Eu usei sed
para permitir apenas o formato numérico e csv, usei sed '/[^0-9, -, .]/d'
como regex. Ainda tem "_","1-0","1-1"
caractere como o / p que me preocupa.
sed 's/'"$(printf '5')"'//g' ./abc.CSV | sed '/[^0-9, -, .]/d'
1,090216002122,128,126,134,124,-0,00,220,28,292688,077,040911,0858,0159
1,090216002123,128,126,134,124, 000,00,_220,28,292688,077,040911,0858,0159
1,090216002125,128,126,134,124, 000,00,220,28,292688,077,040911,0858,170
1,090216002126,128,126,134,124, 000,-.00,220,28,292688,077,040911,0858,0159
1,090216002130,128,126,134,124, 000,00,220,28,292688,077,040911,0858,
1,090216002134,128,126,134,124, 000,00,220,28,,077,040911,0858,0159
1,090216002135,128,126,134,124, 000,00,220,28,077,040911,0858,0159
11,090216002137,128,126,134,124, 000,00,220,28,292688,077,040911,0858,0159
1-1,090216002137,128,126,134,124, 000,00,220,28,292688,077,040911,0858,0159
1,000000000000,000,0,0,0,0,0,0,0,0,0,0,0,0
1,040216002137,128,126,134,124, 000,00,220,28,292688,077,040911,0858,0159
1,090216002137,128,126,134,124, 000,00,220,28,292688,077,040911,0858,0-1
Eu tentei com ($9 ~ /^-?[0-9]+([.][0-9]+)?$/ )
parece não funcionar com o número do espaço. Eu também tentei com
Qual é a melhor maneira de analisar apenas numérico de um arquivo csv? Obrigado.