[GitHub] [hadoop] steveloughran commented on pull request #4220: HADOOP-18216. io.file.buffer.size must be greater than zero

steveloughran commented on PR #4220: URL: https://github.com/apache/hadoop/pull/4220#issuecomment-084200 oh, and the ioutil code would be easy to test, wouldn't it, pass in null stream values and a negative offset and expect the precondition failures, not any null pointer exception

[GitHub] [hadoop] steveloughran commented on pull request #4220: HADOOP-18216. io.file.buffer.size must be greater than zero

steveloughran commented on PR #4220: URL: https://github.com/apache/hadoop/pull/4220#issuecomment-082405 could have some private method in ioutils to get and validate the value, use from both places it is called. filesystem could have its own copy of the check -- This is an automated

[GitHub] [hadoop] steveloughran commented on pull request #4220: HADOOP-18216. io.file.buffer.size must be greater than zero

steveloughran commented on PR #4220: URL: https://github.com/apache/hadoop/pull/4220#issuecomment-1110139088 @Hexiaoqiao yes, you can add a hadoop.util.Precondition check where it is loaded, but do it in a new PR. let's merge this doc change in first, as it self contained and low risk to

[GitHub] [hadoop] steveloughran commented on pull request #4220: HADOOP-18216. io.file.buffer.size must be greater than zero

steveloughran commented on PR #4220: URL: https://github.com/apache/hadoop/pull/4220#issuecomment-1109750693 i am about to merge this. should i give the credit to Jingxuan Fu? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub