Functions — github.com/GVCLab/CutClaw

hub / github.com/GVCLab/CutClaw / functions

Functions377 in github.com/GVCLab/CutClaw

⨍ Functions 377◇ Types & classes 16

↓ 29 callersFunction_cfg

Get a config value as a plain string (strips quotes).

app.py:31

↓ 23 callersFunctionseconds_to_mmss

Convert seconds to MM:SS.f format (with one decimal place). Args: seconds: Time in seconds Returns: Time string in MM:S

src/audio/audio_caption_madmom.py:116

↓ 21 callersMethodrun

Run the ReAct-style loop with OpenAI Function Calling. Args: shot_plan_path: Path to a pre-generated shot_plan.json file

src/core.py:1381

↓ 18 callersFunctionsave_config

Overwrite a single key in config.py.

app.py:41

↓ 16 callersFunctionmmss_to_seconds

Convert MM:SS or MM:SS.f format to seconds. Args: mmss: Time string in MM:SS or MM:SS.f format Returns: Time in seconds

src/audio/audio_caption_madmom.py:279

↓ 16 callersFunctionparse_time_to_seconds

Convert time string to seconds, supports formats: '01:23.45' or '83.45

src/audio/short_music.py:14

↓ 14 callersFunctionhhmmss_to_seconds

Convert time string to seconds. Supported formats: - HH:MM:SS or HH:MM:SS.mmm (standard) - HH:MM:SS:FF (with frame number,

src/utils/media_utils.py:54

↓ 13 callersFunction_to_sec

(t)

src/Screenwriter_scene_short.py:242

↓ 10 callersFunction_seconds_to_srt_time

(seconds: float)

src/Screenwriter_scene_short.py:576

↓ 10 callersFunctionformat_table

(data_list: List[Dict], type_key='type')

src/audio/interactive/visualization.py:454

↓ 10 callersFunctionhhmmss_to_seconds

Convert HH:MM:SS, HH:MM:SS.s, HH:MM:SS:FF, or MM:SS to seconds.

src/utils/time_format_convert.py:9

↓ 9 callersMethod_append_tool_msg

(self, tool_call_id, name, content, msgs)

src/core.py:1125

↓ 8 callersFunctionescape_drawtext_path

Escape file path for ffmpeg drawtext arguments.

render/render_video.py:118

↓ 8 callersFunctionhhmmss_to_seconds

(time_str: str)

src/core.py:143

↓ 8 callersFunctionnode

(stage, label, state_override=None)

app.py:588

↓ 8 callersFunctionsrt_time_to_seconds

Convert SRT-style HH:MM:SS,mmm to seconds.

render/render_video.py:34

↓ 7 callersFunction_unique_keypoint_types

(keypoints: List[Dict])

src/audio/interactive/visualization.py:505

↓ 7 callersFunctionseconds_to_hhmmss

Convert seconds to 'HH:MM:SS.s' string (one decimal place).

src/utils/media_utils.py:45

↓ 6 callersFunction_filter_keypoints_by_types

(keypoints: List[Dict], selected_types: List[str])

src/audio/interactive/visualization.py:516

↓ 6 callersFunction_to_audio_seconds

Normalize section timestamps into seconds.

src/Screenwriter_scene_short.py:118

↓ 6 callersFunctioncreate_full_width_player

Generates the HTML for the full-width player. CRITICAL FIX: Uses requestAnimationFrame for smooth playback visualization. Args:

src/audio/interactive/visualization.py:207

↓ 6 callersFunctionparse_time_str

解析时间字符串为秒数支持格式: "MM:SS" 或 "HH:MM:SS" 或直接数字

src/audio/audio_Madmom.py:861

↓ 6 callersFunctionseconds_to_time_str

Convert seconds to time string format MM:SS.xx

src/audio/short_music.py:27

↓ 5 callersMethod_result_ranges

(self, result_data: dict)

src/core.py:1579

↓ 5 callersFunction_seconds_to_mmss

Convert seconds to MM:SS.f format (e.g. 90.5 → '1:30.5').

