xiaoxia-saas/packages/domain/plan_generator_utils.py

"""剪辑计划生成 — 纯逻辑工具函数.

从 PlanGeneratorService 提取的纯业务逻辑：
- 素材分配策略（4 种 editing_mode）
- 默认 clip 结构生成
- clip_type 按模式映射
- 从 TemplateClipConfig 创建 EditPlanClip

纯函数，无副作用，不依赖 DB/外部服务。
"""

from __future__ import annotations

import random
from typing import List

from packages.domain.edit_plan_clip import EditPlanClip
from packages.domain.editing_mode import EditingMode
from packages.domain.template_clip_config import ClipType, TemplateClipConfig

# ── 默认片段时长（秒） ────────────────────────────────────────────────────────
DEFAULT_CLIP_DURATION = 5.0
DEFAULT_INTRO_DURATION = 3.0
DEFAULT_OUTRO_DURATION = 3.0


# ── 素材分配 ────────────────────────────────────────────────────────────────


def distribute_assets(
    clips: List[EditPlanClip],
    asset_ids: List[str],
    editing_mode: str,
    *,
    random_selection: bool = False,
    asset_durations: dict[str, float] | None = None,
) -> None:
    """按 editing_mode 将素材分配到 clips（就地修改）.

    分配策略:
    - ONE_TAKE: 素材按顺序依次分配给 main 类型 clips
    - PIP: 第1个素材→main(全屏背景)，其余→overlay clips
    - VOICE_OVER: 素材→main clips (B-roll)
    - VOICE_PIP: 第1个→background, 第2个→corner_voice, 其余→b_roll

    Args:
        clips: 剪辑片段列表（就地修改 asset_id）
        asset_ids: 素材 ID 列表
        editing_mode: 剪辑模式字符串
        random_selection: 是否随机选择素材（用于预览生成）
        asset_durations: 素材 ID -> 时长（秒）映射，用于设置随机 start_time
    """
    if not asset_ids or not clips:
        return

    # 如果需要随机选择，先打乱素材顺序
    if random_selection:
        asset_ids = list(asset_ids)  # 复制避免修改原列表
        random.shuffle(asset_ids)

    if editing_mode == EditingMode.ONE_TAKE.value:
        _distribute_one_take(clips, asset_ids, asset_durations)
    elif editing_mode == EditingMode.PIP.value:
        _distribute_pip(clips, asset_ids, asset_durations)
    elif editing_mode == EditingMode.VOICE_OVER.value:
        _distribute_voice_over(clips, asset_ids, asset_durations)
    elif editing_mode == EditingMode.VOICE_PIP.value:
        _distribute_voice_pip(clips, asset_ids, asset_durations)
    else:
        # 未知模式，退化为 one_take
        _distribute_one_take(clips, asset_ids, asset_durations)


def _distribute_one_take(
    clips: List[EditPlanClip],
    asset_ids: List[str],
    asset_durations: dict[str, float] | None = None,
) -> None:
    """ONE_TAKE: 素材按顺序依次分配给 main 类型 clips."""
    used_segments: dict[str, list[tuple[float, float]]] = {}
    main_clips = [c for c in clips if c.clip_type == ClipType.MAIN.value]
    for i, clip in enumerate(main_clips):
        if i < len(asset_ids):
            asset_id = asset_ids[i]
            start_time = _calc_random_start_time(asset_id, clip.duration, asset_durations, used_segments)
            clip.assign_asset(asset_id, start_time=start_time)
            # Record used segment
            if start_time is not None and asset_durations is not None:
                if asset_id not in used_segments:
                    used_segments[asset_id] = []
                used_segments[asset_id].append((start_time, start_time + clip.duration))


def _distribute_pip(
    clips: List[EditPlanClip],
    asset_ids: List[str],
    asset_durations: dict[str, float] | None = None,
) -> None:
    """PIP: 第1个素材→main(全屏背景)，其余→overlay clips."""
    used_segments: dict[str, list[tuple[float, float]]] = {}
    # 第1个素材 → main clip
    main_clips = [c for c in clips if c.clip_type == ClipType.MAIN.value]
    if main_clips and asset_ids:
        asset_id = asset_ids[0]
        start_time = _calc_random_start_time(asset_id, main_clips[0].duration, asset_durations, used_segments)
        main_clips[0].assign_asset(asset_id, start_time=start_time)
        # Record used segment
        if start_time is not None and asset_durations is not None:
            if asset_id not in used_segments:
                used_segments[asset_id] = []
            used_segments[asset_id].append((start_time, start_time + main_clips[0].duration))

    # 其余素材 → overlay clips
    overlay_clips = [c for c in clips if c.clip_type == "overlay"]
    remaining = asset_ids[1:]
    for i, clip in enumerate(overlay_clips):
        if i < len(remaining):
            asset_id = remaining[i]
            start_time = _calc_random_start_time(asset_id, clip.duration, asset_durations, used_segments)
            clip.assign_asset(asset_id, start_time=start_time)
            # Record used segment
            if start_time is not None and asset_durations is not None:
                if asset_id not in used_segments:
                    used_segments[asset_id] = []
                used_segments[asset_id].append((start_time, start_time + clip.duration))


def _distribute_voice_over(
    clips: List[EditPlanClip],
    asset_ids: List[str],
    asset_durations: dict[str, float] | None = None,
) -> None:
    """VOICE_OVER: 素材→main clips (B-roll)."""
    used_segments: dict[str, list[tuple[float, float]]] = {}
    main_clips = [c for c in clips if c.clip_type == ClipType.MAIN.value]
    for i, clip in enumerate(main_clips):
        if i < len(asset_ids):
            asset_id = asset_ids[i]
            start_time = _calc_random_start_time(asset_id, clip.duration, asset_durations, used_segments)
            clip.assign_asset(asset_id, start_time=start_time)
            # Record used segment
            if start_time is not None and asset_durations is not None:
                if asset_id not in used_segments:
                    used_segments[asset_id] = []
                used_segments[asset_id].append((start_time, start_time + clip.duration))


def _distribute_voice_pip(
    clips: List[EditPlanClip],
    asset_ids: List[str],
    asset_durations: dict[str, float] | None = None,
) -> None:
    """VOICE_PIP: 第1个→background, 第2个→corner_voice, 其余→b_roll."""
    used_segments: dict[str, list[tuple[float, float]]] = {}
    bg_clips = [c for c in clips if c.clip_type == "background"]
    voice_clips = [c for c in clips if c.clip_type == "corner_voice"]
    broll_clips = [c for c in clips if c.clip_type == "b_roll"]

    idx = 0

    # 第1个 → background
    if idx < len(asset_ids) and bg_clips:
        asset_id = asset_ids[idx]
        start_time = _calc_random_start_time(asset_id, bg_clips[0].duration, asset_durations, used_segments)
        bg_clips[0].assign_asset(asset_id, start_time=start_time)
        # Record used segment
        if start_time is not None and asset_durations is not None:
            if asset_id not in used_segments:
                used_segments[asset_id] = []
            used_segments[asset_id].append((start_time, start_time + bg_clips[0].duration))
        idx += 1

    # 第2个 → corner_voice
    if idx < len(asset_ids) and voice_clips:
        asset_id = asset_ids[idx]
        start_time = _calc_random_start_time(asset_id, voice_clips[0].duration, asset_durations, used_segments)
        voice_clips[0].assign_asset(asset_id, start_time=start_time)
        # Record used segment
        if start_time is not None and asset_durations is not None:
            if asset_id not in used_segments:
                used_segments[asset_id] = []
            used_segments[asset_id].append((start_time, start_time + voice_clips[0].duration))
        idx += 1

    # 剩余 → b_roll clips
    remaining = asset_ids[idx:]
    for i, clip in enumerate(broll_clips):
        if i < len(remaining):
            asset_id = remaining[i]
            start_time = _calc_random_start_time(asset_id, clip.duration, asset_durations, used_segments)
            clip.assign_asset(asset_id, start_time=start_time)
            # Record used segment
            if start_time is not None and asset_durations is not None:
                if asset_id not in used_segments:
                    used_segments[asset_id] = []
                used_segments[asset_id].append((start_time, start_time + clip.duration))


# ── 随机 start_time 计算 ────────────────────────────────────────────────────


def _calc_random_start_time(
    asset_id: str,
    clip_duration: float,
    asset_durations: dict[str, float] | None,
    used_segments: dict[str, list[tuple[float, float]]] | None = None,
) -> float | None:
    """计算随机 start_time，避开已使用的时间段.

