clip_ids))
        menu.addMenu(Split_Audio_Channels_Menu)

        # If Playhead overlapping clip
        if clip:
            start_of_clip = float(clip.data["start"])
            end_of_clip = float(clip.data["end"])
            position_of_clip = float(clip.data["position"])
            if (
                playhead_position >= position_of_clip
                and playhead_position <= (position_of_clip + (end_of_clip - start_of_clip))
            ):
                # Add split clip menu
                Slice_Menu = QMenu(_("Slice"), self)
                Slice_Keep_Both = Slice_Menu.addAction(_("Keep Both Sides"))
                Slice_Keep_Both.triggered.connect(partial(
                    self.Slice_Triggered, MENU_SLICE_KEEP_BOTH, [clip_id], [], playhead_position))
                Slice_Keep_Left = Slice_Menu.addAction(_("Keep Left Side"))
                Slice_Keep_Left.triggered.connect(partial(
                    self.Slice_Triggered, MENU_SLICE_KEEP_LEFT, [clip_id], [], playhead_position))
                Slice_Keep_Right = Slice_Menu.addAction(_("Keep Right Side"))
                Slice_Keep_Right.triggered.connect(partial(
                    self.Slice_Triggered, MENU_SLICE_KEEP_RIGHT, [clip_id], [], playhead_position))
                menu.addMenu(Slice_Menu)

        # Transform menu
        Transform_Action = self.window.actionTransform
        Transform_Action.triggered.connect(
            partial(self.Transform_Triggered, MENU_TRANSFORM, clip_ids))
        menu.addAction(Transform_Action)

        # Add clip display menu (waveform or thumbnail)
        menu.addSeparator()
        Waveform_Menu = QMenu(_("Display"), self)
        ShowWaveform = Waveform_Menu.addAction(_("Show Waveform"))
        ShowWaveform.triggered.connect(partial(self.Show_Waveform_Triggered, clip_ids))
        HideWaveform = Waveform_Menu.addAction(_("Show Thumbnail"))
        HideWaveform.triggered.connect(partial(self.Hide_Waveform_Triggered, clip_ids))
        menu.addMenu(Waveform_Menu)

        # Properties
        menu.addAction(self.window.actionProperties)

        # Remove Clip Menu
        menu.addSeparator()
        menu.addAction(self.window.actionRemoveClip)

        # Show Context menu
        return menu.exec_(QCursor.pos())

    def Transform_Triggered(self, action, clip_ids):
        log.debug("Transform_Triggered")

        # Emit signal to transform this clip (for the 1st clip id)
        if clip_ids:
            # Transform first clip in list
            self.window.TransformSignal.emit(clip_ids[0])
        else:
            # Clear transform
            self.window.TransformSignal.emit("")

    def Show_Waveform_Triggered(self, clip_ids):
        """Show a waveform for the selected clip"""

        # Loop through each selected clip
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            file_path = clip.data["reader"]["path"]

            # Find actual clip object from libopenshot
            c = self.window.timeline_sync.timeline.GetClip(clip_id)
            if c and c.Reader() and not c.Reader().info.has_single_image:
                # Find frame 1 channel_filter property
                channel_filter = c.channel_filter.GetInt(1)

                # Set cursor to waiting
                get_app().setOverrideCursor(QCursor(Qt.WaitCursor))

                # Get audio data in a separate thread (so it doesn't block the UI)
                channel_filter = channel_filter
                get_audio_data(clip_id, file_path, channel_filter, c.volume)

    def Hide_Waveform_Triggered(self, clip_ids):
        """Hide the waveform for the selected clip"""

        # Loop through each selected clip
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)

            if clip:
                # Pass to javascript timeline (and render)
                self.run_js(JS_SCOPE_SELECTOR + ".hideAudioData('" + clip_id + "');")

    def Waveform_Ready(self, clip_id, audio_data):
        """Callback when audio waveform is ready"""
        log.info("Waveform_Ready for clip ID: %s" % (clip_id))

        # Convert waveform data to JSON
        serialized_audio_data = json.dumps(audio_data)

        # Set waveform cache (with clip_id as key)
        self.waveform_cache[clip_id] = serialized_audio_data

        # Pass to javascript timeline (and render)
        self.run_js(JS_SCOPE_SELECTOR + ".setAudioData('" + clip_id + "', " + serialized_audio_data + ");")

        # Restore normal cursor
        get_app().restoreOverrideCursor()

        # Start timer to redraw audio
        self.redraw_audio_timer.start()

    def Thumbnail_Updated(self, clip_id):
        """Callback when thumbnail needs to be updated"""
        # Pass to javascript timeline (and render)
        self.run_js(JS_SCOPE_SELECTOR + ".updateThumbnail('" + clip_id + "');")

    def Split_Audio_Triggered(self, action, clip_ids):
        """Callback for split audio context menus"""
        log.debug("Split_Audio_Triggered")

        # Get translation method
        _ = get_app()._tr

        # Loop through each selected clip
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            # Get # of tracks
            all_tracks = get_app().project.get("layers")

            # Clear audio override
            p = openshot.Point(1, -1.0, openshot.CONSTANT)  # Override has_audio keyframe to False
            p_object = json.loads(p.Json())
            clip.data["has_audio"] = {"Points": [p_object]}

            # Remove the ID property from the clip (so it becomes a new one)
            clip.id = None
            clip.type = 'insert'
            clip.data.pop('id')
            clip.key.pop(1)

            # Get title of clip
            clip_title = clip.data["title"]

            if action == MENU_SPLIT_AUDIO_SINGLE:
                # Clear channel filter on new clip
                p = openshot.Point(1, -1.0, openshot.CONSTANT)
                p_object = json.loads(p.Json())
                clip.data["channel_filter"] = {"Points": [p_object]}

                # Filter out video on the new clip
                p = openshot.Point(1, 0.0, openshot.CONSTANT)  # Override has_video keyframe to False
                p_object = json.loads(p.Json())
                clip.data["has_video"] = {"Points": [p_object]}
                # Also set scale to None
                # Workaround for https://github.com/OpenShot/openshot-qt/issues/2882
                clip.data["scale"] = openshot.SCALE_NONE

                # Get track below selected track (if any)
                next_track_number = clip.data['layer']
                found_track = False
                for track in reversed(sorted(all_tracks, key=itemgetter('number'))):
                    if found_track:
                        next_track_number = track.get("number")
                        break
                    if track.get("number") == clip.data['layer']:
                        found_track = True
                        continue

                # Adjust the layer, so this new audio clip doesn't overlap the parent
                clip.data['layer'] = next_track_number  # Add to layer below clip

                # Adjust the clip title
                channel_label = _("(all channels)")
                clip.data["title"] = clip_title + " " + channel_label
                # Save changes
                clip.save()

                # Generate waveform for new clip
                log.info("Generate waveform for split audio track clip id: %s" % clip.id)
                self.Show_Waveform_Triggered([clip.id])

            if action == MENU_SPLIT_AUDIO_MULTIPLE:
                # Get # of channels on clip
                channels = int(clip.data["reader"]["channels"])

                # Loop through each channel
                for channel in range(0, channels):
                    log.debug("Adding clip for channel %s" % channel)

                    # Each clip is filtered to a different channel
                    p = openshot.Point(1, channel, openshot.CONSTANT)
                    p_object = json.loads(p.Json())
                    clip.data["channel_filter"] = {"Points": [p_object]}

                    # Filter out video on the new clip
                    p = openshot.Point(1, 0.0, openshot.CONSTANT)  # Override has_video keyframe to False
                    p_object = json.loads(p.Json())
                    clip.data["has_video"] = {"Points": [p_object]}
                    # Also set scale to None
                    # Workaround for https://github.com/OpenShot/openshot-qt/issues/2882
                    clip.data["scale"] = openshot.SCALE_NONE

                    # Get track below selected track (if any)
                    next_track_number = clip.data['layer']
                    found_track = False
                    for track in reversed(sorted(all_tracks, key=itemgetter('number'))):
                        if found_track:
                            next_track_number = track.get("number")
                            break
                        if track.get("number") == clip.data['layer']:
                            found_track = True
                            continue

                    # Adjust the layer, so this new audio clip doesn't overlap the parent
                    clip.data['layer'] = max(next_track_number, 0)  # Add to layer below clip

                    # Adjust the clip title
                    channel_label = _("(channel %s)") % (channel + 1)
                    clip.data["title"] = clip_title + " " + channel_label

                    # Save changes
                    clip.save()

                    # Generate waveform for new clip
                    log.info("Generate waveform for split audio track clip id: %s" % clip.id)
                    self.Show_Waveform_Triggered([clip.id])

                    # Remove the ID property from the clip (so next time, it will create a new clip)
                    clip.id = None
                    clip.type = 'insert'
                    clip.data.pop('id')

        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            # Filter out audio on the original clip
            p = openshot.Point(1, 0.0, openshot.CONSTANT)  # Override has_audio keyframe to False
            p_object = json.loads(p.Json())
            clip.data["has_audio"] = {"Points": [p_object]}

            # Save filter on original clip
            self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)
            clip.save()

    def Layout_Triggered(self, action, clip_ids):
        """Callback for the layout context menus"""
        log.debug(action)

        # Loop through each selected clip
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            new_gravity = openshot.GRAVITY_CENTER
            if action == MENU_LAYOUT_CENTER:
                new_gravity = openshot.GRAVITY_CENTER
            if action == MENU_LAYOUT_TOP_LEFT:
                new_gravity = openshot.GRAVITY_TOP_LEFT
            elif action == MENU_LAYOUT_TOP_RIGHT:
                new_gravity = openshot.GRAVITY_TOP_RIGHT
            elif action == MENU_LAYOUT_BOTTOM_LEFT:
                new_gravity = openshot.GRAVITY_BOTTOM_LEFT
            elif action == MENU_LAYOUT_BOTTOM_RIGHT:
                new_gravity = openshot.GRAVITY_BOTTOM_RIGHT

            if action == MENU_LAYOUT_NONE:
                # Reset scale mode
                clip.data["scale"] = openshot.SCALE_FIT
                clip.data["gravity"] = openshot.GRAVITY_CENTER

                # Clear scale keyframes
                p = openshot.Point(1, 1.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data["scale_x"] = {"Points": [p_object]}
                clip.data["scale_y"] = {"Points": [p_object]}

                # Clear location keyframes
                p = openshot.Point(1, 0.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data["location_x"] = {"Points": [p_object]}
                clip.data["location_y"] = {"Points": [p_object]}

            if action in [MENU_LAYOUT_CENTER,
                          MENU_LAYOUT_TOP_LEFT,
                          MENU_LAYOUT_TOP_RIGHT,
                          MENU_LAYOUT_BOTTOM_LEFT,
                          MENU_LAYOUT_BOTTOM_RIGHT]:
                # Reset scale mode
                clip.data["scale"] = openshot.SCALE_FIT
                clip.data["gravity"] = new_gravity

