Closed rhansen closed 8 years ago
I've reviewed up to and including e64be6e, feel free to merge up to that point after addressing my comments. (I intend to finish reviewing the other commits later; I won't get to them tonight.)
(nit) Why are the "use * type instead of int for " commits not next to the "use an enum for " commits in the graph?
I've finished reviewing this, up to and including b954107. Overall it looks great!
(nit) Why are the "use * type instead of int for " commits not next to the "use an enum for " commits in the graph?
All of the commits between those two batches are preparing for the "use * type instead of int for *" commits. Some of those commits depend on the new ERR_SCM_UNSPECIFIED
error code, and I wanted to add that after converting the error codes to an enum so that I wouldn't need to renumber all of the #define
s to make room for a new -1
.
On 2016-02-09 17:57, Richard Hansen wrote:
(nit) Why are the "use * type instead of int for " commits not next to the "use an enum for " commits in the graph?
All of the commits between those two batches are preparing for the "use * type instead of int for *" commits. Some of those commits depend on the new ERR_SCM_UNSPECIFIED error code, and I wanted to add that after converting the error codes to an enum so that I wouldn't need to renumber all of the #defines to make room for a new -1.
That's a good reason. Sorry, my eyes must have glazed over too much to notice that.
I pushed a bunch of untested fixups that I believe address your comments except for your configure.ac
comment. There are additional issues with configure.ac
that I noticed, and it's easiest to deal with those in a rebase.
I'll test these untested fixups at some point. :smiley:
I've reviewed all but the last of your fixups (8026c94^..6c8499a), and they look good.
Rebased and ready for re-review. Aside from autosquashing, the only changes are in the commits affecting configure.ac
. Only the commits touching configure.ac
should be new to you.
Merge it so!
see #27