在jenkins管道上，我执行：
def devicesText = sh(script: "xcrun xctrace list devices", returnStdout: true)
printf devicesText
它具有奇怪的行为，因为它在控制台上打印输出，但不将输出保存在变量中。控制台如下(ID已被XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX替换(：
+ xcrun xctrace list devices
== Devices ==
SAM Team - Build Server 2018 (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
== Simulators ==
Apple Watch Series 5 - 40mm (7.0) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
Apple Watch Series 5 - 44mm (7.0) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPad (7th generation) (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPad Air (3rd generation) (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPad Pro (11-inch) (2nd generation) (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPad Pro (12.9-inch) (4th generation) (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPad Pro (9.7-inch) (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPhone 11 (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPhone 11 Pro (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPhone 11 Pro Max (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPhone 6s (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPhone 8 (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPhone 8 Plus (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
iPhone SE (2nd generation) (13.5) (XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX)
[Pipeline] echo
正如您所看到的，echo是空的，因为它没有被复制到变量中。如何将输出保存在变量中？
(它曾经使用不推荐使用的命令instruments -s devices(
不知何故，xcrun xctrace将输出打印到stderr。
因此，为了在stdout中获得stderr日志，只需使用2>&1将它们重定向到stdout即可。
->xcrun xctrace list devices 2>&1 | grep -oE 'iPhone.*?[^(]+' | head -1 | awk '{$1=$1;print}'
我在使用GitHub Actions的xcodebuild中遇到了类似的问题。我的工作是找到一个带有xcrun xctrace列表设备的设备，并直接在build命令ex:-destination中使用它；platform=$platform，name＝iPhone 8"；。
我试着使用默认的regEx和grep 1行匹配的iPhone*，但它从来没有对我起作用。