Unix替换文件中的特殊字符_Unix_Sed

Unix替换文件中的特殊字符

unix sed

Unix替换文件中的特殊字符,unix,sed,Unix,Sed,我有一个文件如下： ["2015.01.07 10:06:24 UTC","2015.01.07 10:06:34 UTC","2015.01.07 10:06:48 UTC","<a href='/jobhistory/job/job_1418436639844_231528'>job_1418436639844_231528</a>","PigLatin:agg_daily_total_searches_and_unique_searc","data_svc","ma

我有一个文件如下：

["2015.01.07 10:06:24 UTC","2015.01.07 10:06:34 UTC","2015.01.07 10:06:48 UTC","<a href='/jobhistory/job/job_1418436639844_231528'>job_1418436639844_231528</a>","PigLatin:agg_daily_total_searches_and_unique_searc","data_svc","marathon","SUCCEEDED","1","1","0","0"],
["2015.01.07 10:06:28 UTC","2015.01.07 10:06:35 UTC","2015.01.07 10:13:28 UTC","<a href='/jobhistory/job/job_1418436639844_231529'>job_1418436639844_231529</a>","EndorsementsPendingImpressionEvent-c1201ed1-fb23-4","kafkaetl","marathon","SUCCEEDED","1","1","1","1"],

[“2015.01.07 10:06:24 UTC”，“2015.01.07 10:06:34 UTC”，“2015.01.07 10:06:48 UTC”，“PigLatin:agg_daily_total_searc_and_unique_searc”，“data_svc”，“marathon”，“succeed”，“1”，“1”，“0”，“0”]，
[“2015.01.07 10:06:28 UTC”、“2015.01.07 10:06:35 UTC”、“2015.01.07 10:13:28 UTC”、“背书暂停印象事件-c1201ed1-fb23-4”、“卡夫卡特尔”、“马拉松”、“成功”、“1”、“1”、“1”]，

我需要分别从每行的开头和结尾删除

和

]，

。

您可以尝试下面的sed命令

sed 's/^\[\|\],$//g' file

断言我们在一行的开头，

断言我们在一行的结尾。由于basic sed使用BRE（基本正则表达式），因此需要转义

符号，以便像逻辑OR运算符一样工作

例如：

$ cat ri
["2015.01.07 10:06:24 UTC","2015.01.07 10:06:34 UTC","2015.01.07 10:06:48 UTC","<a href='/jobhistory/job/job_1418436639844_231528'>job_1418436639844_231528</a>","PigLatin:agg_daily_total_searches_and_unique_searc","data_svc","marathon","SUCCEEDED","1","1","0","0"],
["2015.01.07 10:06:28 UTC","2015.01.07 10:06:35 UTC","2015.01.07 10:13:28 UTC","<a href='/jobhistory/job/job_1418436639844_231529'>job_1418436639844_231529</a>","EndorsementsPendingImpressionEvent-c1201ed1-fb23-4","kafkaetl","marathon","SUCCEEDED","1","1","1","1"],
$ sed 's/^\[\|\],$//g' ri
"2015.01.07 10:06:24 UTC","2015.01.07 10:06:34 UTC","2015.01.07 10:06:48 UTC","<a href='/jobhistory/job/job_1418436639844_231528'>job_1418436639844_231528</a>","PigLatin:agg_daily_total_searches_and_unique_searc","data_svc","marathon","SUCCEEDED","1","1","0","0"
"2015.01.07 10:06:28 UTC","2015.01.07 10:06:35 UTC","2015.01.07 10:13:28 UTC","<a href='/jobhistory/job/job_1418436639844_231529'>job_1418436639844_231529</a>","EndorsementsPendingImpressionEvent-c1201ed1-fb23-4","kafkaetl","marathon","SUCCEEDED","1","1","1","1"

$cat ri
[“2015.01.07 10:06:24 UTC”，“2015.01.07 10:06:34 UTC”，“2015.01.07 10:06:48 UTC”，“PigLatin:agg_daily_搜索总数”和“unique_searc”，“data_svc”，“marathon”，“successed”，“1”，“1”，“0”，“0”]，
[“2015.01.07 10:06:28 UTC”、“2015.01.07 10:06:35 UTC”、“2015.01.07 10:13:28 UTC”、“背书暂停印象事件-c1201ed1-fb23-4”、“卡夫卡特尔”、“马拉松”、“成功”、“1”、“1”、“1”]，
$sed's/^\[\\\\]，$//g'ri
“2015.01.07 10:06:24 UTC”、“2015.01.07 10:06:34 UTC”、“2015.01.07 10:06:48 UTC”、“PigLatin:agg_daily_搜索总数_和_unique_searc”、“data_svc”、“marathon”、“successed”、“1”、“0”、“0”
“2015.01.07 10:06:28 UTC”、“2015.01.07 10:06:35 UTC”、“2015.01.07 10:13:28 UTC”、“背书PendingImpressionEvent-c1201ed1-fb23-4”、“卡夫卡特尔”、“马拉松”、“成功”、“1”、“1”、“1”

您可以尝试下面的sed命令

sed 's/^\[\|\],$//g' file

断言我们在一行的开头，

断言我们在一行的结尾。由于basic sed使用BRE（基本正则表达式），因此需要转义

符号，以便像逻辑OR运算符一样工作

例如：

$ cat ri
["2015.01.07 10:06:24 UTC","2015.01.07 10:06:34 UTC","2015.01.07 10:06:48 UTC","<a href='/jobhistory/job/job_1418436639844_231528'>job_1418436639844_231528</a>","PigLatin:agg_daily_total_searches_and_unique_searc","data_svc","marathon","SUCCEEDED","1","1","0","0"],
["2015.01.07 10:06:28 UTC","2015.01.07 10:06:35 UTC","2015.01.07 10:13:28 UTC","<a href='/jobhistory/job/job_1418436639844_231529'>job_1418436639844_231529</a>","EndorsementsPendingImpressionEvent-c1201ed1-fb23-4","kafkaetl","marathon","SUCCEEDED","1","1","1","1"],
$ sed 's/^\[\|\],$//g' ri
"2015.01.07 10:06:24 UTC","2015.01.07 10:06:34 UTC","2015.01.07 10:06:48 UTC","<a href='/jobhistory/job/job_1418436639844_231528'>job_1418436639844_231528</a>","PigLatin:agg_daily_total_searches_and_unique_searc","data_svc","marathon","SUCCEEDED","1","1","0","0"
"2015.01.07 10:06:28 UTC","2015.01.07 10:06:35 UTC","2015.01.07 10:13:28 UTC","<a href='/jobhistory/job/job_1418436639844_231529'>job_1418436639844_231529</a>","EndorsementsPendingImpressionEvent-c1201ed1-fb23-4","kafkaetl","marathon","SUCCEEDED","1","1","1","1"

$cat ri
[“2015.01.07 10:06:24 UTC”，“2015.01.07 10:06:34 UTC”，“2015.01.07 10:06:48 UTC”，“PigLatin:agg_daily_搜索总数”和“unique_searc”，“data_svc”，“marathon”，“successed”，“1”，“1”，“0”，“0”]，
[“2015.01.07 10:06:28 UTC”、“2015.01.07 10:06:35 UTC”、“2015.01.07 10:13:28 UTC”、“背书暂停印象事件-c1201ed1-fb23-4”、“卡夫卡特尔”、“马拉松”、“成功”、“1”、“1”、“1”]，
$sed's/^\[\\\\]，$//g'ri
“2015.01.07 10:06:24 UTC”、“2015.01.07 10:06:34 UTC”、“2015.01.07 10:06:48 UTC”、“PigLatin:agg_daily_搜索总数_和_unique_searc”、“data_svc”、“marathon”、“successed”、“1”、“0”、“0”
“2015.01.07 10:06:28 UTC”、“2015.01.07 10:06:35 UTC”、“2015.01.07 10:13:28 UTC”、“背书PendingImpressionEvent-c1201ed1-fb23-4”、“卡夫卡特尔”、“马拉松”、“成功”、“1”、“1”、“1”

tr-d'[]'new\u文件

tr-d'[]'new\u文件

将所有东西都放在外部支架内，如果没有外部支架，则保持不变

将所有内容都放在外部括号内，如果没有外部括号保持不变，如果行中有更多的

或

怎么办？如果行中有更多的

或

怎么办？您可能需要

sed的s/^\[/；s/\]，$/'文件，因为您正在删除最后一个，
是的，op不想要尾随的，
（包括逗号）：/我以为这是句子中他不想删除的部分：P建议编辑您可能需要sed的/^\[/；s/\]，$/，/'文件
，因为您正在删除最后一个，
是的，op不想要尾随的，
（包括逗号）：/I我认为这是句子的一部分，而不是他想删除的部分：P建议编辑