                # Add scale keyframes
                p = openshot.Point(1, 0.5, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data["scale_x"] = {"Points": [p_object]}
                clip.data["scale_y"] = {"Points": [p_object]}

                # Add location keyframes
                p = openshot.Point(1, 0.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data["location_x"] = {"Points": [p_object]}
                clip.data["location_y"] = {"Points": [p_object]}

            if action == MENU_LAYOUT_ALL_WITH_ASPECT:
                # Update all intersecting clips
                self.show_all_clips(clip, False)

            elif action == MENU_LAYOUT_ALL_WITHOUT_ASPECT:
                # Update all intersecting clips
                self.show_all_clips(clip, True)

            else:
                # Save changes
                self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

    def Animate_Triggered(self, action, clip_ids, position="Entire Clip"):
        """Callback for the animate context menus"""
        log.debug(action)

        # Loop through each selected clip
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            # Get framerate
            fps = get_app().project.get("fps")
            fps_float = float(fps["num"]) / float(fps["den"])

            # Get existing clip object
            start_of_clip = round(float(clip.data["start"]) * fps_float) + 1
            end_of_clip = round(float(clip.data["end"]) * fps_float) + 1

            # Determine the beginning and ending of this animation
            # ["Start of Clip", "End of Clip", "Entire Clip"]
            start_animation = start_of_clip
            end_animation = end_of_clip
            if position == "Start of Clip":
                start_animation = start_of_clip
                end_animation = min(start_of_clip + (1.0 * fps_float), end_of_clip)
            elif position == "End of Clip":
                start_animation = max(1.0, end_of_clip - (1.0 * fps_float))
                end_animation = end_of_clip

            if action == MENU_ANIMATE_NONE:
                # Clear all keyframes
                default_zoom = openshot.Point(start_animation, 1.0, openshot.BEZIER)
                default_zoom_object = json.loads(default_zoom.Json())
                default_loc = openshot.Point(start_animation, 0.0, openshot.BEZIER)
                default_loc_object = json.loads(default_loc.Json())
                clip.data["gravity"] = openshot.GRAVITY_CENTER
                clip.data["scale_x"] = {"Points": [default_zoom_object]}
                clip.data["scale_y"] = {"Points": [default_zoom_object]}
                clip.data["location_x"] = {"Points": [default_loc_object]}
                clip.data["location_y"] = {"Points": [default_loc_object]}

            if action in [
                MENU_ANIMATE_IN_50_100,
                MENU_ANIMATE_IN_75_100,
                MENU_ANIMATE_IN_100_150,
                MENU_ANIMATE_OUT_100_75,
                MENU_ANIMATE_OUT_100_50,
                MENU_ANIMATE_OUT_150_100
            ]:
                # Scale animation
                start_scale = 1.0
                end_scale = 1.0
                if action == MENU_ANIMATE_IN_50_100:
                    start_scale = 0.5
                elif action == MENU_ANIMATE_IN_75_100:
                    start_scale = 0.75
                elif action == MENU_ANIMATE_IN_100_150:
                    end_scale = 1.5
                elif action == MENU_ANIMATE_OUT_100_75:
                    end_scale = 0.75
                elif action == MENU_ANIMATE_OUT_100_50:
                    end_scale = 0.5
                elif action == MENU_ANIMATE_OUT_150_100:
                    start_scale = 1.5

                # Add keyframes
                start = openshot.Point(start_animation, start_scale, openshot.BEZIER)
                start_object = json.loads(start.Json())
                end = openshot.Point(end_animation, end_scale, openshot.BEZIER)
                end_object = json.loads(end.Json())
                clip.data["gravity"] = openshot.GRAVITY_CENTER
                self.AddPoint(clip.data["scale_x"], start_object)
                self.AddPoint(clip.data["scale_x"], end_object)
                self.AddPoint(clip.data["scale_y"], start_object)
                self.AddPoint(clip.data["scale_y"], end_object)

            if action in [
                MENU_ANIMATE_CENTER_TOP,
                MENU_ANIMATE_CENTER_LEFT,
                MENU_ANIMATE_CENTER_RIGHT,
                MENU_ANIMATE_CENTER_BOTTOM,
                MENU_ANIMATE_TOP_CENTER,
                MENU_ANIMATE_LEFT_CENTER,
                MENU_ANIMATE_RIGHT_CENTER,
                MENU_ANIMATE_BOTTOM_CENTER,
                MENU_ANIMATE_TOP_BOTTOM,
                MENU_ANIMATE_LEFT_RIGHT,
                MENU_ANIMATE_RIGHT_LEFT,
                MENU_ANIMATE_BOTTOM_TOP
            ]:
                # Location animation
                animate_start_x = 0.0
                animate_end_x = 0.0
                animate_start_y = 0.0
                animate_end_y = 0.0
                # Center to edge...
                if action == MENU_ANIMATE_CENTER_TOP:
                    animate_end_y = -1.0
                elif action == MENU_ANIMATE_CENTER_LEFT:
                    animate_end_x = -1.0
                elif action == MENU_ANIMATE_CENTER_RIGHT:
                    animate_end_x = 1.0
                elif action == MENU_ANIMATE_CENTER_BOTTOM:
                    animate_end_y = 1.0

                # Edge to Center
                elif action == MENU_ANIMATE_TOP_CENTER:
                    animate_start_y = -1.0
                elif action == MENU_ANIMATE_LEFT_CENTER:
                    animate_start_x = -1.0
                elif action == MENU_ANIMATE_RIGHT_CENTER:
                    animate_start_x = 1.0
                elif action == MENU_ANIMATE_BOTTOM_CENTER:
                    animate_start_y = 1.0

                # Edge to Edge
                elif action == MENU_ANIMATE_TOP_BOTTOM:
                    animate_start_y = -1.0
                    animate_end_y = 1.0
                elif action == MENU_ANIMATE_LEFT_RIGHT:
                    animate_start_x = -1.0
                    animate_end_x = 1.0
                elif action == MENU_ANIMATE_RIGHT_LEFT:
                    animate_start_x = 1.0
                    animate_end_x = -1.0
                elif action == MENU_ANIMATE_BOTTOM_TOP:
                    animate_start_y = 1.0
                    animate_end_y = -1.0

                # Add keyframes
                start_x = openshot.Point(start_animation, animate_start_x, openshot.BEZIER)
                start_x_object = json.loads(start_x.Json())
                end_x = openshot.Point(end_animation, animate_end_x, openshot.BEZIER)
                end_x_object = json.loads(end_x.Json())
                start_y = openshot.Point(start_animation, animate_start_y, openshot.BEZIER)
                start_y_object = json.loads(start_y.Json())
                end_y = openshot.Point(end_animation, animate_end_y, openshot.BEZIER)
                end_y_object = json.loads(end_y.Json())
                clip.data["gravity"] = openshot.GRAVITY_CENTER
                self.AddPoint(clip.data["location_x"], start_x_object)
                self.AddPoint(clip.data["location_x"], end_x_object)
                self.AddPoint(clip.data["location_y"], start_y_object)
                self.AddPoint(clip.data["location_y"], end_y_object)

            if action == MENU_ANIMATE_RANDOM:
                # Location animation
                animate_start_x = uniform(-0.5, 0.5)
                animate_end_x = uniform(-0.15, 0.15)
                animate_start_y = uniform(-0.5, 0.5)
                animate_end_y = uniform(-0.15, 0.15)

                # Scale animation
                start_scale = uniform(0.5, 1.5)
                end_scale = uniform(0.85, 1.15)

                # Add keyframes
                start = openshot.Point(start_animation, start_scale, openshot.BEZIER)
                start_object = json.loads(start.Json())
                end = openshot.Point(end_animation, end_scale, openshot.BEZIER)
                end_object = json.loads(end.Json())
                clip.data["gravity"] = openshot.GRAVITY_CENTER
                self.AddPoint(clip.data["scale_x"], start_object)
                self.AddPoint(clip.data["scale_x"], end_object)
                self.AddPoint(clip.data["scale_y"], start_object)
                self.AddPoint(clip.data["scale_y"], end_object)

                # Add keyframes
                start_x = openshot.Point(start_animation, animate_start_x, openshot.BEZIER)
                start_x_object = json.loads(start_x.Json())
                end_x = openshot.Point(end_animation, animate_end_x, openshot.BEZIER)
                end_x_object = json.loads(end_x.Json())
                start_y = openshot.Point(start_animation, animate_start_y, openshot.BEZIER)
                start_y_object = json.loads(start_y.Json())
                end_y = openshot.Point(end_animation, animate_end_y, openshot.BEZIER)
                end_y_object = json.loads(end_y.Json())
                clip.data["gravity"] = openshot.GRAVITY_CENTER
                self.AddPoint(clip.data["location_x"], start_x_object)
                self.AddPoint(clip.data["location_x"], end_x_object)
                self.AddPoint(clip.data["location_y"], start_y_object)
                self.AddPoint(clip.data["location_y"], end_y_object)

            # Save changes
            self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

    def AddPoint(self, keyframe, new_point):
        """Add a Point to a Keyframe dict. Always remove existing points,
        if any collisions are found"""
        # Get all points that don't match new point coordinate
        cleaned_points = [
            point
            for point in keyframe["Points"]
            if point.get("co", {}).get("X") != new_point.get("co", {}).get("X")
        ]
        cleaned_points.append(new_point)

        # Replace points with new list
        keyframe["Points"] = cleaned_points


    def Copy_Triggered(self, action, clip_ids, tran_ids):
        """Callback for copy context menus"""
        log.debug(action)

