unpopular opinion: Rust: not a fan
-
I don't know why people like it. It's like they took a perfectly good language (C) and pythonized it. No, thankfully it doesn't use significant whitespace, but I get the impression that whoever designed Rust actually hates the C language family and wants bad things to happen to them. You can keep this nonsense. If I wanted training wheels, I'd use VB. Joy. Yet another language I get to learn enough of simply to port things away into a proper language. Edited to add: anybody who designs a grammar with this construction "fn main()" needs to have their compiler taken away and forced to use scripting languages until they can prove to the world that they can use context free grammars properly
To err is human. Fortune favors the monsters.
-
I don't know why people like it. It's like they took a perfectly good language (C) and pythonized it. No, thankfully it doesn't use significant whitespace, but I get the impression that whoever designed Rust actually hates the C language family and wants bad things to happen to them. You can keep this nonsense. If I wanted training wheels, I'd use VB. Joy. Yet another language I get to learn enough of simply to port things away into a proper language. Edited to add: anybody who designs a grammar with this construction "fn main()" needs to have their compiler taken away and forced to use scripting languages until they can prove to the world that they can use context free grammars properly
To err is human. Fortune favors the monsters.
-
I was in total agreement until you wrote:
honey the codewitch wrote:
a perfectly good language (C)
Neither perfect nor good, IMO.
If you think 'goto' is evil, try writing an Assembly program without JMP.
Well I was being generous. I prefer C++ myself.
To err is human. Fortune favors the monsters.
-
I started reading the rust documentation -- in the first section it says that by default variables are immutable. Hokay ... it appears the inventors of the language didn't know what "variable" means and didn't know how to type "dictionary.com" in their browser. I didn't feel the need to read further ...
BryanFazekas wrote:
in the first section it says that by default variables are immutable. Hokay ... it appears the inventors of the language didn't know what "variable" means
One rule I remember from my student days: "Constants ain't, variables won't". I have experienced the truth of this rule quite a few times.
-
Greg Utas wrote:
Talk about a culture clash. C++ would make far more sense, but maybe the Linux crowd also feel that C has abused them
It won't happen: Linus hates C++. He thinks that C++ is like a box of chocolates: you never know what you are gonna get.
...
one = 1;
two = 2;
x = one + two;In C++ '+' can be redefined to do whatever you want. I see his point but any language that doesn't allow you to shoot yourself in the foot occasionally is probably too boring. In the words of Spider-Man: "with great power comes great responsibility" :)
Mircea
Mircea Neacsu wrote:
In C++ '+' can be redefined to do whatever you want.
I never programmed Forth myself, but a former coworker who was a Forth worshipper (quoting the Scriptures: "Go Forth, and Multiply") showed med how to really obfuscate code: Forth lets you redefine any token. So he redefined '17' to have the value of '5'. That brings the concept of 'off by one' to new levels :-)
-
Well I was being generous. I prefer C++ myself.
To err is human. Fortune favors the monsters.
I can respect that. In my current and previous roles in the financial and music industries (respectively), both immersed in the Microsoft world with no need for hardware-level programming, I prefer C#. But if I had the opportunity to do hardware-level stuff again my preference would probably shift to Assembly. I love speaking CPU.
If you think 'goto' is evil, try writing an Assembly program without JMP.
-
Mircea Neacsu wrote:
In C++ '+' can be redefined to do whatever you want.
I never programmed Forth myself, but a former coworker who was a Forth worshipper (quoting the Scriptures: "Go Forth, and Multiply") showed med how to really obfuscate code: Forth lets you redefine any token. So he redefined '17' to have the value of '5'. That brings the concept of 'off by one' to new levels :-)
I had a love affair with Forth. When I read Leo Brodie's "Starting Forth" book I fell so much in love with it that I wrote my own interpreter (in MACRO11 - the PDP11 assembler no less). Talk about being young and foolish :laugh:
Mircea
-
I can respect that. In my current and previous roles in the financial and music industries (respectively), both immersed in the Microsoft world with no need for hardware-level programming, I prefer C#. But if I had the opportunity to do hardware-level stuff again my preference would probably shift to Assembly. I love speaking CPU.
If you think 'goto' is evil, try writing an Assembly program without JMP.
I like the challenge of developing highly usable type safe interfaces that resolve to really efficient assembly like you'd have typed by hand. A recent endeavor had me bitbanging the SPI hardware registers on an ESP32 MCU to get faster performance. It was all microsecond sensitive, requiring me to have total control over what got inlined and not. I had to use constexpr everywhere to forward as much of the configuration, like pin assignments off to compile time evaluations. It replaced a nasty set of C preprocessor macros that accomplished the same. Byte for byte it generates the same assembly instructions. The guts of this file are not clean, but what it presents to the user of this header *is* htcw_tft_io/tft_spi.hpp at master · codewitch-honey-crisis/htcw_tft_io · GitHub[^]
To err is human. Fortune favors the monsters.
-
I like the challenge of developing highly usable type safe interfaces that resolve to really efficient assembly like you'd have typed by hand. A recent endeavor had me bitbanging the SPI hardware registers on an ESP32 MCU to get faster performance. It was all microsecond sensitive, requiring me to have total control over what got inlined and not. I had to use constexpr everywhere to forward as much of the configuration, like pin assignments off to compile time evaluations. It replaced a nasty set of C preprocessor macros that accomplished the same. Byte for byte it generates the same assembly instructions. The guts of this file are not clean, but what it presents to the user of this header *is* htcw_tft_io/tft_spi.hpp at master · codewitch-honey-crisis/htcw_tft_io · GitHub[^]
To err is human. Fortune favors the monsters.