    在素材总时长范围内随机取点，确保 clip_duration 不超出素材边界。
    如果 asset_durations 为 None 或素材不在其中，返回 None（使用默认 0.0）。
    如果提供了 used_segments，会避开已使用的时间区间。

    Args:
        asset_id: 素材 ID
        clip_duration: 片段时长（秒）
        asset_durations: 素材 ID -> 时长映射
        used_segments: {asset_id: [(start1, end1), (start2, end2), ...]} 已使用的时间段

    Returns:
        随机 start_time 或 None
    """
    if asset_durations is None:
        return None

    total_duration = asset_durations.get(asset_id)
    if total_duration is None or total_duration <= 0:
        return None

    # 最大起始点 = 素材总时长 - 片段时长
    max_start = max(0.0, total_duration - clip_duration)
    if max_start <= 0:
        return 0.0

    # 如果没有已使用段，直接随机
    if not used_segments or asset_id not in used_segments:
        return random.uniform(0.0, max_start)

    # 尝试找到一个不与已使用段重叠的起始点
    used = sorted(used_segments[asset_id])
    max_attempts = 100

    for _ in range(max_attempts):
        candidate = random.uniform(0.0, max_start)
        candidate_end = candidate + clip_duration

        # 检查是否与任何已使用段重叠
        overlap = False
        for seg_start, seg_end in used:
            # 两个区间 [a, b] 和 [c, d] 重叠的条件: a < d and c < b
            if candidate < seg_end and seg_start < candidate_end:
                overlap = True
                break

        if not overlap:
            return candidate

    # 如果尝试多次仍找不到，缩短时长使用素材末尾
    # 找到最后一个已使用段之后的可用空间
    last_used_end = 0.0
    for _seg_start, seg_end in used:
        last_used_end = max(last_used_end, seg_end)

    if last_used_end < total_duration:
        # 返回从最后使用点开始的位置
        return min(last_used_end, max_start)

    # 实在没有空间，返回0（可能会重叠，但至少能执行）
    return 0.0


# ── clip_type 映射 ────────────────────────────────────────────────────────


def map_clip_types_for_mode(
    clips: List[EditPlanClip],
    editing_mode: str,
) -> None:
    """将 MAIN 类型片段按 editing_mode 映射为对应角色类型.

    模板的 clip_config 使用 ClipType 枚举（main/intro/outro 等），
    但 PIP / VOICE_PIP 模式的素材分配和渲染分层依赖特定的 clip_type 命名
    （overlay / background / corner_voice / b_roll）。

    映射规则（仅修改 MAIN 类型片段，非 MAIN 片段保持原类型）：
    - PIP: 第1个 MAIN → main（背景），其余 MAIN → overlay（画中画）
    - VOICE_PIP: 第1个 → background，第2个 → corner_voice，第3+个 → b_roll
    - ONE_TAKE / VOICE_OVER: 保持 main 不变

    Args:
        clips: 剪辑片段列表（就地修改 clip_type）
        editing_mode: 剪辑模式字符串
    """
    main_clips = [c for c in clips if c.clip_type == ClipType.MAIN.value]
    if not main_clips:
        return

    if editing_mode == EditingMode.PIP.value:
        # 第1个 main 保持（背景层），其余改为 overlay（画中画层）
        for i, clip in enumerate(main_clips):
            if i > 0:
                clip.clip_type = "overlay"

    elif editing_mode == EditingMode.VOICE_PIP.value:
        for i, clip in enumerate(main_clips):
            if i == 0:
                clip.clip_type = "background"
            elif i == 1:
                clip.clip_type = "corner_voice"
            else:
                clip.clip_type = "b_roll"

    # ONE_TAKE / VOICE_OVER: 保持 main 不变，无需处理


# ── 默认 clip 生成 ────────────────────────────────────────────────────────


def generate_default_clips(
    plan_id: str,
    editing_mode: str,
    asset_count: int,
) -> List[EditPlanClip]:
    """无 clip_configs 时，根据 editing_mode 生成默认 clip 结构.

