abaixo está o formato em um arquivo. alguém pode ajudar na conversão das duas primeiras colunas que estão no formato de data para o formato de época?
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702058496,0,0,0,0,0,0,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702058496,2,0,0,0,0,0,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702058496,3,332,0,0,0,332,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702058496,4,0,0,0,0,0,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702062592,0,0,0,0,0,0,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702062592,2,0,0,0,0,919,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702062592,3,919,0,0,0,0,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702062592,4,0,0,0,0,0,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702066688,0,0,0,0,0,0,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702066688,2,0,0,0,0,0,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702066688,3,0,0,0,0,919,0,0
02-01-2018 05:44:00,2018-01-02 05:59:00,1,702066688,4,0,0,0,0,0,0,0
02-01-2018 10:44:00,2018-01-02 10:59:00,1,702058496,0,0,0,0,0,0,0,0
02-01-2018 10:44:00,2018-01-02 10:59:00,1,702058496,2,0,0,0,0,0,0,0
eu tentei com o seguinte comando.
aa= cat ATMIF.180102-0615 |grep -|awk -F ',' '{print $1}'; for i in cat ATMIF.180102-0615 |grep -|awk -F ',' '{print $1}';do date -d "$aa" "+%s"; feito
mas está dando erro como: date: invalid date 2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00' date: invalid date2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05 :44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018 -01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 10: 44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018- 01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44 :00\n2018-01-02 10:44:00\n2018-01-02 10:44:00' data: data inválida `2018-01-02 05:44:00\n2018-01-02 05:44: 00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01- 02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00\n2018-01-02 05:44:00 \n2018-01-02 05:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\ n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00\n2018-01-02 10:44:00'
Responder1
Isso seria suficiente, assumindo que os dados estão em:
IFS=,;while read line; do
ff=($line);
ff[0]=`date -d "${ff[0]}" +%s`;
ff[1]=`date -d "${ff[1]}" +%s`;
echo "${ff[*]}";
done < file.in > file.out
Nota: pode ser necessário adicionar 'UTC' no comando de data de acordo com o formato de origem.
Agora, para fazer isso no local, conforme solicitado pelo seu comentário, você precisa fazer um loop em todos os arquivos. Por exemplo, usando o comando find para corresponder todos os arquivos .csv recursivamente em /mypath/to-files/:
for file in $(find /mypath/to-files -name "*.csv"); do
cp $file /tmp/tmp.csv
IFS=,;while read line; do
ff=($line);
ff[0]=`date -d "${ff[0]}" +%s`;
ff[1]=`date -d "${ff[1]}" +%s`;
echo "${ff[*]}";
done < /tmp/tmp.csv > $file
done