I've got my pipeline up and running, but I need to save a number to a file, which will allow me to come back to previous processing in case something goes wrong ( or when I want to deploy new version of pipeline, and I cannot use the update-pipeline feature ).
我已经启动并运行了我的管道，但我需要将一个数字保存到一个文件中，这将允许我回到以前的处理，以防出现问题（或者当我想部署新版本的管道时，我无法使用更新管道功能）。
I was trying to add libraries from google-cloud-storage, but there are errors in Runtime (it does not find some classes from google-cloud-core even when I have all dependencies up to date ). There need to be a simple way, just to save ( and then read ) a file. I will be grateful for some snippet or link.
我试图从google-cloud-storage添加库，但运行时有错误（即使我最新的所有依赖项，它也没有找到google-cloud-core的一些类）。需要一种简单的方法，只需保存（然后读取）一个文件。我会感谢一些片段或链接。
1 个解决方案
#1
0
If you are stuck in dependency hell, and must do this step from within dataflow, then I would recommend looking at side-inputs and outputs...and then pairing that with some simple TextIO operations.
如果您陷入依赖地狱，并且必须从数据流中执行此步骤，那么我建议查看侧输入和输出...然后将其与一些简单的TextIO操作配对。
#1
0
If you are stuck in dependency hell, and must do this step from within dataflow, then I would recommend looking at side-inputs and outputs...and then pairing that with some simple TextIO operations.
如果您陷入依赖地狱，并且必须从数据流中执行此步骤，那么我建议查看侧输入和输出...然后将其与一些简单的TextIO操作配对。