    - ONE_TAKE: N 个 main clips（N = asset_count，至少1个）
    - PIP: 1 个 main + (N-1) 个 overlay（N = asset_count）
    - VOICE_OVER: N 个 main clips + 标记需要配音
    - VOICE_PIP: 1 个 background + 1 个 corner_voice + (N-2) 个 b_roll

    Args:
        plan_id: 剪辑计划 ID
        editing_mode: 剪辑模式字符串
        asset_count: 素材数量

    Returns:
        List[EditPlanClip]: 生成的默认剪辑片段列表
    """
    n = max(asset_count, 1)
    clips: List[EditPlanClip] = []
    order = 0

    if editing_mode == EditingMode.PIP.value:
        # 1 个 main（全屏背景）
        clips.append(
            EditPlanClip.create(
                plan_id=plan_id,
                clip_type=ClipType.MAIN.value,
                order=order,
                duration=DEFAULT_CLIP_DURATION,
            )
        )
        order += 1
        # 剩余为 overlay
        for _ in range(1, n):
            clips.append(
                EditPlanClip.create(
                    plan_id=plan_id,
                    clip_type="overlay",
                    order=order,
                    duration=DEFAULT_CLIP_DURATION,
                )
            )
            order += 1

    elif editing_mode == EditingMode.VOICE_OVER.value:
        # N 个 main clips（B-roll）
        for _ in range(n):
            clips.append(
                EditPlanClip.create(
                    plan_id=plan_id,
                    clip_type=ClipType.MAIN.value,
                    order=order,
                    duration=DEFAULT_CLIP_DURATION,
                    config={"role": "b_roll"},
                )
            )
            order += 1

    elif editing_mode == EditingMode.VOICE_PIP.value:
        # 1 个 background
        clips.append(
            EditPlanClip.create(
                plan_id=plan_id,
                clip_type="background",
                order=order,
                duration=DEFAULT_CLIP_DURATION,
            )
        )
        order += 1
        # 1 个 corner_voice（至少有1个素材就有）
        if n >= 2:
            clips.append(
                EditPlanClip.create(
                    plan_id=plan_id,
                    clip_type="corner_voice",
                    order=order,
                    duration=DEFAULT_CLIP_DURATION,
                )
            )
            order += 1
        # 剩余为 b_roll
        for _ in range(2, n):
            clips.append(
                EditPlanClip.create(
                    plan_id=plan_id,
                    clip_type="b_roll",
                    order=order,
                    duration=DEFAULT_CLIP_DURATION,
                )
            )
            order += 1

    else:
        # ONE_TAKE / 未知模式：N 个 main clips
        for _ in range(n):
            clips.append(
                EditPlanClip.create(
                    plan_id=plan_id,
                    clip_type=ClipType.MAIN.value,
                    order=order,
                    duration=DEFAULT_CLIP_DURATION,
                )
            )
            order += 1

    return clips


# ── 从配置创建 clips ────────────────────────────────────────────────────────


def create_clips_from_configs(
    plan_id: str,
    clip_configs: List[TemplateClipConfig],
) -> List[EditPlanClip]:
    """从 TemplateClipConfig 列表创建 EditPlanClip 列表.

    Args:
        plan_id: 剪辑计划 ID
        clip_configs: 模板片段配置列表

    Returns:
        List[EditPlanClip]: 创建的剪辑片段列表（按 order 排序）
    """
    clips: List[EditPlanClip] = []
    # 按 order 排序
    sorted_configs = sorted(clip_configs, key=lambda c: c.order)

    for cfg in sorted_configs:
        # 计算时长：取 min_duration 和 max_duration 的中间值
        if cfg.min_duration > 0 and cfg.max_duration > 0:
            duration = (cfg.min_duration + cfg.max_duration) / 2
        elif cfg.min_duration > 0:
            duration = cfg.min_duration
        elif cfg.max_duration > 0:
            duration = cfg.max_duration
        else:
            duration = DEFAULT_CLIP_DURATION

        # clip_type 可能是枚举或字符串
        clip_type = cfg.clip_type.value if hasattr(cfg.clip_type, "value") else cfg.clip_type

        # transition_effect 可能是枚举或字符串
        transition = cfg.transition_effect.value if hasattr(cfg.transition_effect, "value") else cfg.transition_effect

        # 从 clip config 中解析 playback_speed（兼容 speed_ratio 字段名）
        clip_cfg = cfg.config or {}
        playback_speed = clip_cfg.get("playback_speed", clip_cfg.get("speed_ratio", 1.0)) or 1.0

        clip = EditPlanClip.create(
            plan_id=plan_id,
            clip_type=clip_type,
            order=cfg.order,
            template_clip_config_id=cfg.id,
            text_content=getattr(cfg, "text_template", "") or "",
            duration=duration,
            transition_effect=transition or "cut",
            playback_speed=playback_speed,
            config=clip_cfg,
        )
        clips.append(clip)

    return clips