Files
Linux_post_install/bin/pos-media-grab
T
Your Name 072a8e72c1
gates / consistency-and-conventions (push) Successful in 2m14s
feat: pos media grab — smart URL classifier for auto-download
New tool that classifies URLs by domain and delegates to pos media mp3
(audio) or pos media mp4 (video). Listener gains URL detection step
between prefix map and AI bridge — bare URLs auto-download.

Domain rules: music.youtube.com/soundcloud/bandcamp → mp3,
youtube/vimeo/twitch → mp4 --best, unknown → configurable default.

28 test cases / 70 assertions / 0 failed.
make gen && make check green, make lint 0 FAIL / 0 WARN.
2026-09-04 10:48:49 -04:00

228 lines
7.7 KiB
Bash
Executable File

#!/usr/bin/env bash
set -euo pipefail
# POS: media grab — Auto-download URL as audio or video (classify + route)
# POS_FLAGS: --audio --video --best --worst --output --no-playlist --cookies --dry-run
# POS_CONFIG: grab | grab.env | GRAB_DEFAULT=:Default mode for unknown domains (video or audio, default video)
source "$(dirname "$0")/../lib/common.sh" 2>/dev/null || source "$(dirname "$0")/common.sh"
# Load grab.env config (env-seam: GRAB_DEFAULT)
load_grab_config() {
local f="$HOME/.config/linux_post_install/grab.env" k v
[ -f "$f" ] || return 0
while IFS='=' read -r k v; do
[ -n "$k" ] || continue
case "$k" in
\#*) continue ;;
esac
v="${v%\"}"; v="${v#\"}"; v="${v%\'}"; v="${v#\'}"
if [ -z "${!k:-}" ]; then
export "$k"="$v"
fi
done < <(grep -E '^[A-Z_]+=' "$f" || true)
}
load_grab_config
# ── URL classification ─────────────────────────────────────────
classify_url() {
local url="$1" mode="${GRAB_DEFAULT:-video}"
case "$url" in
*music.youtube.com*) echo "audio" ;;
*soundcloud.com*) echo "audio" ;;
*bandcamp.com*) echo "audio" ;;
*youtube.com*|*youtu.be*) echo "video" ;;
*vimeo.com*) echo "video" ;;
*twitch.tv*) echo "video" ;;
*) echo "$mode" ;;
esac
}
usage() {
cat <<EOF
Usage: pos media grab [options] <url>
Auto-download a URL as audio or video. Classifies the domain and delegates
to 'pos media mp3' (audio) or 'pos media mp4' (video).
Options:
--audio Force audio (mp3) download
--video Force video (mp4) download
--best Best quality for video (default for non-interactive)
--worst Lowest quality for video
-o, --output <dir> Output directory (passed to mp3/mp4)
--no-playlist Download only the single video
--cookies <file> Netscape cookies.txt for age-gated content
--dry-run Print the command that would run, don't execute
-h, --help This help
Examples:
pos media grab https://music.youtube.com/watch?v=abc
pos media grab https://youtube.com/watch?v=xyz
pos media grab --audio https://vimeo.com/123
pos media grab --worst https://youtu.be/abc
pos media grab --dry-run https://soundcloud.com/artist/track
EOF
exit 0
}
# ── Arg parsing ────────────────────────────────────────────────
URL=""
FORCE_AUDIO=0
FORCE_VIDEO=0
BEST=0
WORST=0
DRY_RUN=0
EXTRA_ARGS=()
while [[ $# -gt 0 ]]; do
case "$1" in
-h|--help) usage ;;
--audio) FORCE_AUDIO=1; shift ;;
--video) FORCE_VIDEO=1; shift ;;
--best) BEST=1; shift ;;
--worst) WORST=1; shift ;;
-o|--output)
[ $# -ge 2 ] || err "pos media grab: --output needs a value"
EXTRA_ARGS+=(--output "$2"); shift 2 ;;
--no-playlist) EXTRA_ARGS+=(--no-playlist); shift ;;
--cookies)
[ $# -ge 2 ] || err "pos media grab: --cookies needs a value"
EXTRA_ARGS+=(--cookies "$2"); shift 2 ;;
--dry-run) DRY_RUN=1; shift ;;
-*) err "pos media grab: Unknown option: $1 (see --help)" ;;
*)
[ -z "$URL" ] && URL="$1" && shift || err "pos media grab: Unexpected argument: $1" ;;
esac
done
[ -n "$URL" ] || usage
