shortGPT/editing_framework/README.md

Editing Framework Module Documentation

The editing_framework module provides a set of classes and functions for editing videos and images. This module is part of the shortGPT project and is designed to be used with the CoreEditingEngine class to generate videos and images based on a specified editing schema.

Module Files

The editing_framework module consists of three files:

1. rendering_logger.py: This file contains the MoviepyProgressLogger class, which is used for logging the progress of the rendering process.
2. editing_engine.py: This file contains the EditingStep and Flow enums, as well as the EditingEngine class, which is the main class for managing the editing process.
3. core_editing_engine.py: This file contains the CoreEditingEngine class, which is responsible for generating videos and images based on the editing schema.

rendering_logger.py

This file defines the MoviepyProgressLogger class, which is a subclass of ProgressBarLogger from the proglog module. It provides a callback function for logging the progress of the rendering process. The MoviepyProgressLogger class has the following methods:

__init__(self, callBackFunction=None)

• Initializes a new instance of the MoviepyProgressLogger class.
• Parameters:
  • callBackFunction: An optional callback function that will be called with the progress string.

bars_callback(self, bar, attr, value, old_value=None)

• This method is called every time the logger progress is updated.
• It calculates the rendering progress and the estimated time left.
• It calls the callback function with the progress string or prints the progress string if no callback function is provided.
• Parameters:
  • bar: The progress bar name.
  • attr: The progress attribute name.
  • value: The current progress value.
  • old_value: The previous progress value.

format_time(self, seconds)

• Formats the given time in seconds to the format "mm:ss".
• Parameters:
  • seconds: The time in seconds.
• Returns:
  • The formatted time string.

editing_engine.py

This file defines the EditingStep and Flow enums, as well as the EditingEngine class, which is responsible for managing the editing process. The EditingEngine class has the following methods:

__init__(self)

• Initializes a new instance of the EditingEngine class.
• It initializes the editing step tracker and the editing schema.

addEditingStep(self, editingStep: EditingStep, args: Dict[str, any] = {})

• Adds an editing step to the editing schema with the specified arguments.
• Parameters:
  • editingStep: The editing step to add.
  • args: The arguments for the editing step.
• Raises:
  • Exception: If a required argument is missing.

ingestFlow(self, flow: Flow, args)

• Ingests a flow into the editing schema with the specified arguments.
• Parameters:
  • flow: The flow to ingest.
  • args: The arguments for the flow.
• Raises:
  • Exception: If a required argument is missing.

dumpEditingSchema(self)

• Returns the current editing schema.

renderVideo(self, outputPath, logger=None)

• Renders the video based on the editing schema and saves it to the specified output path.
• Parameters:
  • outputPath: The path to save the rendered video.
  • logger: An optional logger object for logging the rendering progress.

renderImage(self, outputPath)

• Renders the image based on the editing schema and saves it to the specified output path.
• Parameters:
  • outputPath: The path to save the rendered image.

core_editing_engine.py

This file defines the CoreEditingEngine class, which is responsible for generating videos and images based on the editing schema. The CoreEditingEngine class has the following methods:

generate_image(self, schema:Dict[str, Any], output_file)

• Generates an image based on the editing schema and saves it to the specified output file.
• Parameters:
  • schema: The editing schema.
  • output_file: The path to save the generated image.
• Returns:
  • The path to the saved image.

generate_video(self, schema:Dict[str, Any], output_file, logger=None)

• Generates a video based on the editing schema and saves it to the specified output file.
• Parameters:
  • schema: The editing schema.
  • output_file: The path to save the generated video.
  • logger: An optional logger object for logging the rendering progress.
• Returns:
  • The path to the saved video.

process_common_actions(self, clip: Union[VideoFileClip, ImageClip, TextClip, AudioFileClip], actions: List[Dict[str, Any]])

• Processes common actions for the given clip.
• Parameters:
  • clip: The clip to process.
  • actions: The list of actions to apply to the clip.
• Returns:
  • The processed clip.

process_common_visual_actions(self, clip: Union[VideoFileClip, ImageClip, TextClip], actions: List[Dict[str, Any]])

• Processes common visual clip actions for the given clip.
• Parameters:
  • clip: The clip to process.
  • actions: The list of actions to apply to the clip.
• Returns:
  • The processed clip.

process_audio_actions(self, clip: AudioFileClip, actions: List[Dict[str, Any]])

• Processes audio actions for the given audio clip.
• Parameters:
  • clip: The audio clip to process.
  • actions: The list of actions to apply to the audio clip.
• Returns:
  • The processed audio clip.

process_video_asset(self, asset: Dict[str, Any])

• Processes a video asset based on the asset parameters and actions.
• Parameters:
  • asset: The video asset to process.
• Returns:
  • The processed video clip.

process_image_asset(self, asset: Dict[str, Any])

• Processes an image asset based on the asset parameters and actions.
• Parameters:
  • asset: The image asset to process.
• Returns:
  • The processed image clip.

process_text_asset(self, asset: Dict[str, Any])

• Processes a text asset based on the asset parameters and actions.
• Parameters:
  • asset: The text asset to process.
• Returns:
  • The processed text clip.

process_audio_asset(self, asset: Dict[str, Any])

• Processes an audio asset based on the asset parameters and actions.
• Parameters:
  • asset: The audio asset to process.
• Returns:
  • The processed audio clip.

__normalize_image(self, clip)

• Normalizes the image clip.
• Parameters:
  • clip: The image clip to normalize.
• Returns:
  • The normalized image clip.

__normalize_frame(self, frame)

• Normalizes the given frame.
• Parameters:
  • frame: The frame to normalize.
• Returns:
  • The normalized frame.