src/Screenwriter_scene_short.py:125

↓ 4 callersFunction_call_agent_litellm

Call the agent LLM via litellm. Returns content string or None on failure.

src/Screenwriter_scene_short.py:84

↓ 4 callersFunction_create_decord_reader

( video_path: str, target_resolution: Optional[Tuple[int, int]] = None, )

src/video/preprocess/video_utils.py:60

↓ 4 callersFunction_make_reader

Try to open VideoReader; fall back to CPU if decord lacks CUDA support.

src/video/preprocess/video_utils.py:66

↓ 4 callersFunction_try_select

(window_mode: str, attempt_side: str, start_index: int | None = None)

src/Screenwriter_scene_short.py:823

↓ 4 callersFunction_validate

(data)

src/utils/media_utils.py:309

↓ 4 callersFunctionarray_to_base64

Encode a numpy uint8 RGB array (H, W, C) to a base64 JPEG string.

src/utils/media_utils.py:189

↓ 4 callersFunctionceil_by_factor

Returns the smallest integer greater than or equal to 'number' that is divisible by 'factor'.

src/utils/video_tools.py:50

↓ 4 callersFunctionextract_json_from_text

Extract and parse JSON from text that may contain additional content. Args: text: Text that may contain JSON Returns: P

src/audio/audio_caption_madmom.py:134

↓ 4 callersFunctionfilter_significant_keypoints

过滤关键点，只保留显著的点 Args: keypoints: 原始关键点列表 min_interval: 最小间隔（秒），间隔内只保留最强的点 top_k: 只保留强度最高的前K个点，0表示不限制 energy_pe

src/audio/audio_Madmom.py:756

↓ 4 callersFunctionfloor_by_factor

Returns the largest integer less than or equal to 'number' that is divisible by 'factor'.

src/utils/video_tools.py:55

↓ 4 callersFunctionformat_srt_timestamp

Convert milliseconds to SRT timestamp HH:MM:SS,mmm.

src/utils/time_format_convert.py:34

↓ 4 callersFunctiongenerate_structure_proposal

Generate a structure proposal for the video editing based on scene summaries.

src/Screenwriter_scene_short.py:343

↓ 4 callersFunctionget_audio_data

Load audio data with caching

src/audio/interactive/utils.py:56

↓ 4 callersFunctionget_section_duration

Calculate section duration in seconds

src/audio/short_music.py:34

↓ 4 callersFunctionload_audio_no_librosa

Load audio file without using librosa. Args: path_or_buffer: File path, URL, or buffer sr: Target sample rate (default:

src/audio/audio_utils.py:20

↓ 4 callersFunctionsmart_resize

Rescales the image so that the following conditions are met: 1. Both dimensions (height and width) are divisible by 'factor'. 2. The to

src/utils/video_tools.py:60

↓ 4 callersFunctiontype_matches

检查关键点类型是否匹配优先类型列表

src/audio/audio_Madmom.py:949

↓ 3 callersFunction_clear_thread_video_reader

()

src/core.py:65

↓ 3 callersMethod_compute_frame_indices

Compute frame indices for a time range using native fps with a hard cap.

src/Reviewer.py:266

↓ 3 callersMethod_get_cache_filename

Generate cache filename from cache key

src/audio/interactive/structure.py:24

↓ 3 callersFunction_get_setting

Read setting from src/config.py first, then fallback to environment/.env.

src/audio/litellm_client.py:31

↓ 3 callersFunction_get_thread_video_reader

(video_path: str)

src/Reviewer.py:28

↓ 3 callersFunction_has_meaningful_value

Check whether a JSON field is present with usable content.

