keith-turner commented on code in PR #5240:
URL: https://github.com/apache/accumulo/pull/5240#discussion_r1909215790
##########
core/src/main/java/org/apache/accumulo/core/metadata/StoredTabletFile.java:
##########
@@ -292,26 +292,26 @@ public TabletFileCq(Path path, Range range) {
}
static class TabletFileCqMetadataGson {
- private String path;
- private byte[] startRow;
- private byte[] endRow;
+ protected String metadataEntry;
+ protected String path;
+ protected byte[] startRow;
+ protected byte[] endRow;
Review Comment:
> @keith-turner - I guess the question is whether or not the convenience of
having the file path in an easier to read format in the external compaction
json is worth having it in addition to the metadata entry. If not, then we
should revert https://github.com/apache/accumulo/pull/5237. If it is worth it,
then I can try and clean this up.
I think the extra data may make it hard to read. I was looking at the
serialized compaction data in the manager logs w/ compaction bugs that caused
thousands of files to build up per tablet. Having more data for that case may
make it more confusing.
> Could we write a tool or something to make scanning and reading the entry
easier?
Yeah that seems good. The data in the metadata table should probably be
compact and correct (so it would only have the escaped data) and maybe we can
have tooling to make the data more readable in logs and provide some mechanism
to make the data more readable when scanning the metadata table.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
