Add Pixabay music fallback

README.md

@@ -23,7 +23,7 @@ sudo apt install ffmpeg
 The default title/stamp styling uses these fonts:
 
 - Bebas Neue for the opening title, date, and location
-- Noto Sans SemiBold for the bottom-left/bottom-right stamps and per-file captions
+- Noto Sans Bold for the bottom-left/bottom-right stamps and per-file captions
 
 Both fonts support common Danish and Croatian characters such as `æ ø å Æ Ø Å č ć ž š đ Č Ć Ž Š Đ`.
 
@@ -33,9 +33,8 @@ Install them locally for the current user:
 mkdir -p ~/.local/share/fonts/google
 curl -L -o ~/.local/share/fonts/google/BebasNeue-Regular.ttf \
   'https://github.com/google/fonts/raw/main/ofl/bebasneue/BebasNeue-Regular.ttf'
-curl -L -o ~/.local/share/fonts/google/NotoSans-SemiBold.ttf \
+sudo apt install fonts-noto-core
-  'https://github.com/google/fonts/raw/main/ofl/notosans/NotoSans%5Bwdth,wght%5D.ttf'
+fc-cache -f
-fc-cache -f ~/.local/share/fonts/google
 ```
 
 Check that fonts are available:
@@ -45,6 +44,36 @@ fc-match 'Bebas Neue'
 fc-match 'Noto Sans'
 ```
 
+## Optional Pixabay music download
+
+If the input folder has no audio file, movmaker can download one random track from Pixabay Music.
+
+Save your Pixabay API key in either an environment variable:
+
+```bash
+export PIXABAY_API_KEY='your_key_here'
+```
+
+or in a local config file:
+
+```bash
+mkdir -p ~/.config/movmaker
+echo 'PIXABAY_API_KEY=your_key_here' > ~/.config/movmaker/pixabay.env
+chmod 600 ~/.config/movmaker/pixabay.env
+```
+
+The default music search query is:
+
+```text
+cinematic punk rock
+```
+
+Override it with:
+
+```bash
+python3 movmaker.py input --music-genre "upbeat rock"
+```
+
 ## Input folder
 
