music_folder_scan2

#!/bin/bash

#todo: handle commas within double quotes in csv files
# e.g. "blah,blah"
# possible oneliner:
# awk -F'"' -v OFS='' '{ for (i=2; i<=NF; i+=2) gsub(",", "", $i) } 1' source_file.csv >processed_file.csv

# this Script does the following:
#1.iterate through the wav files in a directory
#2.scan the files with acrcustom.py to get metadata from acrcloud
#3.create one csv per directory containg data about the files.
#4.remove the temporary files.
dirname=${PWD##*/}
id=mfs


for f in *.wav
do
  echo "processing file $f with ACRCloud"
  #music_scan "$f"
  python /Users/ewancolsell/Documents/GitHub/acrcloud_scan_files_python/acrcustom.py  "$f"
  #acr custom.py produces a .csv with the format:
  # filename,timestamp,custom_files_title,custom_acrid,title,artists,album,acrid,played_duration,label,isrc,upc,dezzer,spotify,itunes,youtube

  # for files with no match acrcloud produces a csv file with just an index and no records
  # the following will add a record showing "unknown_title" etc for files with no records

  if [[ $(wc -l <"$f.csv") -eq 1 ]]; then
    #statements
    echo "no acrcloud results for file $f adding a blank record..."
    echo "$f,,,,unknown_title,unknown_artists,unknown_album,,,,,,,,," >> "$f".csv

  fi

  # acrcustom generates a csv file with a load of fields we don't care about
  # here we remove the extra feilds.
  echo "remove extra fields from $f.csv"

  cut -d, -f1,5,6,7,9,10 "$f.csv" > "$f.$id.tmp0"

  # echo "remove duplicate records and sort"
  # sort -s -u "$0.tmp" > "$f"_processed.csv.tmp0
  echo "$f.$id.tmp0"
  cat "$f.$id.tmp0"
done
echo "finished scanning with acrcloud"

# get the index for the csv files
# do this just for the first .csv file
for f in *.$id.tmp0
do
  echo "getting index from $f"
head -n 1 "$f" > $id.index.tmp
cat $id.index.tmp
break
done

for f in *.$id.tmp0
do
  # echo "remove index and sort $f"
  # sed 1d $f | sort -s -u > "$f".$id.tmp1
  echo "remove index from $f"
  sed 1d "$f" > "$f.$id.tmp1"

  echo "remove duplicate records and sort in $f.$id.tmp1"
  sort -s -u "$f.$id.tmp1" > "$f.$id".tmp2
done


echo "merging csv files...."
cat *.$id.tmp2 > "all.$id.tmp"


echo "restore index line at begining of file."
cat "$id.index.tmp" "all.$id.tmp" > "$dirname.csv"

echo "removing temporary files..."
rm *.tmp *.tmp1 *.tmp0 *.tmp2 *.wav.csv  > /dev/null

cat "$dirname.csv"
echo "wrote file $dirname.csv"