1from __future__ import annotations
2 
3from dataclasses import dataclass, field
4from pathlib import Path
5from typing import Any
6 
7 
8@dataclass(frozen=True)
9class MediaInfo:
10    duration: float
11    has_audio: bool
12 
13 
14@dataclass(frozen=True)
15class CropBox:
16    x: float
17    y: float
18    width: float
19    height: float
20 
21 
22@dataclass(frozen=True)
23class FitSpec:
24    mode: str = "contain"
25    anchor: str = "center"
26    background: str | None = "black"
27 
28 
29@dataclass(frozen=True)
30class TransformSpec:
31    hflip: bool = False
32    vflip: bool = False
33    zoom: float = 1.0
34    video_pts: float = 1.0
35    audio_tempo: float = 1.0
36    eq: dict[str, float] = field(default_factory=dict)
37 
38 
39@dataclass(frozen=True)
40class OverlaySpec:
41    text: str
42    style: str = "default"
43    start: float | None = None
44    end: float | None = None
45    options: dict[str, Any] = field(default_factory=dict)
46 
47 
48@dataclass(frozen=True)
49class CaptionTrack:
50    transcript: str | Path
51    style: str = "karaoke"
52    mode: str = "karaoke"
53    trim: str = "word"
54    max_words_per_cue: int = 4
55    source_offset: float = 0.0
56    replacements: dict[str, str] = field(default_factory=dict)
57    skip_tokens: tuple[str, ...] = ()
58 
59 
60@dataclass(frozen=True)
61class Project:
62    width: int
63    height: int
64    fps: int
65    background: str
66    output: str | Path
67    work_dir: str | Path
68    open_after: bool = False
69    overlay_styles: dict[str, dict[str, Any]] = field(default_factory=dict)
70    caption_styles: dict[str, dict[str, Any]] = field(default_factory=dict)
71 
72 
73@dataclass(frozen=True)
74class Clip:
75    id: str
76    source: str | Path
77    label: str
78    src_in: float | None = None
79    src_out: float | None = None
80    trim_start: float = 0.0
81    trim_end: float = 0.0
82    speed: float = 1.0
83    fit: FitSpec = field(default_factory=FitSpec)
84    crop: CropBox | None = None
85    transform: TransformSpec = field(default_factory=TransformSpec)
86    overlays: tuple[OverlaySpec, ...] = ()
87    captions: CaptionTrack | None = None
88    muted: bool = False
89    volume: float = 1.0
90 
91 
92@dataclass(frozen=True)
93class WordToken:
94    text: str
95    start: float
96    end: float
97 
98 
99@dataclass(frozen=True)
100class ClipTiming:
101    nominal_in: float
102    nominal_out: float
103    actual_in: float
104    actual_out: float
105    video_time_scale: float
106    audio_time_scale: float
107    output_duration: float
108    word_aligned: bool = False
109 
110 
111@dataclass(frozen=True)
112class ClipAnalysis:
113    clip: Clip
114    source_path: Path
115    media: MediaInfo
116    timing: ClipTiming
117    caption_words: tuple[WordToken, ...] = ()
118 
119