ffmpeg-skill 3.1.0 → 3.1.2

package/scripts/graphics.py

@@ -32,7 +32,7 @@ import sys
 from typing import List, Optional, Tuple
 
 from _platforms import PLATFORMS, PLATFORM_CHOICES, safe_margins_px, resolve as resolve_platform
-from _common import (aac_args, add_common, brand_caption_style, script_font_for_text, apply_common, cfr_args,
+from _common import (require_filter, aac_args, add_common, brand_caption_style, script_font_for_text, apply_common, cfr_args,
                      color_hex, default_output, die, draws_chroma, drawn_chain, emit, escape_drawtext, escape_filter_path,
                      ffmpeg_base, info, label_chain, load_brand, parse_time, probe, rgb_chain_to_bt709, run, run_keeping_subtitles, video_args,
                      drawtext_boxborderw, drawtext_center_align, X264_PRESETS, time_arg, fmt_secs, STATE, drawtext_text_opts,
@@ -237,13 +237,18 @@ def main(force_route: Optional[str] = None) -> int:
     emoji_mode = None
     emoji_overlays: List[dict] = []
     if has_emoji(all_text):
-        support = emoji_support(emoji_assets, probe=True)
+        _ff = args.font_file or brand.get("font_file")
+        support = emoji_support(emoji_assets, probe=True, fonts_dir=os.path.dirname(os.path.abspath(_ff)) if _ff else None)
         emoji_mode = support["mode"] if args.emoji == "auto" else args.emoji
         if args.emoji == "color" and not support["libass_color"]:
             die("--emoji color: this ffmpeg renders emoji monochrome through libass "
                 f"({support['detail']}) -- pass --emoji-assets DIR for colour, or --emoji mono", kind="input")
         if args.emoji == "png" and not emoji_assets:
             die("--emoji png: no emoji assets directory resolved -- " + EMOJI_ASSET_HINT, kind="input")
+        if args.emoji == "mono" and support.get("libass_glyph") is False:
+            die("--emoji mono: this ffmpeg draws no emoji glyph through libass "
+                f"({support['detail']}), so they would come out as empty boxes -- pass --emoji-assets DIR "
+                "for colour, or --emoji none", kind="input")
     if args.text_render == "drawtext" and shaping:
         die(f"{LANGUAGE_NAMES.get(_script, _script)} text cannot be shaped by drawtext on any ffmpeg "
             "build (the marks are reordered by harfbuzz, which drawtext does not use): drop "
@@ -642,6 +647,7 @@ def main(force_route: Optional[str] = None) -> int:
                             "assets": emoji_assets, "missing": missing,
                             "overlays": len(emoji_overlays)}
         ass_path = args.write_ass or os.path.splitext(output)[0] + "_gfx.ass"
+        require_filter("ass", "drawing this text through libass")  # before the sidecar is written
         if STATE.dry_run:
             info(f"[dry-run] would write {ass_path} ({len(elements)} text elements)")
         else:
@@ -682,7 +688,7 @@ def main(force_route: Optional[str] = None) -> int:
         chains = list(fc) if fc else [f"[0:v]{text or 'null'}[vout]"]
         last = "vout"
         if route == "ass":
-            vf = f"ass={escape_filter_path(ass_path)}"
+            vf = f"ass=filename={escape_filter_path(ass_path)}"
             fdir = ass_fonts_dir()
             if fdir:
                 vf += f":fontsdir={escape_filter_path(fdir)}"

package/scripts/look.py

@@ -19,6 +19,8 @@ from typing import List, Optional
 from _platforms import PLATFORMS, PLATFORM_CHOICES, resolve as resolve_platform
 from _common import STATE, require_tool, add_common, apply_common, default_font_file, die, emit, escape_drawtext, ffmpeg_base, info, parse_time, probe, run, run_analysis, time_arg
 from _common import drawtext_fontfile  # noqa: E402
+from _common import ffmpeg_filters, filter_missing_hint  # noqa: E402
+from _common.emit import note, record_fallback  # noqa: E402
 
 FONT = "fontcolor=white:fontsize=h/18:box=1:boxcolor=black@0.55:boxborderw=6:x=8:y=8"
 
@@ -162,6 +164,14 @@ def main() -> int:
     # a resolvable font file, given as fontfile=, is the only form confirmed not to crash drawtext's
     # own fontconfig resolution on some real Windows ffmpeg builds (#100); font= is the fallback when
     # nothing can be resolved, unchanged from before this existed.
+    if not args.no_timecode and ffmpeg_filters() and "drawtext" not in ffmpeg_filters():
+        # a build without drawtext (Homebrew's plain ffmpeg) cannot stamp the timecodes, but the
+        # sheet itself only needs tile: draw it without them and say so, rather than failing the
+        # step every picture change ends with (#394)
+        args.no_timecode = True
+        record_fallback("this ffmpeg build has no drawtext filter, so the sheet has no timecode stamps "
+                        "(the same as --no-timecode)", "look_no_drawtext")
+        note("no timecode stamps: this ffmpeg build has no drawtext filter (" + filter_missing_hint("drawtext") + ")")
     if args.no_timecode:
         font_prefix, tc = "", ""   # nothing is drawn, so no font is resolved or copied
     else:

package/scripts/silence.py

@@ -246,8 +246,12 @@ def resolve_filler(args, meta):
         # faster-whisper word_timestamps=True, openai-whisper --word_timestamps True). An SRT
         # cannot answer this question: a cue has a start and an end, a word does not.
         # The call's own result, not module state: engine, routing and notes belong to this run.
+        # Whisper leaves "uh"/"um" out of its text unless the context shows them: prime it with the
+        # filler list of the language asked for (English under auto, where most filler cuts are)
+        primer = FILLER_WORDS.get(args.filler_lang if args.filler_lang != "auto" else "en") or []
+        prompt = (", ".join(sorted(primer)[:8]).capitalize() + ". So, I mean...") if primer else None
         asr_run = _asr.transcribe_words_result(args.input, args.filler_lang if args.filler_lang != "auto" else None,
-                                               model=args.model, engine=args.engine)
+                                               model=args.model, engine=args.engine, prompt=prompt)
         words, engine = asr_run.words, asr_run.engine
         transcription = dict(asr_run.facts)
         args._asr_notes = list(asr_run.notes)