From: Steven Baltakatei Sandoval Date: Fri, 27 Oct 2023 13:01:47 +0000 (+0000) Subject: feat(user/bk-copy-rand-music):Optimize file processing X-Git-Url: https://zdv2.bktei.com/gitweb/BK-2020-03.git/commitdiff_plain/7e772cd7416dcaef1b0f9a8cd09ce5e8d34c05c9?ds=sidebyside feat(user/bk-copy-rand-music):Optimize file processing - Now ignores blacklist of extensions. - Increase minimum file duration to 30s. - Now adds files until 95% of target size or duration is reached instead of stopping as soon as a large or long file is considered. - Now skips `ffprobe` of files (slow) if `du` size check (fast) fails. --- diff --git a/user/bk-copy-rand-music b/user/bk-copy-rand-music index b807286..2303d50 100755 --- a/user/bk-copy-rand-music +++ b/user/bk-copy-rand-music @@ -1,7 +1,7 @@ #!/usr/bin/env bash # Desc: Copies random audio files # Usage: bk-copy-rand-music [dir SOURCE] [dir DEST] [int DURATION] ([int BYTES]) -# Version: 0.2.0 +# Version: 0.3.0 # Depends: BK-2020-03: bkshuf v0.1.0 declare -Ag appRollCall # Associative array for storing app status @@ -11,8 +11,9 @@ declare -a music_codecs # Array for storing valid codec names (e.g. "aac" "mp3") # Adjustable parameters music_codecs=("vorbis" "aac" "mp3" "flac" "opus"); # whitelist of valid codec_names ffprobe might return +ext_ignore=".ots\$|.mid\$|.json\$|.gz\$|.jpg\$|.png\$|.asc\$|.pdf\$|.txt\$|.vtt\$|\.SUM|.zip\$|.xz\$|.org\$|.txt\$"; # blacklist of file extensions for 'grep -Evi' max_filename_length="255"; # max output filename length -min_file_duration="10"; # minimum duration per music file +min_file_duration="30"; # minimum duration per music file max_file_duration="3600"; # maximum duration per music file min_file_size="100000"; # minimum size per music file (bytes) max_file_size="100000000"; # maximum size per music file (bytes) @@ -438,7 +439,9 @@ main() { # Populate list_files array while read -r line; do list_files+=("$line"); - done < <(find -L "$dir_source" -maxdepth "$max_find_depth" -type f | sort); + done < <(find -L "$dir_source" -maxdepth "$max_find_depth" -type f | \ + grep -Ev "$ext_ignore" | \ + sort); # Test and add random elements of list_files to list_copy dur=0; # Initialize duration @@ -449,12 +452,23 @@ main() { ## Get element count of list_files array file_count="${#list_files[@]}"; while read -r line && \ - [[ $dur -le $dur_dest ]] && \ - [[ $siz -le $siz_dest ]] && \ + [[ $dur -le $((dur_dest * 95 / 100)) ]] && \ + [[ $siz -le $((siz_dest * 95 / 100)) ]] && \ [[ $n -le $file_count ]]; do - #yell "DEBUG:list_copy building loop:$n"; + ((n++)); + + yell "DEBUG:list_copy building loop:$n/$file_count"; # debug + printf "DEBUG:%8d,%8d,%8d/%8d,%8d/%8d\n" "$dur_cand" "$siz_cand" "$dur" "$dur_dest" "$siz" "$siz_dest"; # debug + path_candfile="$line"; # path of candidate file + ### Check size + siz_cand="$(du -b "$path_candfile" | awk '{ print $1 }')"; # size in bytes + if ! checkInt "$siz_cand"; then continue; fi; # reject + if [[ "$((siz + siz_cand))" -gt "$siz_dest" ]]; then continue; fi; # reject + if [[ "$siz_cand" -lt "$min_file_size" ]]; then continue; fi; # reject + if [[ "$siz_cand" -gt "$max_file_size" ]]; then continue; fi; # reject + ### Check if has valid codec if ! check_parsable_audio_ffprobe "$path_candfile"; then continue; fi; # reject @@ -462,31 +476,28 @@ main() { file_format="$(get_audio_format "$path_candfile")"; if ! checkIsInArray "$file_format" "${music_codecs[@]}"; then continue; fi; # reject - ### Check and save duration + ### Check duration dur_cand="$(get_media_length "$path_candfile")"; dur_cand="${dur_cand%%.*}"; # convert float to int - if [[ "$((dur + dur_cand))" -gt "$dur_dest" ]]; then break; fi; # no more - dur_cand_wnow="$(printf "%s" "$dur_cand" | wc -m)"; # duration width count - if [[ $dur_cand_wnow -gt $dur_cand_w ]]; then - dur_cand_w="$dur_cand_wnow"; fi; - if ! checkInt "$dur_cand"; then continue; fi; # reject + if ! checkInt "$dur_cand"; then continue; fi; # reject + if [[ "$((dur + dur_cand))" -gt "$dur_dest" ]]; then continue; fi; # reject if [[ "$dur_cand" -lt "$min_file_duration" ]]; then continue; fi; # reject if [[ "$dur_cand" -gt "$max_file_duration" ]]; then continue; fi; # reject - ### Check and save size - siz_cand="$(du -b "$path_candfile" | awk '{ print $1 }')"; # size in bytes - if [[ "$((siz + siz_cand))" -gt "$siz_dest" ]]; then break; fi; # no more + ### Update stats digits widths + #### duration + dur_cand_wnow="$(printf "%s" "$dur_cand" | wc -m)"; # duration width count + if [[ $dur_cand_wnow -gt $dur_cand_w ]]; then + dur_cand_w="$dur_cand_wnow"; fi; + #### size siz_cand_wnow="$(printf "%s" "$siz_cand" | wc -m)"; # size width count if [[ $siz_cand_wnow -gt $siz_cand_w ]]; then siz_cand_w="$siz_cand_wnow"; fi; - if ! checkInt "$siz_cand"; then continue; fi; # reject - if [[ "$siz_cand" -lt "$min_file_size" ]]; then continue; fi; # reject - if [[ "$siz_cand" -gt "$max_file_size" ]]; then continue; fi; # reject - + ### Add/update candfile to array: ### list_copy (array with "duration, size, path") #yell "DEBUG:Adding $path_candfile"; - #printf "DEBUG:%8d,%8d,%s\n" "$dur_cand" "$siz_cand" "$path_candfile" 1>&2; + printf "DEBUG:%8d,%8d,%s\n" "$dur_cand" "$siz_cand" "$path_candfile" 1>&2; #printf "DEBUG:dur:%s\n" "$dur" 1>&2; #printf "DEBUG:siz:%s\n" "$siz" 1>&2; list_copy+=("$dur_cand,$siz_cand,$path_candfile"); # for copying with order @@ -494,10 +505,8 @@ main() { ### Update total duration $dur and total size $siz dur="$((dur + dur_cand))"; siz="$((siz + siz_cand))"; - #yell "DEBUG:dur:$dur"; - #yell "DEBUG:siz:$siz"; - - ((n++)); + yell "DEBUG:dur:$dur"; + yell "DEBUG:siz:$siz"; done < <(printf "%s\n" "${list_files[@]}" | bkshuf); #yell "DEBUG:BKSHUF_PARAM_LINEC:$BKSHUF_PARAM_LINEC"; @@ -531,7 +540,7 @@ main() { path_output="$dir_dest"/"$file_name"; ## Copy - must cp "$fpath" "$path_output" && yell "NOTICE:Copied ($fdur seconds): $fpath "; + must cp "$fpath" "$path_output" && yell "NOTICE:Copied ($(printf "%""$dur_cand_w"d "$fdur") seconds): $fpath "; #yell "DEBUG:Copied $file_basename to $dur_dest."; ## Append log