        # Empty previous clipboard
        self.copy_clipboard = {}
        self.copy_transition_clipboard = {}

        # Loop through clip objects
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            self.copy_clipboard[clip_id] = {}

            if action in [MENU_COPY_CLIP, MENU_COPY_ALL]:
                self.copy_clipboard[clip_id] = clip.data
            elif action == MENU_COPY_KEYFRAMES_ALL:
                self.copy_clipboard[clip_id]['alpha'] = clip.data['alpha']
                self.copy_clipboard[clip_id]['crop_height'] = clip.data['crop_height']
                self.copy_clipboard[clip_id]['crop_width'] = clip.data['crop_width']
                self.copy_clipboard[clip_id]['crop_x'] = clip.data['crop_x']
                self.copy_clipboard[clip_id]['crop_y'] = clip.data['crop_y']
                self.copy_clipboard[clip_id]['gravity'] = clip.data['gravity']
                self.copy_clipboard[clip_id]['scale_x'] = clip.data['scale_x']
                self.copy_clipboard[clip_id]['scale_y'] = clip.data['scale_y']
                self.copy_clipboard[clip_id]['rotation'] = clip.data['rotation']
                self.copy_clipboard[clip_id]['location_x'] = clip.data['location_x']
                self.copy_clipboard[clip_id]['location_y'] = clip.data['location_y']
                self.copy_clipboard[clip_id]['time'] = clip.data['time']
                self.copy_clipboard[clip_id]['volume'] = clip.data['volume']
            elif action == MENU_COPY_KEYFRAMES_ALPHA:
                self.copy_clipboard[clip_id]['alpha'] = clip.data['alpha']
            elif action == MENU_COPY_KEYFRAMES_SCALE:
                self.copy_clipboard[clip_id]['gravity'] = clip.data['gravity']
                self.copy_clipboard[clip_id]['scale_x'] = clip.data['scale_x']
                self.copy_clipboard[clip_id]['scale_y'] = clip.data['scale_y']
            elif action == MENU_COPY_KEYFRAMES_ROTATE:
                self.copy_clipboard[clip_id]['gravity'] = clip.data['gravity']
                self.copy_clipboard[clip_id]['rotation'] = clip.data['rotation']
            elif action == MENU_COPY_KEYFRAMES_LOCATION:
                self.copy_clipboard[clip_id]['gravity'] = clip.data['gravity']
                self.copy_clipboard[clip_id]['location_x'] = clip.data['location_x']
                self.copy_clipboard[clip_id]['location_y'] = clip.data['location_y']
            elif action == MENU_COPY_KEYFRAMES_TIME:
                self.copy_clipboard[clip_id]['time'] = clip.data['time']
            elif action == MENU_COPY_KEYFRAMES_VOLUME:
                self.copy_clipboard[clip_id]['volume'] = clip.data['volume']
            elif action == MENU_COPY_EFFECTS:
                self.copy_clipboard[clip_id]['effects'] = clip.data['effects']

        # Loop through transition objects
        for tran_id in tran_ids:

            # Get existing transition object
            tran = Transition.get(id=tran_id)
            if not tran:
                # Invalid transition, skip to next item
                continue

            self.copy_transition_clipboard[tran_id] = {}

            if action in [MENU_COPY_TRANSITION, MENU_COPY_ALL]:
                self.copy_transition_clipboard[tran_id] = tran.data
            elif action == MENU_COPY_KEYFRAMES_ALL:
                self.copy_transition_clipboard[tran_id]['brightness'] = tran.data['brightness']
                self.copy_transition_clipboard[tran_id]['contrast'] = tran.data['contrast']
            elif action == MENU_COPY_KEYFRAMES_BRIGHTNESS:
                self.copy_transition_clipboard[tran_id]['brightness'] = tran.data['brightness']
            elif action == MENU_COPY_KEYFRAMES_CONTRAST:
                self.copy_transition_clipboard[tran_id]['contrast'] = tran.data['contrast']

    def Paste_Triggered(self, action, position, layer_id, clip_ids, tran_ids):
        """Callback for paste context menus"""
        log.debug(action)

        # Get list of clipboard items (that are complete clips or transitions)
        # i.e. ignore partial clipboard items (keyframes / effects / etc...)
        clipboard_clip_ids = [k for k, v in self.copy_clipboard.items() if v.get('id')]
        clipboard_tran_ids = [k for k, v in self.copy_transition_clipboard.items() if v.get('id')]

        # Determine left most copied clip, and top most track (the top left point of the copied objects)
        if len(clipboard_clip_ids) + len(clipboard_tran_ids):
            left_most_position = -1.0
            top_most_layer = -1
            # Loop through each copied clip (looking for top left point)
            for clip_id in clipboard_clip_ids:
                # Get existing clip object
                clip = Clip()
                clip.data = self.copy_clipboard.get(clip_id, {})
                if clip.data['position'] < left_most_position or left_most_position == -1.0:
                    left_most_position = clip.data['position']
                if clip.data['layer'] > top_most_layer or top_most_layer == -1.0:
                    top_most_layer = clip.data['layer']
            # Loop through each copied transition (looking for top left point)
            for tran_id in clipboard_tran_ids:
                # Get existing transition object
                tran = Transition()
                tran.data = self.copy_transition_clipboard.get(tran_id, {})
                if tran.data['position'] < left_most_position or left_most_position == -1.0:
                    left_most_position = tran.data['position']
                if tran.data['layer'] > top_most_layer or top_most_layer == -1.0:
                    top_most_layer = tran.data['layer']

            # Default layer if not known
            if layer_id == -1:
                layer_id = top_most_layer

            # Determine difference from top left and paste location
            position_diff = position - left_most_position
            layer_diff = layer_id - top_most_layer

            # Loop through each copied clip
            for clip_id in clipboard_clip_ids:
                # Get existing clip object
                clip = Clip()
                clip.data = self.copy_clipboard.get(clip_id, {})

                # Remove the ID property from the clip (so it becomes a new one)
                clip.type = 'insert'
                clip.data.pop('id')

                # Adjust the position and track
                clip.data['position'] += position_diff
                clip.data['layer'] += layer_diff

                # Save changes
                clip.save()

            # Loop through all copied transitions
            for tran_id in clipboard_tran_ids:
                # Get existing transition object
                tran = Transition()
                tran.data = self.copy_transition_clipboard.get(tran_id, {})

                # Remove the ID property from the transition (so it becomes a new one)
                tran.type = 'insert'
                tran.data.pop('id')

                # Adjust the position and track
                tran.data['position'] += position_diff
                tran.data['layer'] += layer_diff

                # Save changes
                tran.save()

        # Loop through each full clip object copied
        if self.copy_clipboard:
            for clip_id in clip_ids:

                # Get existing clip object
                clip = Clip.get(id=clip_id)
                if not clip:
                    # Invalid clip, skip to next item
                    continue

                # Apply clipboard to clip (there should only be a single key in this dict)
                for k, v in self.copy_clipboard[list(self.copy_clipboard)[0]].items():
                    if k != 'id':
                        # Overwrite clips properties (which are in the clipboard)
                        clip.data[k] = v

                # Save changes
                clip.save()

        # Loop through each full transition object copied
        if self.copy_transition_clipboard:
            for tran_id in tran_ids:

                # Get existing transition object
                tran = Transition.get(id=tran_id)
                if not tran:
                    # Invalid transition, skip to next item
                    continue

                # Apply clipboard to transition (there should only be a single key in this dict)
                for k, v in self.copy_transition_clipboard[list(self.copy_transition_clipboard)[0]].items():
                    if k != 'id':
                        # Overwrite transition properties (which are in the clipboard)
                        tran.data[k] = v

                # Save changes
                tran.save()

    def Nudge_Triggered(self, action, clip_ids, tran_ids):
        """Callback for clip nudges"""
        log.debug("Nudging clip(s) and/or transition(s)")
        left_edge = -1.0
        right_edge = -1.0

        # Determine how far we're going to nudge (1/2 frame or 0.01s, whichever is larger)
        fps = get_app().project.get("fps")
        fps_float = float(fps["num"]) / float(fps["den"])
        nudgeDistance = float(action) / float(fps_float)
        nudgeDistance /= 2.0  # 1/2 frame
        if abs(nudgeDistance) < 0.01:
            nudgeDistance = 0.01 * action  # nudge is less than the minimum of +/- 0.01s
        log.debug("Nudging by %s sec" % nudgeDistance)

        # Loop through each selected clip (find furthest left and right edge)
        for clip_id in clip_ids:
            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            position = float(clip.data["position"])
            start_of_clip = float(clip.data["start"])
            end_of_clip = float(clip.data["end"])

            if position < left_edge or left_edge == -1.0:
                left_edge = position
            if position + (end_of_clip - start_of_clip) > right_edge or right_edge == -1.0:
                right_edge = position + (end_of_clip - start_of_clip)

            # Do not nudge beyond the start of the timeline
            if left_edge + nudgeDistance < 0.0:
                log.info("Cannot nudge beyond start of timeline")
                nudgeDistance = 0

        # Loop through each selected transition (find furthest left and right edge)
        for tran_id in tran_ids:
            # Get existing transition object
            tran = Transition.get(id=tran_id)
            if not tran:
                # Invalid transition, skip to next item
                continue

            position = float(tran.data["position"])
            start_of_tran = float(tran.data["start"])
            end_of_tran = float(tran.data["end"])

            if position < left_edge or left_edge == -1.0:
                left_edge = position
            if position + (end_of_tran - start_of_tran) > right_edge or right_edge == -1.0:
                right_edge = position + (end_of_tran - start_of_tran)

            # Do not nudge beyond the start of the timeline
            if left_edge + nudgeDistance < 0.0:
                log.info("Cannot nudge beyond start of timeline")
                nudgeDistance = 0

        # Loop through each selected clip (update position to align clips)
        for clip_id in clip_ids:
            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            # Do the nudge
            clip.data['position'] += nudgeDistance

            # Save changes
            self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

        # Loop through each selected transition (update position to align clips)
        for tran_id in tran_ids:
            # Get existing transition object
            tran = Transition.get(id=tran_id)
            if not tran:
                # Invalid transition, skip to next item
                continue

            # Do the nudge
            tran.data['position'] += nudgeDistance

            # Save changes
            self.update_transition_data(tran.data, only_basic_props=False)

    def Align_Triggered(self, action, clip_ids, tran_ids):
        """Callback for alignment context menus"""
        log.debug(action)

        left_edge = -1.0
        right_edge = -1.0

        # Loop through each selected clip (find furthest left and right edge)
        for clip_id in clip_ids:
            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            position = float(clip.data["position"])
            start_of_clip = float(clip.data["start"])
            end_of_clip = float(clip.data["end"])

            if position < left_edge or left_edge == -1.0:
                left_edge = position
            if position + (end_of_clip - start_of_clip) > right_edge or right_edge == -1.0:
                right_edge = position + (end_of_clip - start_of_clip)

        # Loop through each selected transition (find furthest left and right edge)
        for tran_id in tran_ids:
            # Get existing transition object
            tran = Transition.get(id=tran_id)
            if not tran:
                # Invalid transition, skip to next item
                continue

            position = float(tran.data["position"])
            start_of_tran = float(tran.data["start"])
            end_of_tran = float(tran.data["end"])

            if position < left_edge or left_edge == -1.0:
                left_edge = position
            if position + (end_of_tran - start_of_tran) > right_edge or right_edge == -1.0:
                right_edge = position + (end_of_tran - start_of_tran)

        # Loop through each selected clip (update position to align clips)
        for clip_id in clip_ids:
            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            if action == MENU_ALIGN_LEFT:
                clip.data['position'] = left_edge
            elif action == MENU_ALIGN_RIGHT:
                position = float(clip.data["position"])
                start_of_clip = float(clip.data["start"])
                end_of_clip = float(clip.data["end"])
                right_clip_edge = position + (end_of_clip - start_of_clip)

                clip.data['position'] = position + (right_edge - right_clip_edge)

