[Lldb-commits] [PATCH] D121844: Applying clang-tidy modernize-use-equals-default over LLDB

[Pavel Labath via Phabricator via lldb-commits]

labath added inline comments.


================
Comment at: lldb/source/Core/Value.cpp:667
 
-const ValueList &ValueList::operator=(const ValueList &rhs) {
+const ValueList &ValueList::operator=(const ValueList &rhs) {  // NOLINT(modernize-use-equals-default)
   m_values = rhs.m_values;
----------------
shafik wrote:
> labath wrote:
> > shafik wrote:
> > > I have to look into why we return a `const &` here. We do this in a few other places too.
> > I don't think there's a good reason for that. Most people aren't aware that built-in assignment operators return lvalues. And some of the people who are aware of that think that it's a bad idea, so they make sure their operators don't do it...
> It produced build errors, so some of the users are relying on this. I didn't want to plumb into this since it was orthogonal to the change.
I was curious to see what kind of errors could be produced by that change -- I didn't get any, so I committed (c484857b2e77721a4235b0e2d53d335c09fc6af3) my version. :)


================
Comment at: lldb/source/Host/macosx/cfcpp/CFCMutableArray.cpp:18
 CFCMutableArray::CFCMutableArray(const CFCMutableArray &rhs)
-    : CFCReleaser<CFMutableArrayRef>(rhs) // NOTE: this won't make a copy of the
+    : CFCReleaser<CFMutableArrayRef>(rhs) // NOLINT(modernize-use-equals-default)
+                                          // NOTE: this won't make a copy of the
----------------
shafik wrote:
> labath wrote:
> > Why suppress this?
> I wanted to preserve the comment since someone thought it was important.
In that case, maybe you can just put the comment next to the `=default`  clause. Now, this is beginning to look like there is something very subtle going on -- subtle enough to confuse the clang-tidy check. Which, of course, isn't true...


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D121844/new/

https://reviews.llvm.org/D121844