src/Screenwriter_scene_short.py:31

↓ 3 callersMethod_merge_results

(self, existing_list: list, new_results: dict)

src/core.py:1672

↓ 3 callersFunction_ranges_overlap

(a_start: float, a_end: float, b_start: float, b_end: float)

src/core.py:104

↓ 3 callersFunction_scan_files

(folder: str, exts: set)

app.py:208

↓ 3 callersFunction_subtitle_line_text

(sub: dict)

src/Screenwriter_scene_short.py:585

↓ 3 callersFunctionarrow

()

app.py:600

↓ 3 callersFunctioncalculate_section_intensity

Calculate total intensity for a section by summing all intensity values Returns: { 'total_intensity': sum of all intensity values,

src/audio/short_music.py:50

↓ 3 callersMethoddetect

(self, audio_path: str, **kwargs)

src/audio/interactive/detectors.py:44

↓ 3 callersFunctiondetect_keypoints_madmom

Run Madmom keypoint detection and return a result dict. Returns a dict compatible with the interactive UI detector outputs. Core params expo

src/audio/madmom_api.py:87

↓ 3 callersFunctionescape_ffmpeg_expr

Escape commas in ffmpeg expressions to avoid filtergraph parsing issues.

render/render_video.py:126

↓ 3 callersFunctionfilter_by_sections

基于音乐段落（sections）进行关键点过滤（按比例分配模式）根据每个段落的关键点密度，按比例分配总镜头数 Args: keypoints: 原始关键点列表 sections: 段落列表，每个包含 name, start_time, e

src/audio/audio_Madmom.py:1082

↓ 3 callersFunctionfmt_time

(t)

src/audio/interactive/logic.py:339

↓ 3 callersFunctionhhmmss_to_seconds

Convert HH:MM:SS.s or MM:SS.s to seconds.

render/render_video.py:19

↓ 3 callersFunctionload_shots

读取并排序所有 Shot JSON

src/video/deconstruction/scene_merge.py:484

↓ 3 callersFunctionnms_1d

一维非极大值抑制 Args: values: 检测值（用于阈值过滤，如confidence） timestamps: 对应的时间戳（秒） min_distance: 最小间隔（秒），相邻点之间至少间隔这么多秒 thr

src/audio/audio_Madmom.py:132

↓ 3 callersFunctionoutput_path

(ratio: str)

app.py:836

↓ 3 callersFunctionparse_srt_to_dict

Parse an SRT file and return a mapping '{startSec}_{endSec}' -> 'subtitle text'. Timestamps are truncated to integer seconds.

src/utils/media_utils.py:125

↓ 3 callersFunctionround_by_factor

Returns the closest integer to 'number' that is divisible by 'factor'.

src/utils/video_tools.py:45

↓ 3 callersFunctionsave_scenes

保存 Scene JSON，并计算聚合时间

src/video/deconstruction/scene_merge.py:505

↓ 3 callersMethodsegment

(self, shots, threshold=0.5, max_scene_duration_secs=300)

src/video/deconstruction/scene_merge.py:419

↓ 3 callersFunctionsmart_nframes

calculate the number of frames for video used for model inputs. Args: ele (dict): a dict contains the configuration of video.

src/utils/video_tools.py:149

↓ 2 callersFunction_append_clip_spec

(timestamp_key, sampled_start, sampled_end, transcript_start_sec, transcript_end_sec, shot_id, is_sub_clip, su

src/video/deconstruction/video_caption.py:142

↓ 2 callersFunction_apply_section

(idx: int)

src/Screenwriter_scene_short.py:178

↓ 2 callersMethod_build_audio_section_info

(self, audio_section, shot_idx)

src/core.py:895

↓ 2 callersMethod_build_content

构建 VLM 输入

src/video/deconstruction/scene_analysis_video.py:146

↓ 2 callersMethod_call_video_analysis_model

Call VIDEO_ANALYSIS_MODEL using the same settings as fine_grained_shot_trimming.

src/Reviewer.py:285

↓ 2 callersFunction_caption_one

Send one acompletion request with timeout, return (clip, meta, content_or_None).