            # Save changes
            self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

        # Loop through each selected transition (update position to align clips)
        for tran_id in tran_ids:
            # Get existing transition object
            tran = Transition.get(id=tran_id)
            if not tran:
                # Invalid transition, skip to next item
                continue

            if action == MENU_ALIGN_LEFT:
                tran.data['position'] = left_edge
            elif action == MENU_ALIGN_RIGHT:
                position = float(tran.data["position"])
                start_of_tran = float(tran.data["start"])
                end_of_tran = float(tran.data["end"])
                right_tran_edge = position + (end_of_tran - start_of_tran)

                tran.data['position'] = position + (right_edge - right_tran_edge)

            # Save changes
            self.update_transition_data(tran.data, only_basic_props=False)

    def Fade_Triggered(self, action, clip_ids, position="Entire Clip"):
        """Callback for fade context menus"""
        log.debug(action)

        # Get FPS from project
        fps = get_app().project.get("fps")
        fps_float = float(fps["num"]) / float(fps["den"])

        # Loop through each selected clip
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            start_of_clip = round(float(clip.data["start"]) * fps_float) + 1
            end_of_clip = round(float(clip.data["end"]) * fps_float) + 1

            # Determine the beginning and ending of this animation
            # ["Start of Clip", "End of Clip", "Entire Clip"]
            start_animation = start_of_clip
            end_animation = end_of_clip
            if position == "Start of Clip" and action in [MENU_FADE_IN_FAST, MENU_FADE_OUT_FAST]:
                start_animation = start_of_clip
                end_animation = min(start_of_clip + (1.0 * fps_float), end_of_clip)
            elif position == "Start of Clip" and action in [MENU_FADE_IN_SLOW, MENU_FADE_OUT_SLOW]:
                start_animation = start_of_clip
                end_animation = min(start_of_clip + (3.0 * fps_float), end_of_clip)
            elif position == "End of Clip" and action in [MENU_FADE_IN_FAST, MENU_FADE_OUT_FAST]:
                start_animation = max(1.0, end_of_clip - (1.0 * fps_float))
                end_animation = end_of_clip
            elif position == "End of Clip" and action in [MENU_FADE_IN_SLOW, MENU_FADE_OUT_SLOW]:
                start_animation = max(1.0, end_of_clip - (3.0 * fps_float))
                end_animation = end_of_clip

            # Fade in and out (special case)
            if position == "Entire Clip" and action == MENU_FADE_IN_OUT_FAST:
                # Call this method for the start and end of the clip
                self.Fade_Triggered(MENU_FADE_IN_FAST, clip_ids, "Start of Clip")
                self.Fade_Triggered(MENU_FADE_OUT_FAST, clip_ids, "End of Clip")
                return
            elif position == "Entire Clip" and action == MENU_FADE_IN_OUT_SLOW:
                # Call this method for the start and end of the clip
                self.Fade_Triggered(MENU_FADE_IN_SLOW, clip_ids, "Start of Clip")
                self.Fade_Triggered(MENU_FADE_OUT_SLOW, clip_ids, "End of Clip")
                return

            if action == MENU_FADE_NONE:
                # Clear all keyframes
                p = openshot.Point(1, 1.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data['alpha'] = {"Points": [p_object]}

            if action in [MENU_FADE_IN_FAST, MENU_FADE_IN_SLOW]:
                # Add keyframes
                start = openshot.Point(start_animation, 0.0, openshot.BEZIER)
                start_object = json.loads(start.Json())
                end = openshot.Point(end_animation, 1.0, openshot.BEZIER)
                end_object = json.loads(end.Json())
                self.AddPoint(clip.data['alpha'], start_object)
                self.AddPoint(clip.data['alpha'], end_object)

            if action in [MENU_FADE_OUT_FAST, MENU_FADE_OUT_SLOW]:
                # Add keyframes
                start = openshot.Point(start_animation, 1.0, openshot.BEZIER)
                start_object = json.loads(start.Json())
                end = openshot.Point(end_animation, 0.0, openshot.BEZIER)
                end_object = json.loads(end.Json())
                self.AddPoint(clip.data['alpha'], start_object)
                self.AddPoint(clip.data['alpha'], end_object)

            # Save changes
            self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

    @pyqtSlot(str, str, float)
    def RazorSliceAtCursor(self, clip_id, trans_id, cursor_position):
        """Callback from javascript that the razor tool was clicked"""

        # Determine slice mode (keep both [default], keep left [shift], keep right [ctrl]
        slice_mode = MENU_SLICE_KEEP_BOTH
        if int(QCoreApplication.instance().keyboardModifiers() & Qt.ControlModifier) > 0:
            slice_mode = MENU_SLICE_KEEP_RIGHT
        elif int(QCoreApplication.instance().keyboardModifiers() & Qt.ShiftModifier) > 0:
            slice_mode = MENU_SLICE_KEEP_LEFT

        if clip_id:
            # Slice clip
            QTimer.singleShot(0, partial(self.Slice_Triggered, slice_mode, [clip_id], [], cursor_position))
        elif trans_id:
            # Slice transitions
            QTimer.singleShot(0, partial(self.Slice_Triggered, slice_mode, [], [trans_id], cursor_position))

    def Slice_Triggered(self, action, clip_ids, trans_ids, playhead_position=0):
        """Callback for slice context menus"""
        # Get FPS from project
        fps = get_app().project.get("fps")
        fps_num = float(fps["num"])
        fps_den = float(fps["den"])
        fps_float = fps_num / fps_den
        frame_duration = fps_den / fps_num

        # Get the nearest starting frame position to the playhead (this helps to prevent cutting
        # in-between frames, and thus less likely to repeat or skip a frame).
        playhead_position = float(round((playhead_position * fps_num) / fps_den) * fps_den) / fps_num

        # Loop through each clip (using the list of ids)
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            # Determine if waveform needs to be redrawn
            has_audio_data = clip_id in self.waveform_cache

            if action in [MENU_SLICE_KEEP_LEFT, MENU_SLICE_KEEP_BOTH]:
                # Get details of original clip
                position_of_clip = float(clip.data["position"])
                start_of_clip = float(clip.data["start"])

                # Set new 'end' of clip
                clip.data["end"] = start_of_clip + (playhead_position - position_of_clip)

            elif action == MENU_SLICE_KEEP_RIGHT:
                # Get details of original clip
                position_of_clip = float(clip.data["position"])
                start_of_clip = float(clip.data["start"])

                # Set new 'end' of clip
                clip.data["position"] = playhead_position
                clip.data["start"] = start_of_clip + (playhead_position - position_of_clip)

            if action == MENU_SLICE_KEEP_BOTH:
                # Add the 2nd clip (the right side, since the left side has already been adjusted above)
                # Get right side clip object
                right_clip = Clip.get(id=clip_id)
                if not right_clip:
                    # Invalid clip, skip to next item
                    continue

                # Remove the ID property from the clip (so it becomes a new one)
                right_clip.id = None
                right_clip.type = 'insert'
                right_clip.data.pop('id')
                right_clip.key.pop(1)

                # Generate new ID to effects on the right (so they become new ones)
                for clip_propertie_name, propertie_value in right_clip.data.items() :
                    if clip_propertie_name == "effects":
                        for item in propertie_value:
                            item['id'] = get_app().project.generate_id()

                # Set new 'start' of right_clip (need to bump 1 frame duration more, so we don't repeat a frame)
                right_clip.data["position"] = (round(float(playhead_position) * fps_float) + 1) / fps_float
                right_clip.data["start"] = (round(float(clip.data["end"]) * fps_float) + 2) / fps_float

                # Save changes
                right_clip.save()

                # Save changes again (with new thumbnail)
                self.update_clip_data(right_clip.data, only_basic_props=False, ignore_reader=True)

                if has_audio_data:
                    # Add right clip audio to cache
                    self.waveform_cache[right_clip.id] = self.waveform_cache.get(clip_id, '[]')

                    # Pass audio to javascript timeline (and render)
                    self.run_js(JS_SCOPE_SELECTOR + ".setAudioData('{}',{});"
                        .format(right_clip.id, self.waveform_cache.get(right_clip.id)))

            # Save changes
            self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

        # Start or restart timer to redraw audio waveforms
        self.redraw_audio_timer.start()

        # Loop through each transition (using the list of ids)
        for trans_id in trans_ids:
            # Get existing transition object
            trans = Transition.get(id=trans_id)
            if not trans:
                # Invalid transition, skip to next item
                continue

            if action in [MENU_SLICE_KEEP_LEFT, MENU_SLICE_KEEP_BOTH]:
                # Get details of original transition
                position_of_tran = float(trans.data["position"])

                # Set new 'end' of transition
                trans.data["end"] = playhead_position - position_of_tran

            elif action == MENU_SLICE_KEEP_RIGHT:
                # Get details of transition clip
                position_of_tran = float(trans.data["position"])
                end_of_tran = float(trans.data["end"])

                # Set new 'end' of transition
                trans.data["position"] = playhead_position
                trans.data["end"] = end_of_tran - (playhead_position - position_of_tran)

            if action == MENU_SLICE_KEEP_BOTH:
                # Add the 2nd transition (the right side, since the left side has already been adjusted above)
                # Get right side transition object
                right_tran = Transition.get(id=trans_id)
                if not right_tran:
                    # Invalid transition, skip to next item
                    continue

                # Remove the ID property from the transition (so it becomes a new one)
                right_tran.id = None
                right_tran.type = 'insert'
                right_tran.data.pop('id')
                right_tran.key.pop(1)

                # Get details of original transition
                position_of_tran = float(right_tran.data["position"])
                end_of_tran = float(right_tran.data["end"])

                # Set new 'end' of right_tran
                right_tran.data["position"] = playhead_position + frame_duration
                right_tran.data["end"] = end_of_tran - (playhead_position - position_of_tran) + frame_duration

                # Save changes
                right_tran.save()

                # Save changes again (right side)
                self.update_transition_data(right_tran.data, only_basic_props=False)

            # Save changes (left side)
            self.update_transition_data(trans.data, only_basic_props=False)

    def Volume_Triggered(self, action, clip_ids, position="Entire Clip"):
        """Callback for volume context menus"""
        log.debug(action)

        # Callback function, to redraw audio data after an update
        def callback(self, clip_id, callback_data):
            has_audio_data = callback_data
            log.info('has_audio_data: %s', has_audio_data)

            if has_audio_data:
                # Re-generate waveform since volume curve has changed
                self.Show_Waveform_Triggered(clip_id)

        # Get FPS from project
        fps = get_app().project.get("fps")
        fps_float = float(fps["num"]) / float(fps["den"])

        # Loop through each selected clip
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            start_of_clip = round(float(clip.data["start"]) * fps_float) + 1
            end_of_clip = round(float(clip.data["end"]) * fps_float) + 1

