Todd Lipcon has posted comments on this change.

Change subject: log: refactor log reading into an iterator-like interface
......................................................................


Patch Set 2:

(5 comments)

http://gerrit.cloudera.org:8080/#/c/3038/2/src/kudu/consensus/log_util.cc
File src/kudu/consensus/log_util.cc:

Line 187:     return s.CloneAndPrepend(Substitute("Error reading from log $0", 
seg_->path_));
> Nit: lower-case for Error? This will probably be appended to another Status
k, edited this one since i'm editing the code anyway


Line 223: // TODO: can remove some params here?
> What is this referring to? There's just the status param.
yea, left over from an intermediate step of the refactoring, thanks


Line 231:   for (const auto& r : recent_entries_) {
> It's a fair amount of work to maintain recent_entries_ and I see it's just 
yea, I agree it's a fair bit of work, but we added this a while back because 
it's super helpful when debugging various kinds of issues in log 
corruption/replay. One of those things that isn't needed 99% of the time, but 
the 1% of the time when you need it, it's super helpful, so I'm inclined to 
keep it.


http://gerrit.cloudera.org:8080/#/c/3038/2/src/kudu/consensus/log_util.h
File src/kudu/consensus/log_util.h:

Line 130:   int64_t batches_read_;
> Nit: num_batches_read_
Done


Line 201:   std::unique_ptr<LogEntryReader> NewLogEntryReader();
> Curious why you picked this approach to constructing a LogEntryReader inste
Done


-- 
To view, visit http://gerrit.cloudera.org:8080/3038
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I5727951664600b7a591a6c892ea1e62469ceb109
Gerrit-PatchSet: 2
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <[email protected]>
Gerrit-Reviewer: Adar Dembo <[email protected]>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <[email protected]>
Gerrit-HasComments: Yes

Reply via email to