# Validate URL scheme
case "$URL" in
http://*|https://*) ;;
*) err "pos media grab: not a valid URL: $URL (must start with http:// or https://)" ;;
esac
# Validate mutually exclusive overrides
[ "$FORCE_AUDIO" -eq 1 ] && [ "$FORCE_VIDEO" -eq 1 ] && \
err "pos media grab: --audio and --video are mutually exclusive"
[ "$BEST" -eq 1 ] && [ "$WORST" -eq 1 ] && \
err "pos media grab: --best and --worst are mutually exclusive"
# ── Classification ─────────────────────────────────────────────
mode=""
if [ "$FORCE_AUDIO" -eq 1 ]; then
mode="audio"
elif [ "$FORCE_VIDEO" -eq 1 ]; then
mode="video"
else
mode="$(classify_url "$URL")"
fi
# ── Build delegated command ────────────────────────────────────
DELEGATE_ARGS=()
if [ "$mode" = "audio" ]; then
DELEGATE_ARGS=(pos media mp3 "${EXTRA_ARGS[@]}")
else
# mp4 route: --best by default (non-interactive), --worst if user passes it
if [ "$WORST" -eq 1 ]; then
DELEGATE_ARGS=(pos media mp4 --worst "${EXTRA_ARGS[@]}")
else
DELEGATE_ARGS=(pos media mp4 --best "${EXTRA_ARGS[@]}")
fi
fi
# ── Dry run ────────────────────────────────────────────────────
if [ "$DRY_RUN" -eq 1 ]; then
echo "${DELEGATE_ARGS[*]} $URL"
exit 0
fi
# ── Execute ────────────────────────────────────────────────────
output=""
rc=0
if output=$("${DELEGATE_ARGS[@]}" "$URL" 2>&1); then
rc=0
else
rc=$?
fi
if [ "$rc" -ne 0 ]; then
# Summarize stderr for the user
summary="$(printf '%s' "$output" | grep -i 'error\|fail' | head -1 || true)"
[ -z "$summary" ] && summary="exit code $rc"
err "pos media grab: ❌ Download failed: $summary"
fi
# ── Metadata + summary ────────────────────────────────────────
title=""
duration=""
file_path=""
# Determine expected output directory
if [ "$mode" = "audio" ]; then
out_dir="$HOME/Music"
for (( i=0; i<${#EXTRA_ARGS[@]}; i++ )); do
if [ "${EXTRA_ARGS[$i]}" = "--output" ] && [ $(( i + 1 )) -lt ${#EXTRA_ARGS[@]} ]; then
out_dir="${EXTRA_ARGS[$(( i + 1 ))]}"
break
fi
done
file_ext="mp3"
else
out_dir="$HOME/Videos"
for (( i=0; i<${#EXTRA_ARGS[@]}; i++ )); do
if [ "${EXTRA_ARGS[$i]}" = "--output" ] && [ $(( i + 1 )) -lt ${#EXTRA_ARGS[@]} ]; then
out_dir="${EXTRA_ARGS[$(( i + 1 ))]}"
break
fi
done
file_ext="mp4"
fi
# Fetch metadata (fast, no download)
if command -v yt-dlp &>/dev/null; then
meta="$(yt-dlp --print title --print duration_string --no-warnings "$URL" 2>/dev/null || true)"
title="$(printf '%s' "$meta" | sed -n '1p')"
duration="$(printf '%s' "$meta" | sed -n '2p')"
fi
# Find the downloaded file (most recent matching extension in out_dir)
if [ -d "$out_dir" ]; then
file_path="$(find "$out_dir" -maxdepth 1 -name "*.$file_ext" -printf '%T@ %p\n' 2>/dev/null \
| sort -rn | head -1 | cut -d' ' -f2- || true)"
fi
# Build summary
[ -z "$title" ] && title="$(basename "$URL" | sed 's/[?#].*//')"
[ -z "$duration" ] && duration="?"
if [ "$mode" = "audio" ]; then
emoji="🎵"
else
emoji="🎬"
fi
echo "$emoji Downloaded: $title ($duration)"
if [ -n "$file_path" ] && [ -f "$file_path" ]; then
file_size="$(stat --printf='%s' "$file_path" 2>/dev/null || echo "0")"
# Format size in human-readable form
if [ "$file_size" -ge 1073741824 ]; then
size_human="$(awk "BEGIN { printf \"%.1f GB\", $file_size / 1073741824 }")"
elif [ "$file_size" -ge 1048576 ]; then
size_human="$(awk "BEGIN { printf \"%.1f MB\", $file_size / 1048576 }")"
elif [ "$file_size" -ge 1024 ]; then
size_human="$(awk "BEGIN { printf \"%.1f KB\", $file_size / 1024 }")"
else
size_human="${file_size} B"
fi
# Show path relative to HOME
rel_path="${file_path/#$HOME/\~}"
echo "📁 $rel_path ($size_human)"
else
echo "📁 $out_dir/ ($file_ext)"
fi