cli.py - master - domahes/audio_visualizer_improved

cli.py
347 строк16 KB
Евгений Харченко
Optimize visualizer: implement AudioAnalysis and ImageFrames caching, and integrate model enums
02 июн 2026, 11:49
02 июн 2026, 11:49b7b0145
"""
CLI module for Audio Visualizer Pro.

Extracted in Phase 5: contains argument parser, main dispatch, _tk_runtime_available helper,
_expand_inputs (glob support), and the logic that either launches GUI (when no --audio and Tk available)
or runs headless render / batch / merge.

Re-exports / is imported by the thin audio_visualizer_improved.py shim so that
"python audio_visualizer_improved.py ..." and "from audio_visualizer_improved import main, parse_args" continue to work.

GUI launch imports AudioVisualizerApp lazily/conditionally from gui module.
"""

from __future__ import annotations

import os
import sys
import json
import argparse
import glob
import subprocess
from typing import List

# Flat-layout sibling import support (for direct python cli.py or during early refactor)
if __package__ in (None, ""):
    _here = os.path.dirname(os.path.abspath(__file__))
    if _here not in sys.path:
        sys.path.insert(0, _here)

from availability import (
    TK_AVAILABLE,
    tk,
    RenderCancelled,  # not directly used here but for consistency / future
)

from models import RenderSettings
from pipeline import render_video
from batch import render_batch
from primitives import STYLE_REGISTRY  # for --style choices in parser (str keys)
from models import (
    VisualStyle, ColorPalette, BackgroundMode, Quality,
    ClockType, ClockStyle,
)

# GUI class (conditional). We import it so main() can reference the name even if TK is absent (it will be None).
# When TK is present the class is the real one; otherwise the guard in main() + _tk_runtime_available prevents use.
try:
    from gui import AudioVisualizerApp
except Exception:
    AudioVisualizerApp = None  # type: ignore


def _tk_runtime_available() -> bool:
    if not TK_AVAILABLE:
        return False
    try:
        _r = tk.Tk()
        _r.withdraw()
        _r.update_idletasks()
        _r.destroy()
        return True
    except Exception:
        return False


def parse_args(argv=None):
    p = argparse.ArgumentParser(description="Audio Visualizer Pro — CLI/GUI hybrid")
    p.add_argument('--preset', default=None, help='Path to a preset JSON file to load settings from')
    p.add_argument('--audio', nargs='+', help='Path(s) to input audio file(s). Globs allowed (quoted).')
    p.add_argument('--out-dir', default='', help='Output directory (default: alongside first audio)')
    p.add_argument('--resolution', default='1280x720', help='WIDTHxHEIGHT, e.g. 1920x1080')
    p.add_argument('--fps', type=int, default=30)
    p.add_argument('--style', default=VisualStyle.BARS.value, choices=[s.value for s in VisualStyle])
    p.add_argument('--palette', default=ColorPalette.RAINBOW.value, choices=[p.value for p in ColorPalette])
    p.add_argument('--background', default=BackgroundMode.BLACK.value, choices=[b.value for b in BackgroundMode])
    p.add_argument('--quality', default=Quality.HIGH.value, choices=[q.value for q in Quality])
    p.add_argument('--glow', action='store_true')
    p.add_argument('--glow-intensity', type=int, default=3)
    p.add_argument('--blur', action='store_true')
    p.add_argument('--blur-amount', type=int, default=5)
    p.add_argument('--particles', action='store_true')
    p.add_argument('--particle-count', type=int, default=60)
    p.add_argument('--mirror', action='store_true')
    p.add_argument('--no-beat-react', dest='beat_react', action='store_false', help='Disable beat reaction')
    p.set_defaults(beat_react=True)
    # Logo image options
    p.add_argument('--logo-img', dest='logo_img_path', default=None)
    p.add_argument('--logo-pos', default='tr', choices=['tl','tr','bl','br'])
    p.add_argument('--logo-scale', type=float, default=1.0)
    p.add_argument('--logo-opacity', type=float, default=0.8)
    p.add_argument('--logo-start', type=float, default=0.0)
    p.add_argument('--logo-end', type=float, default=None)
    p.add_argument('--logo-follow-progress', action='store_true')

