在模式后匹配单词并使用 sed 在文件中附加后缀？

Question

I have multiple files like this one:

CREATE TABLE VAI_TEST
   (
      TEST VARCHAR2(4000)
   ) ;

Only the first line is relevant. I need to replace VAI_TEST with VAI_TEST_TEMP. The result should be like this:

CREATE TABLE VAI_TEST_TEMP
   (
      TEST VARCHAR2(4000)
   ) ;

I've found multiple regex pattern that match VAI_TEST eg

• (?:CREATE TABLE )([A-Za-z_]*)[\\r\\n]
• (?:.* TABLE )(.*$)
• CREATE TABLE (.*)\\n
• (?:.* TABLE )(.*)\\n

Probably there are more and better patterns. My problem is to replace the content of capture group 1 with the content and the suffix.

I've tried multiple sed combinations but mostly nothing happens or I get an error for a wrong regex pattern (with -r flag) or invalid reference \\1 (with -e) flag.

In the end I want to use the command in a bash script. Additionally it would be helpful if the new name could be saved in a variable to use it in further steps.

What is a good way to achieve this?

Answer 1

sed '/^CREATE TABLE/s/[^ ]*$/&_TEMP/' file

• /^CREATE TABLE/ - on lines beginning with "CREATE TABLE",
• s/[^ ]*$/ - replace non-space characters ( [^ ] ) before the endline ( $ ) with:
• &_TEMP - & -- whatever those characters were-- suffixed with "_TEMP"

Answer 2

Consider

sed -i 's/CREATE TABLE [^[:space:]]\{1,\}$/&_TEMP/' file

[^[:space:]] is for any symbol other than whitespace and \\{1,\\} repeats it one or more times. & stands for the