            # Determine the beginning and ending of this animation
            # ["Start of Clip", "End of Clip", "Entire Clip"]
            start_animation = start_of_clip
            end_animation = end_of_clip
            if position == "Start of Clip" and action in [
                MENU_VOLUME_FADE_IN_FAST,
                MENU_VOLUME_FADE_OUT_FAST
            ]:
                start_animation = start_of_clip
                end_animation = min(start_of_clip + (1.0 * fps_float), end_of_clip)

            elif position == "Start of Clip" and action in [
                MENU_VOLUME_FADE_IN_SLOW,
                MENU_VOLUME_FADE_OUT_SLOW
            ]:
                start_animation = start_of_clip
                end_animation = min(start_of_clip + (3.0 * fps_float), end_of_clip)

            elif position == "End of Clip" and action in [
                MENU_VOLUME_FADE_IN_FAST,
                MENU_VOLUME_FADE_OUT_FAST
            ]:
                start_animation = max(1.0, end_of_clip - (1.0 * fps_float))
                end_animation = end_of_clip

            elif position == "End of Clip" and action in [
                MENU_VOLUME_FADE_IN_SLOW,
                MENU_VOLUME_FADE_OUT_SLOW
            ]:
                start_animation = max(1.0, end_of_clip - (3.0 * fps_float))
                end_animation = end_of_clip

            elif position == "Start of Clip":
                # Only used when setting levels (a single keyframe)
                start_animation = start_of_clip
                end_animation = start_of_clip

            elif position == "End of Clip":
                # Only used when setting levels (a single keyframe)
                start_animation = end_of_clip
                end_animation = end_of_clip

            # Fade in and out (special case)
            if position == "Entire Clip" and action == MENU_VOLUME_FADE_IN_OUT_FAST:
                # Call this method for the start and end of the clip
                self.Volume_Triggered(MENU_VOLUME_FADE_IN_FAST, clip_ids, "Start of Clip")
                self.Volume_Triggered(MENU_VOLUME_FADE_OUT_FAST, clip_ids, "End of Clip")
                return
            elif position == "Entire Clip" and action == MENU_VOLUME_FADE_IN_OUT_SLOW:
                # Call this method for the start and end of the clip
                self.Volume_Triggered(MENU_VOLUME_FADE_IN_SLOW, clip_ids, "Start of Clip")
                self.Volume_Triggered(MENU_VOLUME_FADE_OUT_SLOW, clip_ids, "End of Clip")
                return

            if action == MENU_VOLUME_NONE:
                # Clear all keyframes
                p = openshot.Point(1, 1.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data['volume'] = {"Points": [p_object]}

            if action in [
                MENU_VOLUME_FADE_IN_FAST,
                MENU_VOLUME_FADE_IN_SLOW
            ]:
                # Add keyframes
                start = openshot.Point(start_animation, 0.0, openshot.BEZIER)
                start_object = json.loads(start.Json())
                end = openshot.Point(end_animation, 1.0, openshot.BEZIER)
                end_object = json.loads(end.Json())
                self.AddPoint(clip.data['volume'], start_object)
                self.AddPoint(clip.data['volume'], end_object)

            if action in [
                MENU_VOLUME_FADE_OUT_FAST,
                MENU_VOLUME_FADE_OUT_SLOW
            ]:
                # Add keyframes
                start = openshot.Point(start_animation, 1.0, openshot.BEZIER)
                start_object = json.loads(start.Json())
                end = openshot.Point(end_animation, 0.0, openshot.BEZIER)
                end_object = json.loads(end.Json())
                self.AddPoint(clip.data['volume'], start_object)
                self.AddPoint(clip.data['volume'], end_object)

            if action in [
                MENU_VOLUME_LEVEL_100,
                MENU_VOLUME_LEVEL_90,
                MENU_VOLUME_LEVEL_80,
                MENU_VOLUME_LEVEL_70,
                MENU_VOLUME_LEVEL_60,
                MENU_VOLUME_LEVEL_50,
                MENU_VOLUME_LEVEL_40,
                MENU_VOLUME_LEVEL_30,
                MENU_VOLUME_LEVEL_20,
                MENU_VOLUME_LEVEL_10,
                MENU_VOLUME_LEVEL_0
            ]:
                # Add keyframes
                p = openshot.Point(start_animation, float(action) / 100.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                self.AddPoint(clip.data['volume'], p_object)

            # Save changes
            self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

            # Determine if waveform needs to be redrawn
            self.run_js(JS_SCOPE_SELECTOR + ".hasAudioData('{}');".format(clip.id), partial(callback, self, clip.id))

    def Rotate_Triggered(self, action, clip_ids, position="Start of Clip"):
        """Callback for rotate context menus"""
        log.debug(action)

        # Loop through each selected clip
        for clip_id in clip_ids:

            # Get existing clip object
            clip = Clip.get(id=clip_id)
            if not clip:
                # Invalid clip, skip to next item
                continue

            if action == MENU_ROTATE_NONE:
                # Clear all keyframes
                p = openshot.Point(1, 0.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data['rotation'] = {"Points": [p_object]}

            if action == MENU_ROTATE_90_RIGHT:
                # Add keyframes
                p = openshot.Point(1, 90.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data['rotation'] = {"Points": [p_object]}

            if action == MENU_ROTATE_90_LEFT:
                # Add keyframes
                p = openshot.Point(1, -90.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data['rotation'] = {"Points": [p_object]}

            if action == MENU_ROTATE_180_FLIP:
                # Add keyframes
                p = openshot.Point(1, 180.0, openshot.BEZIER)
                p_object = json.loads(p.Json())
                clip.data['rotation'] = {"Points": [p_object]}

            # Save changes
            self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

    def Time_Triggered(self, action, clip_ids, speed="1X", playhead_position=0.0):
        """Callback for time context menus"""
        log.debug(action)

        # Get FPS from project
        fps = get_app().project.get("fps")
        fps_float = float(fps["num"]) / float(fps["den"])

        # Loop through each selected clip
        for clip_id in clip_ids:
            # Get existing clip object
            clip = Clip.get(id=clip_id)

            if not clip:
                # Invalid clip, skip to next item
                continue

            # Keep original 'end' and 'duration'
            if "original_data" not in clip.data:
                clip.data["original_data"] = {
                    "end": clip.data["end"],
                    "duration": clip.data["duration"],
                    "video_length": clip.data["reader"]["video_length"]
                }

            # Determine the beginning and ending of this animation
            start_animation = 1

            # Freeze or Speed?
            if action in [
                MENU_TIME_FREEZE,
                MENU_TIME_FREEZE_ZOOM
            ]:
                # Get freeze details
                freeze_seconds = float(speed)

                original_duration = clip.data["duration"]
                if "original_data" in clip.data:
                    original_duration = clip.data["original_data"]["duration"]

                log.info('Updating timing for clip ID {}, original duration: {}'
                         .format(clip.id, original_duration))
                log.debug(clip.data)

                # Extend end & duration (due to freeze)
                clip.data["end"] = float(clip.data["end"]) + freeze_seconds
                clip.data["duration"] = float(clip.data["duration"]) + freeze_seconds
                clip.data["reader"]["video_length"] = float(clip.data["reader"]["video_length"]) + freeze_seconds

                # Determine start frame from position
                start_animation_seconds = float(clip.data["start"]) + (playhead_position - float(clip.data["position"]))
                start_animation_frames = round(start_animation_seconds * fps_float) + 1
                start_animation_frames_value = start_animation_frames
                end_animation_seconds = start_animation_seconds + freeze_seconds
                end_animation_frames = round(end_animation_seconds * fps_float) + 1
                end_of_clip_seconds = float(clip.data["duration"])
                end_of_clip_frames = round((end_of_clip_seconds) * fps_float) + 1
                end_of_clip_frames_value = round((original_duration) * fps_float) + 1

                # Determine volume start and end
                start_volume_value = 1.0

                # Do we already have a time curve? Look up intersecting frame # from time curve
                if len(clip.data["time"]["Points"]) > 1:
                    # Delete last time point (which should be the end of the clip). We have a new end of the clip
                    # after inserting this freeze.
                    del clip.data["time"]["Points"][-1]

                    # Find actual clip object from libopenshot
                    c = self.window.timeline_sync.timeline.GetClip(clip_id)
                    if c:
                        # Look up correct position from time curve
                        start_animation_frames_value = c.time.GetLong(start_animation_frames)

                # Do we already have a volume curve? Look up intersecting frame # from volume curve
                if len(clip.data["volume"]["Points"]) > 1:
                    # Find actual clip object from libopenshot
                    c = self.window.timeline_sync.timeline.GetClip(clip_id)
                    if c:
                        # Look up correct volume from time curve
                        start_volume_value = c.volume.GetValue(start_animation_frames)

                # Create Time Freeze keyframe points
                p = openshot.Point(start_animation_frames, start_animation_frames_value, openshot.LINEAR)
                p_object = json.loads(p.Json())
                self.AddPoint(clip.data['time'], p_object)
                p1 = openshot.Point(end_animation_frames, start_animation_frames_value, openshot.LINEAR)
                p1_object = json.loads(p1.Json())
                self.AddPoint(clip.data['time'], p1_object)
                p2 = openshot.Point(end_of_clip_frames, end_of_clip_frames_value, openshot.LINEAR)
                p2_object = json.loads(p2.Json())
                self.AddPoint(clip.data['time'], p2_object)

                # Create Volume mute keyframe points (so the freeze is silent)
                p = openshot.Point(start_animation_frames - 1, start_volume_value, openshot.LINEAR)
                p_object = json.loads(p.Json())
                self.AddPoint(clip.data['volume'], p_object)
                p = openshot.Point(start_animation_frames, 0.0, openshot.LINEAR)
                p_object = json.loads(p.Json())
                self.AddPoint(clip.data['volume'], p_object)
                p2 = openshot.Point(end_animation_frames - 1, 0.0, openshot.LINEAR)
                p2_object = json.loads(p2.Json())
                self.AddPoint(clip.data['volume'], p2_object)
                p3 = openshot.Point(end_animation_frames, start_volume_value, openshot.LINEAR)
                p3_object = json.loads(p3.Json())
                self.AddPoint(clip.data['volume'], p3_object)

                # Create zoom keyframe points
                if action == MENU_TIME_FREEZE_ZOOM:
                    p = openshot.Point(start_animation_frames, 1.0, openshot.BEZIER)
                    p_object = json.loads(p.Json())
                    self.AddPoint(clip.data['scale_x'], p_object)
                    p = openshot.Point(start_animation_frames, 1.0, openshot.BEZIER)
                    p_object = json.loads(p.Json())
                    self.AddPoint(clip.data['scale_y'], p_object)

                    diff_halfed = (end_animation_frames - start_animation_frames) / 2.0
                    p1 = openshot.Point(start_animation_frames + diff_halfed, 1.05, openshot.BEZIER)
                    p1_object = json.loads(p1.Json())
                    self.AddPoint(clip.data['scale_x'], p1_object)
                    p1 = openshot.Point(start_animation_frames + diff_halfed, 1.05, openshot.BEZIER)
                    p1_object = json.loads(p1.Json())
                    self.AddPoint(clip.data['scale_y'], p1_object)

                    p1 = openshot.Point(end_animation_frames, 1.0, openshot.BEZIER)
                    p1_object = json.loads(p1.Json())
                    self.AddPoint(clip.data['scale_x'], p1_object)
                    p1 = openshot.Point(end_animation_frames, 1.0, openshot.BEZIER)
                    p1_object = json.loads(p1.Json())
                    self.AddPoint(clip.data['scale_y'], p1_object)

            else:

                # Calculate speed factor
                speed_label = speed.replace('X', '')
                speed_parts = speed_label.split('/')
                even_multiple = 1
                if len(speed_parts) == 2:
                    speed_factor = float(speed_parts[0]) / float(speed_parts[1])
                    even_multiple = int(speed_parts[1])
                else:
                    speed_factor = float(speed_label)
                    even_multiple = int(speed_factor)

                # Clear all keyframes
                p = openshot.Point(start_animation, 0.0, openshot.LINEAR)
                p_object = json.loads(p.Json())
                clip.data['time'] = {"Points": [p_object]}

                # Get the ending frame
                end_of_clip = round(float(clip.data["end"]) * fps_float) + 1

                # Determine the beginning and ending of this animation
                start_animation = round(float(clip.data["start"]) * fps_float) + 1
                duration_animation = self.round_to_multiple(end_of_clip - start_animation, even_multiple)
                end_animation = start_animation + duration_animation

                if action == MENU_TIME_FORWARD:
                    # Add keyframes
                    start = openshot.Point(start_animation, start_animation, openshot.LINEAR)
                    start_object = json.loads(start.Json())
                    clip.data['time'] = {"Points": [start_object]}
                    end = openshot.Point(
                        start_animation + (duration_animation / speed_factor), end_animation, openshot.LINEAR)
                    end_object = json.loads(end.Json())
                    self.AddPoint(clip.data['time'], end_object)

                    # Adjust end & duration
                    clip.data["end"] = (start_animation + (duration_animation / speed_factor)) / fps_float
                    clip.data["duration"] = self.round_to_multiple(clip.data["duration"] / speed_factor, even_multiple)
                    clip.data["reader"]["video_length"] = self.round_to_multiple(
                        float(clip.data["reader"]["video_length"]) / speed_factor, even_multiple)

                if action == MENU_TIME_BACKWARD:
                    # Add keyframes
                    start = openshot.Point(start_animation, end_animation, openshot.LINEAR)
                    start_object = json.loads(start.Json())
                    clip.data['time'] = {"Points": [start_object]}
                    end = openshot.Point(
                        start_animation + (duration_animation / speed_factor), start_animation, openshot.LINEAR)
                    end_object = json.loads(end.Json())
                    self.AddPoint(clip.data['time'], end_object)

                    # Adjust end & duration
                    clip.data["end"] = (start_animation + (duration_animation / speed_factor)) / fps_float
                    clip.data["duration"] = self.round_to_multiple(clip.data["duration"] / speed_factor, even_multiple)
                    clip.data["reader"]["video_length"] = self.round_to_multiple(
                        float(clip.data["reader"]["video_length"]) / speed_factor, even_multiple)

                if action == MENU_TIME_NONE and "original_data" in clip.data:
                    # Reset original end & duration (if available)
                    orig = clip.data.pop("original_data")
                    clip.data.update({
                        "end": orig["end"],
                        "duration": orig["duration"],
                    })
                    clip.data["reader"]["video_length"] = orig["video_length"]

            # Save changes
            self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

    def round_to_multiple(self, number, multiple):
        """Round this to the closest multiple of a given #"""
        return number - (number % multiple)

    def show_all_clips(self, clip, stretch=False):
        """ Show all clips at the same time (arranged col by col, row by row)  """
        from math import sqrt

        # Get list of nearby clips
        available_clips = []
        start_position = float(clip.data["position"])
        for c in Clip.filter():
            if (float(c.data["position"]) >= (start_position - 0.5)
               and float(c.data["position"]) <= (start_position + 0.5)):
                # add to list
                available_clips.append(c)

        # Get the number of rows
        number_of_clips = len(available_clips)
        number_of_rows = int(sqrt(number_of_clips))
        max_clips_on_row = float(number_of_clips) / float(number_of_rows)

        # Determine how many clips per row
        if max_clips_on_row > float(int(max_clips_on_row)):
            max_clips_on_row = int(max_clips_on_row + 1)
        else:
            max_clips_on_row = int(max_clips_on_row)

        # Calculate Height & Width
        height = 1.0 / float(number_of_rows)
        width = 1.0 / float(max_clips_on_row)

        clip_index = 0

        # Loop through each row of clips
        for row in range(0, number_of_rows):

            # Loop through clips on this row
            for col in range(0, max_clips_on_row):
                if clip_index >= number_of_clips:
                    continue

                # Calculate X & Y
                X = float(col) * width
                Y = float(row) * height

                # Modify clip layout settings
                selected_clip = available_clips[clip_index]
                selected_clip.data["gravity"] = openshot.GRAVITY_TOP_LEFT

                if stretch:
                    selected_clip.data["scale"] = openshot.SCALE_STRETCH
                else:
                    selected_clip.data["scale"] = openshot.SCALE_FIT

                # Set scale keyframes
                w = openshot.Point(1, width, openshot.BEZIER)
                w_object = json.loads(w.Json())
                selected_clip.data["scale_x"] = {"Points": [w_object]}
                h = openshot.Point(1, height, openshot.BEZIER)
                h_object = json.loads(h.Json())
                selected_clip.data["scale_y"] = {"Points": [h_object]}
                x_point = openshot.Point(1, X, openshot.BEZIER)
                x_object = json.loads(x_point.Json())
                selected_clip.data["location_x"] = {"Points": [x_object]}
                y_point = openshot.Point(1, Y, openshot.BEZIER)
                y_object = json.loads(y_point.Json())
                selected_clip.data["location_y"] = {"Points": [y_object]}

                log.info('Updating clip id: %s' % selected_clip.data["id"])
                log.info('width: %s, height: %s' % (width, height))

                # Increment Clip Index
                clip_index += 1

                # Save changes
                self.update_clip_data(selected_clip.data, only_basic_props=False, ignore_reader=True)

    def Reverse_Transition_Triggered(self, tran_ids):
        """Callback for reversing a transition"""
        log.info("Reverse_Transition_Triggered")

        # Loop through all selected transitions
        for tran_id in tran_ids:

            # Get existing clip object
            tran = Transition.get(id=tran_id)
            if not tran:
                # Invalid transition, skip to next item
                continue

            # Loop through brightness keyframes
            tran_data_copy = deepcopy(tran.data)
            new_index = len(tran.data["brightness"]["Points"])
            for point in tran.data["brightness"]["Points"]:
                new_index -= 1
                tran_data_copy["brightness"]["Points"][new_index]["co"]["Y"] = point["co"]["Y"]
                if "handle_left" in point:
                    tran_data_copy["brightness"]["Points"][new_index]["handle_left"]["Y"] = point["handle_left"]["Y"]
                    tran_data_copy["brightness"]["Points"][new_index]["handle_right"]["Y"] = point["handle_right"]["Y"]

            # Save changes
            self.update_transition_data(tran_data_copy, only_basic_props=False)

    @pyqtSlot(str)
    def ShowTransitionMenu(self, tran_id=None):
        log.info('ShowTransitionMenu: %s' % tran_id)

        # Get translation method
        _ = get_app()._tr

        # Get existing transition object
        tran = Transition.get(id=tran_id)
        if not tran:
            # Not a valid transition id
            return

        # Set the selected transition (if needed)
        if tran_id not in self.window.selected_transitions:
            self.window.addSelection(tran_id, 'transition')
        # Get list of all selected transitions
        tran_ids = self.window.selected_transitions
        clip_ids = self.window.selected_clips

        # Get framerate
        fps = get_app().project.get("fps")
        fps_float = float(fps["num"]) / float(fps["den"])

        # Get playhead position
        playhead_position = float(self.window.preview_thread.current_frame) / fps_float

        menu = QMenu(self)

        # Copy Menu
        if len(tran_ids) + len(clip_ids) > 1:
            # Copy All Menu (Clips and/or transitions are selected)
            Copy_All = menu.addAction(_("Copy"))
            Copy_All.setShortcut(QKeySequence(self.window.getShortcutByName("copyAll")))
            Copy_All.triggered.connect(partial(self.Copy_Triggered, MENU_COPY_ALL, clip_ids, tran_ids))
        else:
            # Only a single transitions is selected (show normal transition copy menu)
            Copy_Menu = QMenu(_("Copy"), self)
            Copy_Tran = Copy_Menu.addAction(_("Transition"))
            Copy_Tran.setShortcut(QKeySequence(self.window.getShortcutByName("copyAll")))
            Copy_Tran.triggered.connect(partial(self.Copy_Triggered, MENU_COPY_TRANSITION, [], [tran_id]))

            Keyframe_Menu = QMenu(_("Keyframes"), self)
            Copy_Keyframes_All = Keyframe_Menu.addAction(_("All"))
            Copy_Keyframes_All.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_ALL, [], [tran_id]))
            Keyframe_Menu.addSeparator()
            Copy_Keyframes_Brightness = Keyframe_Menu.addAction(_("Brightness"))
            Copy_Keyframes_Brightness.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_BRIGHTNESS, [], [tran_id]))
            Copy_Keyframes_Scale = Keyframe_Menu.addAction(_("Contrast"))
            Copy_Keyframes_Scale.triggered.connect(partial(
                self.Copy_Triggered, MENU_COPY_KEYFRAMES_CONTRAST, [], [tran_id]))

            # Only show copy->keyframe if a single transitions is selected
            Copy_Menu.addMenu(Keyframe_Menu)
            menu.addMenu(Copy_Menu)

        # Get list of clipboard items (that are complete clips or transitions)
        # i.e. ignore partial clipboard items (keyframes / effects / etc...)
        clipboard_clip_ids = [k for k, v in self.copy_clipboard.items() if v.get('id')]
        clipboard_tran_ids = [k for k, v in self.copy_transition_clipboard.items() if v.get('id')]
        # Determine if the paste menu should be shown
        if self.copy_transition_clipboard and len(clipboard_clip_ids) + len(clipboard_tran_ids) == 0:
            # Paste Menu (Only show when partial transition clipboard available)
            Paste_Tran = menu.addAction(_("Paste"))
            Paste_Tran.triggered.connect(partial(self.Paste_Triggered, MENU_PASTE, 0.0, 0, [], tran_ids))

        menu.addSeparator()

