What gets in the way of your Raspberry Pi project?

Python of course, I hate it! its childish.  you have to indent for levels you have to be kidding me, and NO semicolon. So where is the end of the statement. RUBBISH ! its almost as bad as BASIC! LOL  I just use C with eclipse!  You Cant do this in Python:

for(i, I < 100){ printf("i \n", %n);} return; 

phoenixcomm said:

You Cant do this in Python:

for(i, I < 100){ printf("i \n", %n);} return;

I'm pretty sure that you can make typo's in Python as well... 

I think you mean:

I had to embedd the code as a picture, the E14 insert code thing removed the files from the includes !

%i would work but I used d to avoid conusion with iterator i,

The u typing on the numeric constants is a bit of MISRA pedantry. I've got into the habit !

It's best to initialise the iterator in the for statement, rather than the declaration, so that some time later on, when the code gets modified no one gets caught out.

Pedantic style guides say that you should never use i or j or k as iterators but give them proper names like  line_number or self_destruct_counter.

I still use i

MK

michaelkellett said:

Pedantic style guides say that you should never use i or j or k as iterators but give them proper names like  line_number or self_destruct_counter.

I still use i

I do the same, if the iterators are just that, a counter to loop trough a number of items. When the loop has multiple levels, I may use outer and inner. When it has an obvious meaning, I may call it pageIndex, ..

But most of the times i, j, k 

michaelkellett said:

I had to embedd the code as a picture, the E14 insert code thing removed the files from the includes !

#include <stdint.h>
#include <stdio.h>

uint8_t i;

for (i=0u; i<100u; ++i)
{
    printf("%d \n", i);
}

It looks like it does when you first insert it, but it does it properly when you post it 

I think the difference is between Insert -> Code and Format -> Formats -> Inline -> Code

I'm sure it will be fixed in 'Birdhouse v2.0' 

No malice was ever intended. I'm interested in the whole concise vs verbose code thing and how it affects typos and code readability.

Also interested in perceptions of different programming languages. I tend to see them as different tools for different tasks.

More so amazed that I even spotted it given that I've mainly been reading up on VHDL/Verilog code recently and not on C. I think I've been rewiring my brain instead of the FPGA fabric...

I'm interested in your suggestion that a Python programe (large or small) could be proven correct. I am unaware that any proven correct Python compilers or interpreters exist (enlighten me if you know of any).

There have been claims of provably correct C compilers and other tools.

There is dispute that any useful programme can be proven correct.

some links:

https://stackoverflow.com/questions/4065001/are-there-any-provable-real-world-languages-scala

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5597727/

But you could spend the rest of your life on this - many will.

MK

No python compiler I know of is able to prove a program correct. What is needed is code that can be proven correct against specs and is maintainable if specs change or bugs appear. Python has a number of useful feature such as control of scope of variables, very strict type checking, inheritance to allow modules to be expanded from already proven modules, automatic memory allocation, automatic garbage collection etc... Proving code is correct is all about having a good spec and a test harness that can test against the spec - a larger job than writing the code in the first place. It can be done in any language. Some make it easier than others. I believe Ian Barron, the founder of Inmos claimed that the transputer could be proven to be correct but didn't ever do so. The only processor I know that was claimed to be mathematically proven correct was the viper ( years ago ) but that was found to have a bug. p.s. I think the first Verilog complier was a hacked C compiler. Hence the similarities with C, although system verilog and Python have a number of similar design concepts.

We haven't even got off first base with Python, a "proven correct Python compiler" isn't one which proves (or not)  other programmes to be correct. The issue is that (as far as I know) there is no Python compiler available which can be proved to correctly compile Python code.

Testing code does not prove it to be correct. Testing may give you some (non quantifiable) confidence about the number of errors in your code.

I'm not an expert on this but I would suggest that since high rel applications use only a subset of Ada (the whole thing being too risky) it is very unlikely to be even possible to write provably correct code in Python.

This doesn't mean that Python can't be useful - almost no code at all (possibly none) is provably correct.

Features like garbage collection make it very difficult to prove a compiler correct.

Relevant discussion on Stackoverflow:

https://stackoverflow.com/questions/69673807/python-coding-standard-for-safety-critical-applications

MK

I think Ada (the programming language) was developed such that the compiler caught everything, but I have never tried it to see how close they came.

I think Ada (the programming language) was developed such that the compiler caught everything, but I have never tried it to see how close they came.

It's impossible. Also impossible to assess how close you are from an unachievable goal.