 Example:
@@ -134,6 +163,7 @@ python3 movmaker.py input \
   --audio-fade 10 \
   --audio-fade-in 2 \
   --video-fade 1.5 \
+  --music-genre "cinematic punk rock" \
   --resolution 1920x1080 \
   --fps 30
 ```
@@ -153,10 +183,12 @@ python3 movmaker.py input --preview
 - includes video clips inline
 - crossfades between media items over 3 seconds by default
 - loops/trims music to match the video
+- if no audio file exists in the input folder, searches Pixabay Music using `--music-genre` and downloads one random track; default query is `cinematic punk rock`
+- reads the Pixabay API key from `PIXABAY_API_KEY` or `~/.config/movmaker/pixabay.env`
 - fades music in over 2 seconds and out over 10 seconds by default
 - fades video in/out over 1.5 seconds by default
 - overlays the opening title for the first 6 seconds: large title, then date/location on one line
-- displays optional per-file captions at bottom center with a shadow; use `|` to split caption lines
+- displays optional per-file captions at bottom center with a thin black outline; use `|` to split caption lines
 - adds persistent bottom-left title/date/place and bottom-right `@bubulescu` stamps
 - preserves special characters in rendered overlays and metadata
 - uses safe ASCII only for generated filenames, so special letters become readable equivalents like `Č` -> `C`, `å` -> `a`, `ø` -> `o`

movmaker.py

@@ -14,6 +14,7 @@ import concurrent.futures
 import json
 import math
 import os
+import random
 import re
 import shlex
 import subprocess
@@ -33,6 +34,7 @@ DATA_NAMES = {"data.txt", "info.txt", "title.txt"}
 BASE_DIR = Path(__file__).resolve().parent
 FFMPEG = BASE_DIR / "ffmpeg" / "ffmpeg"
 FFPROBE = BASE_DIR / "ffmpeg" / "ffprobe"
+PIXABAY_ENV = Path.home() / ".config" / "movmaker" / "pixabay.env"
 
 
 def tool_path(local_path: Path, fallback: str) -> str:
@@ -516,6 +518,71 @@ def normalize_clip(item: MediaItem, out: Path, width: int, height: int, fps: int
     run(cmd)
 
 
+def load_pixabay_api_key() -> str | None:
+    if os.environ.get("PIXABAY_API_KEY"):
+        return os.environ["PIXABAY_API_KEY"]
+    if PIXABAY_ENV.exists():
+        for line in PIXABAY_ENV.read_text(encoding="utf-8", errors="replace").splitlines():
+            line = line.strip()
+            if not line or line.startswith("#") or "=" not in line:
+                continue
+            key, value = line.split("=", 1)
+            if key.strip() == "PIXABAY_API_KEY":
+                return value.strip().strip("'\"")
+    return None
+
+
+def download_pixabay_music(genre: str, out_dir: Path) -> Path | None:
+    api_key = load_pixabay_api_key()
+    if not api_key:
+        print("No audio files found and no Pixabay API key configured; continuing without music.")
+        return None
+    query = urllib.parse.urlencode({
+        "key": api_key,
+        "q": genre,
+        "media_type": "music",
+        "safesearch": "true",
+        "per_page": "50",
+    })
+    request = urllib.request.Request(
+        f"https://pixabay.com/api/audio/?{query}",
+        headers={"User-Agent": "movmaker/1.0"},
+    )
+    try:
+        with urllib.request.urlopen(request, timeout=20) as response:
+            data = json.loads(response.read().decode("utf-8"))
+    except Exception as exc:
+        print(f"Could not search Pixabay music: {exc}")
+        return None
+
+    hits = [hit for hit in data.get("hits", []) if hit.get("audio")]
+    if not hits:
+        print(f"No Pixabay music found for genre/query: {genre}")
+        return None
+    hit = random.choice(hits)
+    audio_url = hit["audio"]
+    title = hit.get("title") or "pixabay_music"
+    filename = f"pixabay_{hit.get('id', 'music')}_{slugify(title)}.mp3"
+    out_dir.mkdir(parents=True, exist_ok=True)
+    out = out_dir / filename
+    print(f"Downloading Pixabay music: {title}")
+    try:
+        urllib.request.urlretrieve(audio_url, out)
+    except Exception as exc:
+        print(f"Could not download Pixabay music: {exc}")
+        return None
+    attribution = out.with_suffix(".txt")
+    attribution.write_text(
+        "Pixabay music downloaded by movmaker\n"
+        f"Title: {title}\n"
+        f"Artist/User: {hit.get('user', '')}\n"
+        f"URL: {hit.get('pageURL', '')}\n"
+        "Source: Pixabay\n",
+        encoding="utf-8",
+    )
+    return out
+
+
 def concat_or_mix_audio(audio_files: list[Path], out: Path) -> Path | None:
     if not audio_files:
         return None
@@ -602,8 +669,8 @@ def build_video(
     final_label = "vout"
     title_label = "titled"
     title_font = str(Path.home() / ".local/share/fonts/google/BebasNeue-Regular.ttf")
-    stamp_font = str(Path.home() / ".local/share/fonts/google/NotoSans-SemiBold.ttf")
+    stamp_font = "/usr/share/fonts/truetype/noto/NotoSans-Bold.ttf"
-    stamp_bold_font = str(Path.home() / ".local/share/fonts/google/NotoSans-SemiBold.ttf")
+    stamp_bold_font = "/usr/share/fonts/truetype/noto/NotoSans-Bold.ttf"
     if title_lines:
         title_font_size = max(60, height // 10)
         subtitle_font_size = max(40, height // 18)
@@ -642,7 +709,7 @@ def build_video(
         filters.append(f"[{current}]copy[{title_label}]")
 
     caption_label = title_label
-    caption_size = max(16, int((subtitle_font_size if title_lines else max(40, height // 18)) * 0.5))
+    caption_size = max(24, int((subtitle_font_size if title_lines else max(40, height // 18)) * 0.7))
     for idx, caption in enumerate(captions):
         if not caption:
             continue
@@ -673,18 +740,12 @@ def build_video(
             next_label = f"caption{idx}_{line_idx}"
             filters.append(
                 f"[{caption_label}]drawtext=fontfile='{stamp_bold_font}':{textfile_arg(caption_line)}:x=(w-text_w)/2:y={y_expr}:"
-                f"fontsize={caption_size}:fontcolor=white:borderw=0:shadowcolor=black@0.75:shadowx=2:shadowy=2:"
+                f"fontsize={caption_size}:fontcolor=white:borderw=1:bordercolor=black:"
                 f"alpha='{caption_alpha}':enable='between(t,{start:.3f},{end:.3f})'[{next_label}]"
             )
-            bold_label = f"caption{idx}_{line_idx}b"
+            caption_label = next_label
-            filters.append(
-                f"[{next_label}]drawtext=fontfile='{stamp_bold_font}':{textfile_arg(caption_line)}:x=(w-text_w)/2+1:y={y_expr}:"
-                f"fontsize={caption_size}:fontcolor=white:borderw=0:shadowcolor=black@0.75:shadowx=2:shadowy=2:"
-                f"alpha='{caption_alpha}':enable='between(t,{start:.3f},{end:.3f})'[{bold_label}]"
-            )
-            caption_label = bold_label
 
-    stamp_size = max(14, height // 65)
+    stamp_size = max(16, height // 58)
     stamp_lines = [x for x in [data.title, data.date, data.place] if x]
     stamp_text = drawtext_escape(" | ".join(stamp_lines))
     stamp_left_label = "stamp_left"
@@ -742,6 +803,7 @@ def main(argv: list[str] | None = None) -> int:
     parser.add_argument("--crf", type=int, default=20, help="x264 quality; lower is better/larger, higher is faster/smaller")
     parser.add_argument("--jobs", type=int, default=1, help="parallel clip preparation jobs; try 2 on Raspberry Pi")
     parser.add_argument("--preview", action="store_true", help="fast preview: 1280x720, 24 fps, ultrafast, CRF 28")
+    parser.add_argument("--music-genre", default="cinematic punk rock", help="Pixabay music search query used only when input has no audio files")
     args = parser.parse_args(argv)
 
     require_tool("ffmpeg")
@@ -768,6 +830,11 @@ def main(argv: list[str] | None = None) -> int:
     audio_fade_in = max(0.0, args.audio_fade_in)
     video_fade = max(0.0, args.video_fade)
 
+    if not audio_files:
+        downloaded_audio = download_pixabay_music(args.music_genre, args.input)
+        if downloaded_audio:
+            audio_files.append(downloaded_audio)
+
     output = default_output_path(media[0].path, data, args.out_dir)
 
     # Image duration means fully visible time, not including crossfades.