        # Alignment Menu (if multiple selections)
        if len(clip_ids) > 1:
            Alignment_Menu = QMenu(_("Align"), self)
            Align_Left = Alignment_Menu.addAction(_("Left"))
            Align_Left.triggered.connect(partial(self.Align_Triggered, MENU_ALIGN_LEFT, clip_ids, tran_ids))
            Align_Right = Alignment_Menu.addAction(_("Right"))
            Align_Right.triggered.connect(partial(self.Align_Triggered, MENU_ALIGN_RIGHT, clip_ids, tran_ids))

            # Add menu to parent
            menu.addMenu(Alignment_Menu)

        # If Playhead overlapping transition
        if tran:
            start_of_tran = float(tran.data["start"])
            end_of_tran = float(tran.data["end"])
            position_of_tran = float(tran.data["position"])
            if (playhead_position >= position_of_tran
               and playhead_position <= (position_of_tran + (end_of_tran - start_of_tran))):
                # Add split transition menu
                Slice_Menu = QMenu(_("Slice"), self)
                Slice_Keep_Both = Slice_Menu.addAction(_("Keep Both Sides"))
                Slice_Keep_Both.triggered.connect(partial(
                    self.Slice_Triggered, MENU_SLICE_KEEP_BOTH, [], [tran_id], playhead_position))
                Slice_Keep_Left = Slice_Menu.addAction(_("Keep Left Side"))
                Slice_Keep_Left.triggered.connect(partial(
                    self.Slice_Triggered, MENU_SLICE_KEEP_LEFT, [], [tran_id], playhead_position))
                Slice_Keep_Right = Slice_Menu.addAction(_("Keep Right Side"))
                Slice_Keep_Right.triggered.connect(partial(
                    self.Slice_Triggered, MENU_SLICE_KEEP_RIGHT, [], [tran_id], playhead_position))
                menu.addMenu(Slice_Menu)

        # Reverse Transition menu
        Reverse_Transition = menu.addAction(_("Reverse Transition"))
        Reverse_Transition.triggered.connect(partial(self.Reverse_Transition_Triggered, tran_ids))

        # Properties
        menu.addSeparator()
        menu.addAction(self.window.actionProperties)

        # Remove transition menu
        menu.addSeparator()
        menu.addAction(self.window.actionRemoveTransition)

        # Show menu
        return menu.exec_(QCursor.pos())

    @pyqtSlot(str)
    def ShowTrackMenu(self, layer_id=None):
        log.info('ShowTrackMenu: %s' % layer_id)

        if layer_id not in self.window.selected_tracks:
            self.window.selected_tracks = [layer_id]

        # Get track object
        track = Track.get(id=layer_id)
        locked = track.data.get("lock", False)

        menu = QMenu(self)
        menu.addAction(self.window.actionAddTrackAbove)
        menu.addAction(self.window.actionAddTrackBelow)
        menu.addAction(self.window.actionRenameTrack)
        if locked:
            menu.addAction(self.window.actionUnlockTrack)
            self.window.actionRemoveTrack.setEnabled(False)
        else:
            menu.addAction(self.window.actionLockTrack)
            self.window.actionRemoveTrack.setEnabled(True)
        menu.addSeparator()
        menu.addAction(self.window.actionRemoveTrack)
        return menu.exec_(QCursor.pos())

    @pyqtSlot(str)
    def ShowMarkerMenu(self, marker_id=None):
        log.info('ShowMarkerMenu: %s' % marker_id)

        if marker_id not in self.window.selected_markers:
            self.window.selected_markers = [marker_id]

        menu = QMenu(self)
        menu.addAction(self.window.actionRemoveMarker)
        return menu.exec_(QCursor.pos())

    @pyqtSlot(str, int)
    def PreviewClipFrame(self, clip_id, frame_number):

        # Get existing clip object
        clip = Clip.get(id=clip_id)
        if not clip:
            # Invalid clip
            return

        preview_path = clip.data['reader']['path']

        # Adjust frame # to valid range
        frame_number = max(frame_number, 1)
        frame_number = min(frame_number, int(clip.data['reader']['video_length']))

        # Load the clip into the Player (ignored if this has already happened)
        self.window.LoadFileSignal.emit(preview_path)
        self.window.SpeedSignal.emit(0)

        # Seek to frame
        self.window.SeekSignal.emit(frame_number)

    @pyqtSlot(int)
    def PlayheadMoved(self, position_frames):

        # Load the timeline into the Player (ignored if this has already happened)
        self.window.LoadFileSignal.emit('')

        if self.last_position_frames != position_frames:
            # Update time code (to prevent duplicate previews)
            self.last_position_frames = position_frames

            # Notify main window of current frame
            self.window.previewFrame(position_frames)

    @pyqtSlot(int)
    def movePlayhead(self, position_frames):
        """ Move the playhead since the position has changed inside OpenShot (probably due to the video player) """
        # Get access to timeline scope and set scale to zoom slider value (passed in)
        self.run_js(JS_SCOPE_SELECTOR + ".movePlayheadToFrame(%s);" % (str(position_frames)))

    @pyqtSlot()
    def centerOnPlayhead(self):
        """ Center the timeline on the current playhead position """
        # Execute JavaScript to center the timeline
        self.run_js(JS_SCOPE_SELECTOR + '.centerOnPlayhead();')

    @pyqtSlot(int)
    def SetSnappingMode(self, enable_snapping):
        """ Enable / Disable snapping mode """
        # Init snapping state (1 = snapping, 0 = no snapping)
        self.run_js(JS_SCOPE_SELECTOR + ".setSnappingMode(%s);" % int(enable_snapping))

    @pyqtSlot(int)
    def SetRazorMode(self, enable_razor):
        """ Enable / Disable razor mode """
        # Init razor state (1 = razor, 0 = no razor)
        self.run_js(JS_SCOPE_SELECTOR + ".setRazorMode(%s);" % int(enable_razor))

    @pyqtSlot(int)
    def SetPlayheadFollow(self, enable_follow):
        """ Enable / Disable playhead follow on seek """
        self.run_js(JS_SCOPE_SELECTOR + ".setFollow({});".format(int(enable_follow)))

    @pyqtSlot(str, str, bool)
    def addSelection(self, item_id, item_type, clear_existing=False):
        """ Add the selected item to the current selection """

        # Add to main window
        self.window.addSelection(item_id, item_type, clear_existing)

    @pyqtSlot(str, str)
    def removeSelection(self, item_id, item_type):
        """ Remove the selected clip from the selection """

        # Remove from main window
        self.window.removeSelection(item_id, item_type)

    @pyqtSlot(str, str)
    def qt_log(self, level="INFO", message=None):
        levels = {
            "DEBUG": logging.DEBUG,
            "INFO": logging.INFO,
            "WARN": logging.WARNING,
            "WARNING": logging.WARNING,
            "ERROR": logging.ERROR,
            "CRITICAL": logging.CRITICAL,
            "FATAL": logging.FATAL,
            }
        if isinstance(level, str):
            level = levels.get(level, logging.INFO)
        self.log_fn(level, message)

    # Handle changes to zoom level, update js
    def update_zoom(self, newValue):
        _ = get_app()._tr

        # Convert slider value (passed in) to a scale (in seconds)
        newScale = zoomToSeconds(newValue)

        # Set zoom label
        self.window.zoomScaleLabel.setText(_("{} seconds").format(newScale))

        # Determine X coordinate of cursor (to center zoom on)
        cursor_y = self.mapFromGlobal(self.cursor().pos()).y()
        if cursor_y >= 0:
            cursor_x = self.mapFromGlobal(self.cursor().pos()).x()
        else:
            cursor_x = 0

        # Get access to timeline scope and set scale to new computed value
        self.run_js(JS_SCOPE_SELECTOR + ".setScale(" + str(newScale) + "," + str(cursor_x) + ");")

        # Start or restart timer to redraw audio
        self.redraw_audio_timer.start()

        # Only update scale if different
        current_scale = get_app().project.get("scale")

        # Save current zoom
        if newScale != current_scale:
            get_app().updates.ignore_history = True
            get_app().updates.update(["scale"], newScale)
            get_app().updates.ignore_history = False

    # Capture wheel event to alter zoom slider control
    def wheelEvent(self, event):
        if event.modifiers() & Qt.ControlModifier:
            event.accept()
            zoom = self.window.sliderZoom
            # For each 120 (standard scroll unit) adjust the zoom slider
            tick_scale = 120
            steps = int(event.angleDelta().y() / tick_scale)
            delta = zoom.pageStep() * steps
            log.debug("Zooming by %d steps", -steps)
            zoom.setValue(zoom.value() - delta)
        else:
            super().wheelEvent(event)

    # An item is being dragged onto the timeline (mouse is entering the timeline now)
    def dragEnterEvent(self, event):

        # If a plain text drag accept
        if not self.new_item and not event.mimeData().hasUrls() and event.mimeData().html():
            # get type of dropped data
            self.item_type = event.mimeData().html()

            # Track that a new item is being 'added'
            self.new_item = True

            # Get the mime data (i.e. list of files, list of transitions, etc...)
            data = json.loads(event.mimeData().text())
            pos = event.posF()

            # create the item
            if self.item_type == "clip":
                self.addClip(data, pos)
            elif self.item_type == "transition":
                self.addTransition(data, pos)

            # accept all events, even if a new clip is not being added
            event.accept()

        # Accept a plain file URL (from the OS)
        elif not self.new_item and event.mimeData().hasUrls():
            # Track that a new item is being 'added'
            self.new_item = True
            self.item_type = "os_drop"

            # accept event
            event.accept()

    # Add Clip
    def addClip(self, data, event_position):

        # Callback function, to actually add the clip object
        def callback(self, data, callback_data):
            js_position = callback_data.get('position', 0.0)
            js_nearest_track = callback_data.get('track', 0)

            # Search for matching file in project data (if any)
            file_id = data[0]
            file = File.get(id=file_id)

            if not file:
                # File not found, do nothing
                return

            # Get file name
            filename = os.path.basename(file.data["path"])

            # Convert path to the correct relative path (based on this folder)
            file_path = file.absolute_path()

            # Create clip object for this file
            c = openshot.Clip(file_path)

            # Append missing attributes to Clip JSON
            new_clip = json.loads(c.Json())
            new_clip["file_id"] = file.id
            new_clip["title"] = filename

            # Skip any clips that are missing a 'reader' attribute
            if not new_clip.get("reader"):
                return  # Do nothing

            # Check for optional start and end attributes
            if 'start' in file.data:
                new_clip["start"] = file.data['start']
            if 'end' in file.data:
                new_clip["end"] = file.data['end']

            # Set position and closet track
            new_clip["position"] = js_position
            new_clip["layer"] = js_nearest_track

            # Adjust clip duration, start, and end
            new_clip["duration"] = new_clip["reader"]["duration"]
            if file.data["media_type"] == "image":
                new_clip["end"] = settings.get_settings().get("default-image-length")  # default to 8 seconds

            # Overwrite frame rate (incase the user changed it in the File Properties)
            file_properties_fps = float(file.data["fps"]["num"]) / float(file.data["fps"]["den"])
            file_fps = float(new_clip["reader"]["fps"]["num"]) / float(new_clip["reader"]["fps"]["den"])
            fps_diff = file_fps / file_properties_fps
            new_clip["reader"]["fps"]["num"] = file.data["fps"]["num"]
            new_clip["reader"]["fps"]["den"] = file.data["fps"]["den"]
            # Scale duration / length / and end properties
            new_clip["reader"]["duration"] *= fps_diff
            new_clip["end"] *= fps_diff
            new_clip["duration"] *= fps_diff

            # Add clip to timeline
            self.update_clip_data(new_clip, only_basic_props=False)

            # temp hold item_id
            self.item_id = new_clip.get('id')

            # Init javascript bounding box (for snapping support)
            self.run_js(JS_SCOPE_SELECTOR + ".startManualMove('{}', '{}');".format(self.item_type, self.item_id))

        # Find position from javascript
        self.run_js(JS_SCOPE_SELECTOR + ".getJavaScriptPosition({}, {});"
            .format(event_position.x(), event_position.y()), partial(callback, self, data))

    # Resize timeline
    @pyqtSlot(float)
    def resizeTimeline(self, new_duration):
        """Resize the duration of the timeline"""
        get_app().updates.update(["duration"], new_duration)

    # Add Transition
    def addTransition(self, file_ids, event_position):

        # Callback function, to actually add the transition object
        def callback(self, file_ids, callback_data):
            js_position = callback_data.get('position', 0.0)
            js_nearest_track = callback_data.get('track', 0)

            # Get FPS from project
            fps = get_app().project.get("fps")
            fps_float = float(fps["num"]) / float(fps["den"])

            # Open up QtImageReader for transition Image
            transition_reader = openshot.QtImageReader(file_ids[0])

            brightness = openshot.Keyframe()
            brightness.AddPoint(1, 1.0, openshot.BEZIER)
            brightness.AddPoint(round(10 * fps_float) + 1, -1.0, openshot.BEZIER)
            contrast = openshot.Keyframe(3.0)

            # Create transition dictionary
            transitions_data = {
                "id": get_app().project.generate_id(),
                "layer": js_nearest_track,
                "title": "Transition",
                "type": "Mask",
                "position": js_position,
                "start": 0,
                "end": 10,
                "brightness": json.loads(brightness.Json()),
                "contrast": json.loads(contrast.Json()),
                "reader": json.loads(transition_reader.Json()),
                "replace_image": False
            }

            # Send to update manager
            self.update_transition_data(transitions_data, only_basic_props=False)

            # temp keep track of id
            self.item_id = transitions_data.get('id')

            # Init javascript bounding box (for snapping support)
            self.run_js(JS_SCOPE_SELECTOR + ".startManualMove('{}','{}');".format(self.item_type, self.item_id))

        # Find position from javascript
        self.run_js(JS_SCOPE_SELECTOR + ".getJavaScriptPosition({}, {});"
            .format(event_position.x(), event_position.y()),
            partial(callback, self, file_ids))

    # Add Effect
    def addEffect(self, effect_names, event_position):

        # Callback function, to actually add the effect object
        def callback(self, effect_names, callback_data):
            js_position = callback_data.get('position', 0.0)
            js_nearest_track = callback_data.get('track', 0)

            # Get name of effect
            name = effect_names[0]

            # Loop through clips on the closest layer
            possible_clips = Clip.filter(layer=js_nearest_track)
            for clip in possible_clips:
                if js_position == 0 or (
                    clip.data["position"]
                    <= js_position
                    <= clip.data["position"] + (clip.data["end"] - clip.data["start"])
                ):
                    log.info("Applying effect {} to clip ID {}".format(name, clip.id))
                    log.debug(clip)

                    # Handle custom effect dialogs
                    if name in effect_options:

                        # Get effect options
                        effect_params = effect_options.get(name)

                        # Show effect pre-processing window
                        from windows.process_effect import ProcessEffect

                        try:
                            win = ProcessEffect(clip.id, name, effect_params)

                        except ModuleNotFoundError as e:
                            print("[ERROR]: " + str(e))
                            return

                        print("Effect %s" % name)
                        print("Effect options: %s" % effect_options)

                        # Run the dialog event loop - blocking interaction on this window during this time
                        result = win.exec_()

                        if result == QDialog.Accepted:
                            log.info('Start processing')
                        else:
                            log.info('Cancel processing')
                            return

                        # Create Effect
                        effect = win.effect # effect.Id already set

                        if effect is None:
                            break
                    else:
                        # Create Effect
                        effect = openshot.EffectInfo().CreateEffect(name)

                        # Get Effect JSON
                        effect.Id(get_app().project.generate_id())
                    effect_json = json.loads(effect.Json())

                    # Append effect JSON to clip
                    clip.data["effects"].append(effect_json)

                    # Update clip data for project
                    self.update_clip_data(clip.data, only_basic_props=False, ignore_reader=True)

        # Find position from javascript
        self.run_js(JS_SCOPE_SELECTOR + ".getJavaScriptPosition({}, {});"
            .format(event_position.x(), event_position.y()), partial(callback, self, effect_names))

    # Without defining this method, the 'copy' action doesn't show with cursor
    def dragMoveEvent(self, event):
        # Accept all move events
        event.accept()

        # Get cursor position
        pos = event.posF()

        # Move clip on timeline
        if self.item_type in ["clip", "transition"]:
            self.run_js(JS_SCOPE_SELECTOR + ".moveItem({}, {}, '{}');".format(pos.x(), pos.y(), self.item_type))

    # Drop an item on the timeline
    def dropEvent(self, event):
        log.info("Dropping item on timeline - item_id: %s, item_type: %s" % (self.item_id, self.item_type))

        # Accept event
        event.accept()

        # Get position of cursor
        pos = event.posF()

        if self.item_type in ["clip", "transition"] and self.item_id:
            # Update most recent clip
            self.run_js(JS_SCOPE_SELECTOR + ".updateRecentItemJSON('{}', '{}');".format(self.item_type, self.item_id))

        elif self.item_type == "effect":
            # Add effect only on drop
            data = json.loads(event.mimeData().text())
            self.addEffect(data, pos)

        elif self.item_type == "os_drop":
            # Add new files to project
            self.window.filesView.dropEvent(event)

            # Add clips for each file dropped
            for uri in event.mimeData().urls():
                filepath = uri.toLocalFile()
                if not os.path.exists(filepath) or not os.path.isfile(filepath):
                    continue

                # Valid file, so create clip for it
                log.debug('Adding clip for {}'.format(os.path.basename(filepath)))
                for file in File.filter(path=filepath):
                    # Insert clip for this file at this position
                    self.addClip([file.id], pos)

        # Clear new clip
        self.new_item = False
        self.item_type = None
        self.item_id = None

        # Update the preview and reselct current frame in properties
        self.window.refreshFrameSignal.emit()
        self.window.propertyTableView.select_frame(self.window.preview_thread.player.Position())

    def dragLeaveEvent(self, event):
        """A drag is in-progress and the user moves mouse outside of timeline"""
        log.debug('dragLeaveEvent - Undo drop')

        # Accept event
        event.accept()

        # Clear selected clips
        self.window.removeSelection(self.item_id, self.item_type)

        if self.item_type == "clip":
            # Delete dragging clip
            clips = Clip.filter(id=self.item_id)
            for c in clips:
                c.delete()

        elif self.item_type == "transition":
            # Delete dragging transitions
            transitions = Transition.filter(id=self.item_id)
            for t in transitions:
                t.delete()

        # Clear new clip
        self.new_item = False
        self.item_type = None
        self.item_id = None

    def redraw_audio_onTimeout(self):
        """Timer is ready to redraw audio (if any)"""
        log.debug('redraw_audio_onTimeout')

        # Pass to javascript timeline (and render)
        self.run_js(JS_SCOPE_SELECTOR + ".reDrawAllAudioData();")

    def ClearAllSelections(self):
        """Clear all selections in JavaScript"""

        # Call javascript command
        self.run_js(JS_SCOPE_SELECTOR + ".clearAllSelections();")

    def SelectAll(self):
        """Select all clips and transitions in JavaScript"""

        # Call javascript command
        self.run_js(JS_SCOPE_SELECTOR + ".selectAll();")

    def render_cache_json(self):
        """Render the cached frames to the timeline (called every X seconds), and only if changed"""

        # Get final cache object from timeline
        try:
            cache_object = self.window.timeline_sync.timeline.GetCache()
            if not cache_object or cache_object.Count() <= 0:
                return
            # Get the JSON from the cache object (i.e. which frames are cached)
            cache_json = self.window.timeline_sync.timeline.GetCache().Json()
            cache_dict = json.loads(cache_json)
            cache_version = cache_dict["version"]

            if self.cache_renderer_version == cache_version:
                # Nothing has changed, ignore
                return
            # Cache has changed, re-render it
            self.cache_renderer_version = cache_version
            self.run_js(JS_SCOPE_SELECTOR + ".renderCache({});".format(cache_json))
        except Exception as ex:
            # Log the exception and ignore
            log.warning("Exception processing timeline cache: %s", ex)

    def __init__(self, window):
        super().__init__()
        self.setObjectName("TimelineWebView")

        app = get_app()
        self.window = window
        self.setAcceptDrops(True)
        self.last_position_frames = None

        # Get logger
        self.log_fn = log.log

        # Add self as listener to project data updates (used to update the timeline)
        app.updates.add_listener(self)

        # Connect zoom functionality
        window.sliderZoom.valueChanged.connect(self.update_zoom)

        # Connect waveform generation signal
        window.WaveformReady.connect(self.Waveform_Ready)

        # Local audio waveform cache
        self.waveform_cache = {}

        # Connect update thumbnail signal
        window.ThumbnailUpdated.connect(self.Thumbnail_Updated)

        # Copy clipboard
        self.copy_clipboard = {}
        self.copy_transition_clipboard = {}

        # Init New clip
        self.new_item = False
        self.item_type = None
        self.item_id = None

        # Delayed zoom audio redraw
        self.redraw_audio_timer = QTimer(self)
        self.redraw_audio_timer.setInterval(300)
        self.redraw_audio_timer.setSingleShot(True)
        self.redraw_audio_timer.timeout.connect(self.redraw_audio_onTimeout)

        # QTimer for cache rendering
        self.cache_renderer_version = None
        self.cache_renderer = QTimer(self)
        self.cache_renderer.setInterval(500)
        self.cache_renderer.timeout.connect(self.render_cache_json)

        # Connect shutdown signals
        app.aboutToQuit.connect(self.redraw_audio_timer.stop)
        app.aboutToQuit.connect(self.cache_renderer.stop)
        app.aboutToQuit.connect(self.deleteLater)

        # Delay the start of cache rendering
        QTimer.singleShot(1500, self.cache_renderer.start)