Zakelly commented on code in PR #25030:
URL: https://github.com/apache/flink/pull/25030#discussion_r1667911678
##########
flink-runtime/src/main/java/org/apache/flink/runtime/checkpoint/filemerging/FileMergingSnapshotManagerBase.java:
##########
@@ -773,6 +773,13 @@ static boolean
shouldSyncAfterClosingLogicalFile(FileSystem fileSystem) {
return true;
}
+ private static String uriEscape(String input) {
+ // All reserved characters (RFC 2396) will be removed. This is enough
for flink's resource
+ // id, job id and operator id.
+ // Ref:
https://docs.oracle.com/javase/8/docs/api/index.html?java/net/URI.html
+ return input.replaceAll("[;/?:@&=+$,\\[\\]]", "-");
+ }
Review Comment:
1. I did some investigation. There are some utilities for URI from
third-party dependencies, but the behavior is not same as what we want. Instead
of removing or replacing the reserved characters, they may treat the reserved
characters as valid ones. The `java.net.URLEncoding` seems feasible, however,
URL specification is a little bit different from the URI. So I'd prefer doing a
simple replace, since the ids from flink is neat.
2. Well no, we never did that. We tried Per-Job on yarn and local
mini-cluster (E2E). I'm trying the local standalone cluster.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@flink.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