    p.add_argument('--logo-text', default=None)
    p.add_argument('--bitrate', default='192k')
    p.add_argument('--codec', default='h264', choices=['h264','h265','vp9'])
    p.add_argument('--draft-start', type=float, default=0.0, help='Draft render start (sec)')
    p.add_argument('--draft-duration', type=float, default=None, help='Draft render duration (sec)')
    # overlays
    p.add_argument('--progress-bar', action='store_true', help='Enable video progress bar overlay')
    p.add_argument('--progress-pos', default='bottom', choices=['top','bottom'])
    p.add_argument('--progress-height', type=int, default=10)
    p.add_argument('--progress-countdown', dest='clock_countdown', action='store_true',
                   help='Legacy alias: show reverse countdown next to the clock')
    p.add_argument('--clock-countdown', dest='clock_countdown', action='store_true',
                   help='Show reverse countdown next to the clock')
    p.add_argument('--no-clock-countdown', dest='clock_countdown', action='store_false',
                   help='Hide reverse countdown next to the clock')
    p.set_defaults(clock_countdown=True)
    p.add_argument('--clock', default=ClockType.DIGITAL.value, choices=[c.value for c in ClockType] + ['none'])
    p.add_argument('--clock-mode', default='elapsed', choices=['system','elapsed'])
    p.add_argument('--clock-style', default=ClockStyle.DARK.value, choices=[c.value for c in ClockStyle])
    p.add_argument('--clock-pos', default='tr', choices=['tl','tr','bl','br'])
    p.add_argument('--clock-scale', type=float, default=1.0)

    # batch
    p.add_argument('--merge', dest='merge', action='store_true', help='Merge multiple inputs into one output with same settings')
    p.add_argument('--no-merge', dest='merge', action='store_false', help='Render multiple inputs separately')
    p.set_defaults(merge=True)
    p.add_argument('--chapters', dest='chapters', action='store_true', help='Add chapters from file names on merged output')
    p.add_argument('--no-chapters', dest='chapters', action='store_false')
    p.set_defaults(chapters=True)
    p.add_argument('--gap-duration', type=float, default=0.0, help='Gap/slate duration between tracks (seconds). 0 disables')
    p.add_argument('--gap-slate', dest='gap_slate', action='store_true', help='Show a slate with next track title during the gap')
    p.add_argument('--no-gap-slate', dest='gap_slate', action='store_false')
    p.set_defaults(gap_slate=True)
    p.add_argument('--gap-label-format', default='Next: {title}', help='Label template for slate. Vars: {index} (1-based next track), {title}')
    p.add_argument('--workers', type=int, default=1, help='Parallel workers for per-file rendering')
    p.add_argument('--cleanup-intermediates', action='store_true', help='Delete per-track videos after merging')
    
    # Track title watermark
    p.add_argument('--track-title-watermark', dest='track_title_watermark', action='store_true', 
                   help='Show track title as watermark at start of each track')
    p.add_argument('--no-track-title-watermark', dest='track_title_watermark', action='store_false')
    p.set_defaults(track_title_watermark=True)
    p.add_argument('--track-title-duration', type=float, default=4.0, 
                   help='Duration (seconds) to show track title watermark')
    p.add_argument('--track-title-pos', default='cc',
                   choices=['tl', 'tc', 'tr', 'cl', 'cc', 'cr', 'bl', 'bc', 'br'],
                   help='Track title watermark position')
    p.add_argument('--track-title-offset-x', type=int, default=0,
                   help='Track title watermark X offset in pixels')
    p.add_argument('--track-title-offset-y', type=int, default=0,
                   help='Track title watermark Y offset in pixels')
    p.add_argument('--track-metadata-watermark', dest='track_metadata_watermark', action='store_true',
                   help='Show track metadata watermark (artist/album/genre/year)')
    p.add_argument('--no-track-metadata-watermark', dest='track_metadata_watermark', action='store_false')
    p.set_defaults(track_metadata_watermark=False)
    p.add_argument('--track-metadata-duration', type=float, default=4.0,
                   help='Track metadata watermark duration (sec). 0 or less = full track')
    p.add_argument('--track-metadata-pos', default='bl',
                   choices=['tl', 'tc', 'tr', 'cl', 'cc', 'cr', 'bl', 'bc', 'br'],
                   help='Track metadata watermark position')
    p.add_argument('--track-metadata-offset-x', type=int, default=18,
                   help='Track metadata watermark X offset in pixels')
    p.add_argument('--track-metadata-offset-y', type=int, default=18,
                   help='Track metadata watermark Y offset in pixels')