-
That's some sweet code there, and a great use-case for C. But gawd how I hate pointers. :laugh:
If you think 'goto' is evil, try writing an Assembly program without JMP.
I'm a weirdo, but I feel like i have a knack for pointers. they don't intimidate me at all and i almost never cause AVs with them in C++. I'm much more likely to make bad things happen like that when I'm P/Invoking from C# just because i have less of an immediate idea of what marshalling is happening.
To err is human. Fortune favors the monsters.
-
Mircea Neacsu wrote:
In C++ '+' can be redefined to do whatever you want.
I never programmed Forth myself, but a former coworker who was a Forth worshipper (quoting the Scriptures: "Go Forth, and Multiply") showed med how to really obfuscate code: Forth lets you redefine any token. So he redefined '17' to have the value of '5'. That brings the concept of 'off by one' to new levels :-)
-
All they did was make C/C++
constthe default so that you have use the equivalent ofmutableif it's not constant.Robust Services Core | Software Techniques for Lemmings | Articles
The fox knows many things, but the hedgehog knows one big thing.Greg Utas wrote:
All they did was make C/C++ const the default so that you have use the equivalent of mutable if it's not constant.
Let's think about this. In a typical program I declare dozens of variables and a handful of constants, and of those variables, probably 0.01% need to be immutable once declared. So the Rust designers flip that, requiring me to explicitly specify that a standard entity in all programming actually behaves as expected, and the default behavior is an edge case? It's a solution in search of a problem and failing to find one. OTOH, each new release of C# contains new syntax that is more cryptic, and those designers crow, "with this new syntax you save typing TTWWWWOOOOOOO characters!" like it's the greatest thing in the world. Clowns to the left of me, jokers to the right ....
-
I don't know why people like it. It's like they took a perfectly good language (C) and pythonized it. No, thankfully it doesn't use significant whitespace, but I get the impression that whoever designed Rust actually hates the C language family and wants bad things to happen to them. You can keep this nonsense. If I wanted training wheels, I'd use VB. Joy. Yet another language I get to learn enough of simply to port things away into a proper language. Edited to add: anybody who designs a grammar with this construction "fn main()" needs to have their compiler taken away and forced to use scripting languages until they can prove to the world that they can use context free grammars properly
To err is human. Fortune favors the monsters.
It might be that Rust is designed to solve different problems than the ones you need to solve? Though it's okay to just flat out dislike it as well. If you add up the billions of dollars of damage caused by good C programmers with decades of experience making mistakes that Rust would have disallowed - things like Heartbleed - you can see the motivation to use a language like Rust. Like it or not, year in and year old, a huge percentage of bugs and vulnerabilities in software are caused by memory safety issues. It's certainly not the best choice for every problem. I can't say I *love* writing Rust. But I think it's a good choice for you need C-like performance but can't afford vulnerabilities. Say you're implementing something like OpenSSL - I'd argue that if you're starting a project like that in 2022, you'd be crazy to pick C (or even C++) over Rust. I suppose you could pick something like MISRA C or even Ada as well, but I don't think either of those would be more joyful to write than Rust. For certain classes of problems, the only safe assumption you should make is that you do need training wheels because no matter how much experience you have or how good you are, eventually you will make a costly mistake. For most other applications, Rust probably isn't the best choice. My personal position on it is that I never quite enjoy writing Rust as much as other language; I never seem to get into that state of seamless flow I get into in other languages. But I also think it fills a niche that isn't particularly well-served by other languages. I think that Rust might not be the #1 tool for you or me, but I also think our industry as a whole needs it (or something similar) pretty desperately.
-
It might be that Rust is designed to solve different problems than the ones you need to solve? Though it's okay to just flat out dislike it as well. If you add up the billions of dollars of damage caused by good C programmers with decades of experience making mistakes that Rust would have disallowed - things like Heartbleed - you can see the motivation to use a language like Rust. Like it or not, year in and year old, a huge percentage of bugs and vulnerabilities in software are caused by memory safety issues. It's certainly not the best choice for every problem. I can't say I *love* writing Rust. But I think it's a good choice for you need C-like performance but can't afford vulnerabilities. Say you're implementing something like OpenSSL - I'd argue that if you're starting a project like that in 2022, you'd be crazy to pick C (or even C++) over Rust. I suppose you could pick something like MISRA C or even Ada as well, but I don't think either of those would be more joyful to write than Rust. For certain classes of problems, the only safe assumption you should make is that you do need training wheels because no matter how much experience you have or how good you are, eventually you will make a costly mistake. For most other applications, Rust probably isn't the best choice. My personal position on it is that I never quite enjoy writing Rust as much as other language; I never seem to get into that state of seamless flow I get into in other languages. But I also think it fills a niche that isn't particularly well-served by other languages. I think that Rust might not be the #1 tool for you or me, but I also think our industry as a whole needs it (or something similar) pretty desperately.
This all makes a lot of sense to me, and I think it even goes to explain what I don't like about it in the end - it's not "flowy" enough. I'd agree that our industry needs something like Rust, and maybe Rust is it. But I don't have to like it! :laugh:
To err is human. Fortune favors the monsters.
-
Yes, I've read that Linus hates C++. Which is why it won't happen, even though it makes sense. "Don't know what you're going to get" is mostly nonsense. Some naughty C++ code must have triggered him years ago.
Robust Services Core | Software Techniques for Lemmings | Articles
The fox knows many things, but the hedgehog knows one big thing.Greg Utas wrote:
Some naughty C++ code must have triggered him years ago
Nah, Linus is just a dick.
Software Zen:
delete this;
