[jcelerier]

just opened a file and...ugh

https://android.googlesource.com/platform/art/+/refs/heads/m...

like, just this, basically C:

    if (strncmp(argv[arg_idx], "-XXlib:", strlen("-XXlib:")) == 0) {
       ...

is just so complicated and error-prone over

    if(std::string_view{argv[arg_idx]}.starts_with("-XXlib:"))  
       ...

(which I guess is the intent)

[jmgao]

That code is older than C++11, let alone C++17. Also, your proposal isn't as efficient, since it calls strlen on construction of string_view (not that cost of parsing command line arguments is going to be measurable).

[jjnoakes]

> your proposal isn't as efficient, since it calls strlen on construction of string_view

The code it replaces also had a strlen in it.

[jmgao]

A strlen of a compile-time constant, which will certainly get turned into a constant by the compiler. It does not get the length of the input string.

[jjnoakes]

Why wouldn't the string_view constructor be inlined and get the same treatment?

[pjmlp]

Certainly isn't a fact unless proven by examining the generated machine code.

[jmgao]

This is a worthless statement. Do you examine all of your generated machine code to ensure that it generates correct code? Any compiler that you cannot trust to inline 3 for strlen("foo") is not one you should trust to compile anything, let alone modern C++.

[maccard]

> Do you examine all of your generated machine code to ensure that it generates correct code?

(not OP). I examine the output if someone tells me it's slower, or if I care about performance. I did a very rough benchmark which says in this case they're equivalent [0], however that's not necessarily representative of the places it's called in. There's definitely a compile time overhead though [1]. On my current project there's ~20k files. On an 8 core machine (as an example), adding 150ms per file would add 7 minutes onto a clean build (before optimising the builds).

[0] https://quick-bench.com/q/LBarCDIigwpmXcwZSx-RgV2x4_4 [1] https://build-bench.com/b/OH8qf9AgEB_l-BhANV5qP_uSCys

[pjmlp]

ISO C++ abstract machine and what a random C++ compiler generates out of it, don't require at any moment the realisation of your assertion.

So you either validate it actually takes place or believe in fairy tales.

[ncmncm]

The code is ancient, and Google (like most) sees no value in modernizing old code, particularly for non-revenue-generating product. Furthermore, Google long ago locked themselves into an idiotically anti-RAII coding standard.

So, the example is of bad old code, at a place that enforces bad coding practice. The intent appears to be to suggest this is typical practice, which is not supported.

[pjmlp]

Google seats at ISO C++, is a major contributor to clang and LLVM, and belongs to the group of companies with largest C++ code in production.

If this what a company with deep roots in C++ world is doing, what are the large majority of "dark matter" software factories doing, specially the sweetshop ones.

[ncmncm]

A minuscule fraction of the people coding what they call C++ at Google come to the committee meetings. Google's ancient, idiotic prohibition on RAII is not a thing the current attendees, or other employees, have a say about. Many, attending or no, would rather be coding sensibly. And, in any case, the ancient code you present is not how code is being written even at Google today. So, your point is hopelessly muddled.

There are shops where people still write new C++ like it's 2011, or 2003, or 1998, or 1992, or C. There are plenty of shops where people code the best way their current production compiler allows. There are plenty where different people do some of each of those. Vanishingly few shops make an effort to rewrite ancient code according to current best practice.

I guess a sweetshop company is one with Oompa-loompas.

[pjmlp]

One day I would appreciate to know where that perfect C++ world of yours exist in real world.

I guess I need to catch up with some Twilight Zone episodes.

[ncmncm]

Try "Willy Wonka and the Chocolate Factory"?

[still_grokking]

> just opened a file and...ugh

Looks just like regular PHP. So what's the point? Or did I miss something?

/s