libpq VS spdlog

Scattered files: In CMake, it's quite common for the build script to be split over multiple files. Most commonly it's a top-level CMakeLists.txt, then another for each subdirectory (often corresponding to a target), plus some utilities in a cmake/ subdirectory. But that is nothing compared to build2. Just look at the build script repo for libpq. What on earth is going on? There are a few top-level files, plus a submodule for the upstream (fair enough), but then absolutely heaps of subdirectories mirroring pq directories, plus softlinks into the upstream submodule. There are others that actually look worse to me but I don't feel comfortable linking to them because I'm not sure what's going on!

Why a new lib instead of using or contributing to an existing one as spdlog?

https://github.com/gabime/spdlog

GitHub repo: gabime/spdlog

SpdLog https://github.com/gabime/spdlog

The usual recommendation when logging libraries are brought up is spdlog, which is however header-only. It's available on Conan-center.

* Aha! In digging up the docs for NDC, I found this[1], which does mention a book for your reading list: "Patterns for Logging Diagnostic Messages" part of the book "Pattern Languages of Program Design 3" edited by Martin et al.

[1] https://github.com/gabime/spdlog

FetchContent_Declare( spdlog GIT_REPOSITORY https://github.com/gabime/spdlog GIT_TAG origin/v1.x ) FetchContent_MakeAvailable(spdlog)

Typically, you'd use a logging library to handle stuff like this. I personally like spdlog. You use different logger functions (info, warn, error) and depending on what level you have set for the logger (or globally) some of the functions become no-ops. E.g. When not running in verbose mode all spdlog::info() do nothing.