    # Font settings for Unicode/Emoji support
    p.add_argument('--font', dest='font_path', default=None, 
                   help='Path to TrueType font file (.ttf) for Unicode/Emoji support')
    p.add_argument('--font-size', type=int, default=32, 
                   help='Base font size in pixels (default: 32)')

    # Background image settings
    p.add_argument('--bg-image', dest='bg_image_path', default=None, 
                   help='Background image (PNG/GIF with alpha support)')
    p.add_argument('--bg-opacity', type=float, default=1.0, 
                   help='Background image opacity (0.0-1.0)')
    p.add_argument('--bg-pattern', action='store_true', 
                   help='Tile background image as pattern')

    # Foreground image settings
    p.add_argument('--fg-image', dest='fg_image_path', default=None, 
                   help='Foreground overlay image (PNG/GIF with alpha support)')
    p.add_argument('--fg-opacity', type=float, default=1.0, 
                   help='Foreground image opacity (0.0-1.0)')

    p.add_argument('--run-tests', action='store_true', help='Run unit tests and exit')
    return p.parse_args(argv)


def _expand_inputs(inputs: List[str]) -> List[str]:
    out: List[str] = []
    for item in inputs:
        if any(ch in item for ch in ['*', '?', '[']):
            out.extend(sorted(glob.glob(item)))
        else:
            out.append(item)
    out = [p for p in out if os.path.isfile(p)]
    if not out:
        raise FileNotFoundError("No matching input files found")
    return out


def main(argv=None):
    args = parse_args(argv)
    if args.preset:
        if not os.path.isfile(args.preset):
            print(f"Error: preset file not found: {args.preset}")
            sys.exit(2)
        try:
            with open(args.preset, 'r', encoding='utf-8') as f:
                preset_data = json.load(f)
            settings = preset_data.get("settings", preset_data)
            for k, v in settings.items():
                if k == 'clock_enable':
                    if not v:
                        args.clock = 'none'
                elif k == 'logo_follow':
                    args.logo_follow_progress = bool(v)
                elif k == 'bg_image_path':
                    args.bg_image_path = v
                elif k == 'bg_image_opacity':
                    args.bg_opacity = v
                elif k == 'bg_image_pattern':
                    args.bg_pattern = v
                elif k == 'fg_image_path':
                    args.fg_image_path = v
                elif k == 'fg_image_opacity':
                    args.fg_opacity = v
                elif hasattr(args, k):
                    default_val = getattr(args, k)
                    if default_val is not None:
                        try:
                            setattr(args, k, type(default_val)(v))
                        except Exception:
                            setattr(args, k, v)
                    else:
                        setattr(args, k, v)
        except Exception as e:
            print(f"Error loading preset JSON: {e}")
            sys.exit(2)

    if args.run_tests:
        import unittest
        unittest.main(module='audio_visualizer_improved', argv=[sys.argv[0]], exit=False)
        return

