regex - 解析 git 历史记录并对其进行模式匹配，一旦找到匹配项，就在变量中捕获附近的值

Question

我有很长的 git 历史，我想找到“build(mariadbstuff): Copybara import”的第一个实例，然后在变量中捕获其关联的 GitOrigin-RevId“d7b349b6b6db24047ae32a1618dd5f5544864225”。

我有一种非常hacky，丑陋的方式，我希望有一种更清洁的方式。这是完成这项工作的正确但笨拙且丑陋的 bash cmd（尽管它不会删除 GitOriginRevId 部分）：

sed -e '/build(mariadbstuff)/,/GitOriginRevId/!d' <(git log) > outputfile.txt && sed -n 11p outputfile.txt

---

... // more contents of git history
 
build(mariadbstuff): Copybara import
    
    Squashed commit of the following:
    
    commit d8829cc7df7a6813f93d224916b10e7c148950be
    Author: Copybara <[e-mail]>
    Date:   Fri Jul 10 10:26:41 2020 -0400
    
        Project import generated by Copybara.
    
        GitOrigin-RevId: d7b349b6b6db24047ae32a1618dd5f5544864225

... // more contents of git history

标签： regexgitawksedgrep

Answer 1

您能否尝试在 GNU 中使用显示的示例进行跟踪、编写和测试awk。此解决方案尽快退出 Input_file，它会获得 OP 所需的 id，并且不会读取整个 Input_file，这将比读取整个 Input_file 解决方案更快。

awk '
/build\(mariadbstuff\): Copybara/{
  found=1
  next
}
found && /GitOrigin-RevId:/{
  print $NF
  exit
}
'  Input_file

说明：为上述添加详细说明。

awk '                                    ##Starting awk program from here.
/build\(mariadbstuff\): Copybara/{       ##Checking condition if a line contains string build\(mariadbstuff\): Copybara escaping ( and ) in it to treat it as literal character here.
  found=1                                ##Setting found variable to 1 here.
  next                                   ##next will skip all further statements from here.
}
found && /GitOrigin-RevId:/{             ##Checking condition if found is SET and line has string GitOrigin-RevId: then do following.
  print $NF                              ##Printing last column of current line here.
  exit                                   ##exiting from program from here to save sometime :)
}
'  Input_file                            ##Mentioning Input_file name here.