代码之家 › 专栏 › 技术社区 › Tiago Sousa Neves

使用sed更改多个文件中的字符串

filenames sed bash string

Tiago Sousa Neves · 技术社区 · 7 年前

我有多个这样的文件:

67 572 Cy_aJATC23?????????????????????????GCCCTGCTGAGCCCATAGTCACAGAAGGACAAGTGCCAGACCACCCCCTTTCTGGCCTTATCTGTGGTGGTGCGCGCACTGCCTCCACACACAGGCCTCAGCGCCACTGCTGTGTGTGTGATCAATGCCCTCCAGTCTGATGCCAAGAGATCAAGAGATCAAGATCAGCCAAGATCTCCCGAAGACTTGATGAGTGAAGATGAATTATGATGTGGAAATCCATGCAAGCCAAGGTGAAGAGAGAGAGATGAGTGAGTGAGTGAGTGAGTGAGTGAGTGAGTGAGTGAGTGAGTGAGTGCGCGTTCTCTCTCTCTCTCTCAGATCAAGAAAAAGAGAAAAAAAGAGATAGAGAGATGCGAGTGAGTGAGTGAGTGAGTGAGTGAGTGCGCGCTCTCTCTCTCTCAtatatcagaagaagagttaattgatgatgatgtcatttggcacggcctgttcccATAGAGCCCCCATCTGATATCTGACAATAACAA

我想换第一行

通过

文件名为“ACA\u exon2”。phylip sequential’我只想要第一部分‘ACA_exon2’,但我都同意,我可以在之后更改它。

我对此非常了解

for file in ~/folder/*; do sed -E 's/^ (\w+)( \w+)/\[$file\2]/g' $file ; done;

这就是结果

CYU aJATC23?????????????????????????????????????????????????????????????????TACCACCAGAGATGTTAACAAGAGCTGCATGTTTTGATCAAGATAGAGAAACTGTTTATGTGTTTGGCACAAAACTAATGTAGGAAAGTCATGGCCCTGTTGTGATGTGTGTGAAGCTGAAGAGGCGCGCTCTAATGTGTGTGTGTCAAGTAAGATCAAGACTTGTTATGGCAAGAGACTTGTTAGTCAGATGAACTTGTTTTAGCTGAACTCCTGCCTGCTGTTGCCTCAATGAATAGACTGCACCAGGCATTAGTGAAACAAACACAAGACTGCACCAGGCATGCATTATTGAAACAAACAAGATAGATGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAAGAAAGAGAGAAAGAGAGATGCATTTCCCCCAAGAGAGAGAGAGATAGATAGATAGATCACAGCTCTCTTTTGTTAGAAAGTGAGAAACACACTGGTGTGTGTGCATGTGCATGTGTAAGATGATCAACCGGTGACGCGAACGGACGAAATGAAACTTTTCCCataGTGATCAGAGATGCTCAGAAACTGGATT???

我想要这个

CYU aJATC23?????????????????????????????????????????????????????????????????TACCACCAGAGATGTTAACAAGAGCTGCATGTTTTGATCAAGATAGAGAAACTGTTTATGTGTTTGGCACAAAACTAATGTAGGAAAGTCATGGCCCTGTTGTGATGTGTGTGAAGCTGAAGAGGCGCGCTCTAATGTGTGTGTGTCAAGTAAGATCAAGACTTGTTATGGCAAGAGACTTGTTAGTCAGATGAACTTGTTTTAGCTGAACTCCTGCCTGCTGTTGCCTCAATGAATAGACTGCACCAGGCATTAGTGAAACAAACACAAGACTGCACCAGGCATGCATTATTGAAACAAACAAGATAGATGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAAGAAAGAGAGAAAGAGAGATGCATTTCCCCCAAGAGAGAGAGAGATAGATAGATAGATCACAGCTCTCTTTTGTTAGAAAGTGAGAAACACACTGGTGTGTGTGCATGTGCATGTGTAAGATGATCAACCGGTGACGCGAACGGACGAAATGAAACTTTTCCCataGTGATCAGAGATGCTCAGAAACTGGATT???

你能帮我吗?

提前感谢

3 回复 | 直到 7 年前

dawg 7 年前

鉴于:

$ cat f1.exon
67 572 Cy_aJATC23 ?????????????????????????
GCCCTGCTGAGGCCCATAGTCACAGAAGGACAAAGTGCCAGACCACCCCCTTTCTGGCCTT
$ cat f2.exon
67 573 Cy_aJATC23 ?????????????????????????
GCCCTGCTGAGGCCCATAGTCACAGAAGGACAAAGTGCCAGACCACCCCCTTTCTGGCCTT

你可以在 awk

$ awk 'FNR==1{$1="[" FILENAME; $2=$2 "]"} 1' *.exon
[f1.exon 572] Cy_aJATC23 ?????????????????????????
GCCCTGCTGAGGCCCATAGTCACAGAAGGACAAAGTGCCAGACCACCCCCTTTCTGGCCTT
[f2.exon 573] Cy_aJATC23 ?????????????????????????
GCCCTGCTGAGGCCCATAGTCACAGAAGGACAAAGTGCCAGACCACCCCCTTTCTGGCCTT

sed :

for fn in *.exon; do 
    sed -E '1s/^[[:digit:]]+( [[:digit:]]+)/['"$fn"'\1]/' "$fn"
done    
# same output

mv 临时文件将原始文件重命名为新内容:

for fn in *.exon; do 
    awk 'FNR==1{$1="[" FILENAME; $2=$2 "]"} 1' "$fn" > a_temp_exon
    mv a_temp_exon "$fn"
done

具有 sed公司 -i

说明:

awk 'FNR==1{$1="[" FILENAME; $2=$2 "]"} 1' *.exon
      ^   ^                                       first line only
           ^                          ^           do this action
            ^                                     first field
               ^     ^                            literal [ and FILENAME
                             ^                    second field
                                 ^                it's same value
                                    ^             literal ]
                                        ^         print the line

Maxim Norin 7 年前

试试这个:

sed -E "s/^ (\w+)( \w+)/\[$file\2]/g" $file

sed -E 's/^ (\w+)( \w+)/\['$file'\2]/g' $file

Louise 7 年前

sed -E "s/^ (\\w+)( \\w+)/\\[$file\\2]/g" $file