    # If no audio provided and Tk usable — launch GUI, else headless
    if (not args.audio) and _tk_runtime_available():
        if AudioVisualizerApp is None:
            print("Error: GUI requested but AudioVisualizerApp is not available (Tkinter missing at import time?)")
            sys.exit(2)
        root = tk.Tk()
        app = AudioVisualizerApp(root)
        root.mainloop()
        return

    if not args.audio:
        print("Error: --audio is required in headless mode (Tk not available or GUI cannot start)")
        print("Tip: run with --run-tests to verify environment, or install a desktop Tk (python3-tk) to use the GUI.")
        sys.exit(2)

    try:
        w, h = map(int, args.resolution.lower().split('x'))
    except Exception:
        print("Invalid --resolution. Use WIDTHxHEIGHT, e.g. 1280x720")
        sys.exit(2)

    clock_val = None if (args.clock == 'none') else args.clock
    inputs = _expand_inputs(args.audio)

    base_settings = RenderSettings(
        audio_path=inputs[0],
        out_dir=args.out_dir,
        resolution=(w, h),
        fps=int(args.fps),
        style=args.style,
        palette=args.palette,
        background=args.background,
        quality=args.quality,
        glow=bool(args.glow),
        glow_intensity=int(args.glow_intensity),
        blur=bool(args.blur),
        blur_amount=int(args.blur_amount),
        particles=bool(args.particles),
        particle_count=int(args.particle_count),
        mirror=bool(args.mirror),
        beat_react=bool(args.beat_react),
        logo_text=(args.logo_text if args.logo_text else None),
        logo_img_path=(args.logo_img_path if args.logo_img_path else None),
        logo_pos=args.logo_pos,
        logo_scale=float(args.logo_scale),
        logo_opacity=float(args.logo_opacity),
        logo_start=float(args.logo_start),
        logo_end=(float(args.logo_end) if args.logo_end is not None else None),
        logo_follow_progress=bool(args.logo_follow_progress),
        bitrate=args.bitrate,
        codec=args.codec,
        draft_start=float(args.draft_start),
        draft_duration=(float(args.draft_duration) if args.draft_duration is not None else None),
        progress_bar=bool(args.progress_bar),
        progress_pos=args.progress_pos,
        progress_height=int(args.progress_height),
        clock=clock_val,
        clock_mode=args.clock_mode,
        clock_style=args.clock_style,
        clock_pos=args.clock_pos,
        clock_scale=float(args.clock_scale),
        clock_countdown=bool(args.clock_countdown),
        clock_offset_x=int(getattr(args, 'clock_offset_x', 18)),
        clock_offset_y=int(getattr(args, 'clock_offset_y', 18)),
        # Track title watermark
        track_title_watermark=bool(args.track_title_watermark),
        track_title_duration=float(args.track_title_duration),
        track_title_pos=str(args.track_title_pos),
        track_title_offset_x=int(args.track_title_offset_x),
        track_title_offset_y=int(args.track_title_offset_y),
        track_metadata_watermark=bool(args.track_metadata_watermark),
        track_metadata_duration=float(args.track_metadata_duration),
        track_metadata_pos=str(args.track_metadata_pos),
        track_metadata_offset_x=int(args.track_metadata_offset_x),
        track_metadata_offset_y=int(args.track_metadata_offset_y),
        workers=int(args.workers),
        cleanup_intermediates=bool(args.cleanup_intermediates),
        # Font settings for Unicode/Emoji support
        font_path=args.font_path,
        font_size=int(args.font_size),
        # Background/Foreground image settings
        bg_image_path=(args.bg_image_path if args.bg_image_path else None),
        bg_image_opacity=float(getattr(args, 'bg_opacity', 1.0)),
        bg_image_pattern=bool(getattr(args, 'bg_pattern', False)),
        fg_image_path=(args.fg_image_path if args.fg_image_path else None),
        fg_image_opacity=float(getattr(args, 'fg_opacity', 1.0)),
    )

    try:
        if len(inputs) == 1:
            render_video(base_settings)[0]
        else:
            if not args.merge:
                outs = render_batch(inputs, base_settings, merge=False)
                for p in outs:
                    print(p)
            else:
                final = render_batch(inputs, base_settings, merge=True, add_chapters=bool(args.chapters),
                                     gap_duration=float(args.gap_duration), gap_slate=bool(args.gap_slate),
                                     gap_label_format=str(args.gap_label_format))
                print(final)
    except subprocess.CalledProcessError as e:
        print("ffmpeg failed:", e)
        sys.exit(1)
    except Exception as e:
        print(f"Render failed: {e}")
        sys.exit(1)