[Lldb-commits] [PATCH] D116491: [lldb] Compute fully qualified command names in FindCommandsForApropos

[Jonas Devlieghere via Phabricator via lldb-commits]

JDevlieghere added inline comments.


================
Comment at: lldb/source/Interpreter/CommandInterpreter.cpp:2864
+      for (const auto &subcommand_name : subcommands_found) {
+        auto qualified_name = (command_name + " " + subcommand_name).str();
+        commands_found.AppendString(std::move(qualified_name));
----------------
kastiglione wrote:
> JDevlieghere wrote:
> > It's not obvious what type `qualified_name` is. I would either do:
> > 
> > ```
> >         std::string qualified_name = (command_name + " " + subcommand_name).str();
> > ```
> > 
> > or 
> > 
> > ```
> >         auto qualified_name = std::string(command_name + " " + subcommand_name);
> > ```
> As you've given feedback on use of `auto` before, should we document some standards around `auto`? It seems to vary in the code, and from person to person. 
> 
> I'm cool with explicit `std::string`, but I think a few bits of context here do indicate it's a string type (ex use of `.str()`, use of `+` and a string literal, being passed to `AppendString()`). I'm thinking that some of these could be documented as either sufficient or insufficient for `auto`.
> 
> Somewhat separately, what do you think about adding `AppendString(const Twine&)` (D116682) and then making this:
> 
> ```
> commands_found.AppendString(command_name + " " + subcommand_name);
> ```
The LLVM Coding Standards have a section on `auto`: https://llvm.org/docs/CodingStandards.html#use-auto-type-deduction-to-make-code-more-readable. 

If it were any other type, I might agree with your argument, but we have a bunch of "string" types (`std::string`, `llvm::StringRef` and `lldb_private::ConstString` being the usual suspects) making it non-obvious. 


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D116491/new/

https://reviews.llvm.org/D116491