Todd Lipcon has posted comments on this change.

Change subject: WIP: KUDU-1812. redaction of PB fields
......................................................................


Patch Set 2:

(8 comments)

http://gerrit.cloudera.org:8080/#/c/5553/2/src/kudu/util/pb_util.cc
File src/kudu/util/pb_util.cc:

Line 551:   virtual string PrintFieldName(const Message& message,
> Don't need virtual and override, right?
Done


PS2, Line 552:                         const Reflection* reflection,
             :                         const FieldDescriptor* field) const 
override {
> Nit: indentation.
Done


Line 560:     if (hide_next_string_) {
> Shouldn't we reset hide_next_string_ to false after redacting? Or does it n
Yea, it doesn't make a difference because we always print a new field name 
before printing some other field, but I'll reset anyway because it looks more 
natural.


Line 593:   if (debug_string.size() > 0 &&
> warning: the 'empty' method should be used to check for emptiness instead o
Done


http://gerrit.cloudera.org:8080/#/c/5553/2/src/kudu/util/pb_util.h
File src/kudu/util/pb_util.h:

Line 102: std::string SecureDebugString(const google::protobuf::Message& 
message);
> warning: function 'kudu::pb_util::SecureDebugString' has a definition with 
Done


Line 103: std::string SecureShortDebugString(const google::protobuf::Message& 
message);
> warning: function 'kudu::pb_util::SecureShortDebugString' has a definition 
Done


http://gerrit.cloudera.org:8080/#/c/5553/2/src/kudu/util/pb_util.proto
File src/kudu/util/pb_util.proto:

Line 43:   optional bool REDACT = 50001 [default=false];
> I don't think changing the field # would be a breaking change, since this i
I think although it's not "in-house", it's also not a "public application" in 
the sense that others would be directly interfacing with our protobuf files 
from other applications. I think this warning is mostly meant to discourage 
people implementing language bindings from assigning some special option like 
'erlang_generator_do_foo' to an extension number that application programmers 
might want to use, etc.

Furthermore, I agree that since it's not serialized in any way, we can always 
change it later.


http://gerrit.cloudera.org:8080/#/c/5553/2/src/kudu/util/proto_container_test.proto
File src/kudu/util/proto_container_test.proto:

Line 29: // TODO move me to another file
> Nit: update this?
Done


-- 
To view, visit http://gerrit.cloudera.org:8080/5553
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Ie07bfdcbc33d38d55315faae2c4906899a5450fb
Gerrit-PatchSet: 2
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Todd Lipcon <t...@apache.org>
Gerrit-Reviewer: Adar Dembo <a...@cloudera.com>
Gerrit-Reviewer: Dan Burkert <danburk...@apache.org>
Gerrit-Reviewer: Jean-Daniel Cryans <jdcry...@apache.org>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <t...@apache.org>
Gerrit-HasComments: Yes

Reply via email to