Shell 从测试套件中删除大量测试

我有一个大约8000个测试的列表，我想从中删除某些测试子集（多达432个），这些测试现在已经通过，不再需要

测试条目采用以下格式

  [img.hidden: setAttribute() to "" followed by IDL get]
    expected: FAIL

  [img.hidden: setAttribute() to " foo " followed by IDL get]
    expected: FAIL

img.hidden: setAttribute() to "" followed by IDL get
img.hidden: setAttribute() to " foo " followed by IDL get

我想删除的测试列表以以下格式提供给我

  [img.hidden: setAttribute() to "" followed by IDL get]
    expected: FAIL

  [img.hidden: setAttribute() to " foo " followed by IDL get]
    expected: FAIL

img.hidden: setAttribute() to "" followed by IDL get
img.hidden: setAttribute() to " foo " followed by IDL get

搜索和删除所有这些测试的条目的快速方法是什么

编辑： 请举例说明。参赛作品：

  [img.hidden: IDL set to undefined followed by hasAttribute()]
    expected: FAIL

  [img.hidden: IDL set to undefined followed by IDL get]
    expected: FAIL

  [img.hidden: IDL set to null followed by hasAttribute()]
    expected: FAIL

  [img.hidden: IDL set to null followed by IDL get]
    expected: FAIL

  [img.hidden: IDL set to 7 followed by IDL get]
    expected: FAIL

删除列表：

img.hidden: IDL set to undefined followed by IDL get
img.hidden: IDL set to null followed by IDL get
img.hidden: IDL set to 7 followed by IDL get

预期产出：

  [img.hidden: IDL set to undefined followed by hasAttribute()]
    expected: FAIL

  [img.hidden: IDL set to null followed by hasAttribute()]
    expected: FAIL

---

您可以使用

sed

作为

while read pattern
do 
sed -i "/$pattern/, +1 d" inputFile
done < test

读取模式时
做
sed-i“/$pattern/，+1d”输入文件
完成<测试

测试

$ cat input
[img.hidden: IDL set to undefined followed by hasAttribute()]
    expected: FAIL

[img.hidden: IDL set to undefined followed by IDL get]
    expected: FAIL

[img.hidden: IDL set to null followed by hasAttribute()]
    expected: FAIL

[img.hidden: IDL set to null followed by IDL get]
    expected: FAIL

[img.hidden: IDL set to 7 followed by IDL get]
    expected: FAIL

$ cat test
img.hidden: IDL set to undefined followed by IDL get
img.hidden: IDL set to null followed by IDL get
img.hidden: IDL set to 7 followed by IDL get

$ while read pattern; do sed -i "/$pattern/, +1 d" input ; done < test

$ cat input
[img.hidden: IDL set to undefined followed by hasAttribute()]
    expected: FAIL


[img.hidden: IDL set to null followed by hasAttribute()]
    expected: FAIL

$cat输入
[img.hidden:IDL设置为未定义，后跟hasAttribute（）]
预期：失败
[img.hidden:IDL设置为未定义，后跟IDL get]
预期：失败
[img.hidden:IDL设置为null，后跟hasAttribute（）]
预期：失败
[img.hidden:IDL设置为null，后跟IDL get]
预期：失败
[img.hidden:IDL设置为7，后跟IDL get]
预期：失败
$cat测试
img.hidden:IDL设置为未定义，后跟IDL get
img.hidden:IDL设置为null，后跟IDL get
img.hidden:IDL设置为7，后跟IDL get
$while读取模式；执行sed-i“/$pattern/，+1d”输入；完成<测试
$cat输入
[img.hidden:IDL设置为未定义，后跟hasAttribute（）]
预期：失败
[img.hidden:IDL设置为null，后跟hasAttribute（）]
预期：失败

这似乎不起作用，奇怪的是，它将我的输入文件变成了8000个空行。这是问题的一个例子。事实证明，将模式从

“/$pattern/，+1d”

更改为

“/$pattern/，+1d”

就可以解决问题。想知道为什么。。。？此外，每次测试后删除空行时，应改为

+2d

。@aakashjain空格不能有任何区别。两者都试过了，效果很好+2将删除2条线路+1对于试验后没有空行的情况可能更安全case@aakashjain不管怎样，我很高兴听到这有帮助：）