src/video/deconstruction/video_caption.py:318

↓ 2 callersMethod_compute_quality_score

(self, result_data: dict)

src/core.py:1565

↓ 2 callersFunction_derive_shot_duration_bounds

Derive segment duration bounds from the desired target shot length.

app.py:69

↓ 2 callersFunction_derive_target_shot_length_from_config

Reconstruct the target shot length from persisted min/max segment durations.

app.py:78

↓ 2 callersFunction_ensure_dir

(path: str)

src/video/preprocess/asr.py:17

↓ 2 callersFunction_ensure_dir

(path: str)

src/video/preprocess/video_utils.py:14

↓ 2 callersMethod_extract_long_shot_id

提取 long shot id：优先结构化字段，缺失时回退解析文件名。

src/video/deconstruction/scene_merge.py:121

↓ 2 callersMethod_get_soft_score

查表获取软相似度

src/video/deconstruction/scene_merge.py:73

↓ 2 callersFunction_get_speaker_at_time

Get the dominant speaker in a given time segment. Args: diarization_tracks: List of (segment, track, speaker) tuples start:

src/video/preprocess/asr.py:255

↓ 2 callersFunction_get_thread_video_reader

(video_path: str)

src/core.py:52

↓ 2 callersFunction_in_any

(t: float)

src/audio/madmom_api.py:342

↓ 2 callersMethod_is_same_long_shot

判断两个相邻片段是否来自同一个 long shot。

src/video/deconstruction/scene_merge.py:142

↓ 2 callersFunction_is_valid_scene_analysis_output

Check whether an existing scene analysis output is complete enough to skip.

src/video/deconstruction/scene_analysis_video.py:30

↓ 2 callersFunction_iter_clip_frames

Generator version of gather_clip_frames_from_long_shots. Yields clips one at a time as frames are read, enabling overlap with captioning.

src/video/deconstruction/video_caption.py:116

↓ 2 callersFunction_mmss_to_seconds

Convert MM:SS or numeric value to seconds.

src/audio/interactive/visualization.py:101

↓ 2 callersFunction_norm_name

(s: str)

src/Screenwriter_scene_short.py:1172

↓ 2 callersFunction_normalize_dialogue_text

Normalize dialogue text for robust subtitle matching.

src/Screenwriter_scene_short.py:593

↓ 2 callersFunction_normalize_video_reader

(video_reader)

src/core.py:72

↓ 2 callersFunction_parse_srt_content

Parse SRT format text into list of segments. Returns: [{"text": str, "start_s": float, "end_s": float}, ...]

src/video/preprocess/asr.py:21

↓ 2 callersMethod_prepare_shot_messages

(self, shot, audio_section_info, related_scene_value, guidance_text=None, forbidden_time_ranges=None)

src/core.py:916

↓ 2 callersFunction_run_overlapped

Producer-consumer: read frames synchronously, caption concurrently. decord VideoReader is not thread-safe, so get_batch is called dir

src/video/deconstruction/video_caption.py:346

↓ 2 callersFunction_run_pyannote_diarization

Run pyannote speaker diarization on audio file. Returns list of (segment, track, speaker) tuples.

src/video/preprocess/asr.py:224

↓ 2 callersFunction_run_scenedetect

Run scenedetect on [warmup_start_frame, end_frame), but only return scenes >= start_frame.

src/video/preprocess/video_utils.py:146

↓ 2 callersMethod_run_shot_loop

(self, msgs, max_iterations=None)

src/core.py:941

↓ 2 callersFunction_save_caption_result

Parse model response and save to disk. Returns error string or None on success.

src/video/deconstruction/video_caption.py:262

↓ 2 callersMethod_scene_duration_secs

计算单个 scene 的时长（秒）。

src/video/deconstruction/scene_merge.py:356

↓ 2 callersFunction_timecode_to_seconds

(timecode: str)

src/video/preprocess/video_utils.py:49

next →1–100 of 377, ranked by callers