|
I have come across some right howlers in this code base.
Anyway, C++, of all the languages I have used, from ADA, to Prolog, through VB and Java, allows this kind of sillyness. So it is for that that I condemn it.
And personally I dont see that OO is a massive benefit over a procedural language except in specific instances. And in fact it is often worse. Particularly in control code, code that is not data centric, but process centric.
|
|
|
|
|
I beg to differ, there are always pieces of information in a process which are data-centric, and applying OO correctly will break up processes in subprocess objects that are easier to isolate, replicate, store, observe and even parallelize.
I do intermix procedural and OO because pure OOP more often than not introduces complexity trying to fit square pegs in round holes, but that's precisely why I like C++ and not the oter OOP languages: it's C, but with 100% OO support.
GCS d-- s-/++ a- C++++ U+++ P- L+@ E-- W++ N+ o+ K- w+++ O? M-- V? PS+ PE- Y+ PGP t+ 5? X R+++ tv-- b+(+++) DI+++ D++ G e++ h--- ++>+++ y+++* Weapons extension: ma- k++ F+2 X
|
|
|
|
|
den2k88 wrote: there are always pieces of information in a process which are data-centric
Often not with drivers. In fact many are pure process and have no data of their own. They have states, flags, plenty of that, but no data.
|
|
|
|
|
Munchies_Matt wrote: Often not with drivers.
More like never, I concur. Drivers, firmware and heavy computing procedures get no benefit and mostly troubles at all from an internal OO implementation. As consumers of OOP (object as parameters), or external implementation (i.e. the driver is encapsulated in an object) they might get some benefits though.
GCS d-- s-/++ a- C++++ U+++ P- L+@ E-- W++ N+ o+ K- w+++ O? M-- V? PS+ PE- Y+ PGP t+ 5? X R+++ tv-- b+(+++) DI+++ D++ G e++ h--- ++>+++ y+++* Weapons extension: ma- k++ F+2 X
|
|
|
|
|
Probably why MSFT dont authorise C++ in the kernel, it isnt worth it.
OF course when I see classes with get/set functions I chuckle. Just what is the point of pretending it is OO? 
|
|
|
|
|
If you had ever had the experience of building a large software project using non-OO code, you would sing a different tune.
|
|
|
|
|
You can build a large project from procedural code just as well, it all depends on the architecture you design. Look at the WIndows kernel. All built in C (with a bit of assembler in the HAL)
|
|
|
|
|
And have you, personally, built a million-line code-base in purely procedural code? In C, perhaps? If you have done so, and have done the same using an object-oriented language, then you have standing to dismiss object oriented programming as no better than procedural.
Merely asserting that it is possible is not a very strong claim. Pointing to 30-year-old code like the Windows Kernel, that was developed before the broad availability of OO languages is meaningless.
|
|
|
|
|
|
While I agree with your statement, std::map implements this very [] operator overload as described. Actually that comment is a nice addition. Would I implement something like that though? probably not.
"the debugger doesn't tell me anything because this code compiles just fine" - random QA comment
"Facebook is where you tell lies to your friends. Twitter is where you tell the truth to strangers." - chriselst
"I don't drink any more... then again, I don't drink any less." - Mike Mullikins uncle
|
|
|
|
|
Quite, this is in the map object. Obscene is isnt it, that such an abortion as this is used so widely.
|
|
|
|
|
You are pretty close to the perfect answer....
C++ is as sensible or as stupid and daft as you want it to be.
Unfortunately not helped by the C++ ISO bods adding more and more different ways to make it more complicated without adding very much to real C++ users. The idea of less is more is lost on them.
Could do with minimum C++ with a lot of the bad and new stuff removed.
As for the Template library - great functionality but let down by the ludicrous syntax etc.
|
|
|
|
|
Yes, I still keep the AT&T C++ book on my desk. A slim volume describing a nice language which was an (IMO) elegant OO extension to 'C'.
Now you can look at valid C++ syntax that looks like a cat walked across the keyboard. They seem to be trying to put every feature of every other language into C++ syntax and comprehensibility be damned.
|
|
|
|
|
Hey! You can do stupid things in any language, if you try hard.
Back in my FORTRAN days, I discovered you could write self modifying code with judicious misuse of COMMON ... you can't do it now, but back then? Handy... :cringe:
Sent from my Amstrad PC 1640
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
OriginalGriff wrote: if you try hard.
And the lazies use VB.
|
|
|
|
|
I remember "PEEK(address)" and "POKE address, value" as dearly loved functions, but that was in the DOS days. If my memory is correct, it never made it into Visual Basic, though. (I guess it would be more or less meaningless, too, considering the memory management of Windows.)
|
|
|
|
|
OriginalGriff wrote: FORTRAN
OriginalGriff wrote: self modifying code
OriginalGriff wrote: misuse of COMMON
What. The. ![Elephant | [mastadon]](https://codeproject.global.ssl.fastly.net/script/Forums/Images/elephant.gif) .
Software Zen: delete this;
|
|
|
|
|
COMMON allowed you to share variables between different blocks of code: much like a C++ global variable being accessible from different classes.
But ... it wasn't type checked, so you could declare a float and then use COMMON to import it as a 7 dimensional array of BYTE values if you wanted. Because the array bounds checking worked on the data as declared in the COMMON statement with no actual reference back to the original variable, you could happily use it to access any location in your memory space. And because there was no physical separation between code and data segments (flat memory model in those days) your code was not in a "read only partition" as code is now. Provided you understood machine code you could revise your program while it was running ... 
Sent from my Amstrad PC 1640
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Interesting times
We can’t stop here, this is bat country - Hunter S Thompson RIP
|
|
|
|
|
We committed some serious atrocities in the name of efficiency back then. 
Sent from my Amstrad PC 1640
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Not as much for speed - more to overcome space limitations. I am too young to have any PDP-8 experience, but an older colleauge told that every single one of the 4094 memory locations - 12 bit data values, 12 bit addresses - was precious. If you had to use a constand, you would always look through the entire binary program to see if the value was already used somewhere, usually as an instruction code, rather that wasting a whole 12 bit word on a duplicate...
Yet, with 4096 twelve-bit words, they managed to make a complete newspaper typesetting system (they had hand coded a segment swapping system to read code from a 5 Mbyte (later: 10 Mbyte) disk. Only once did a customer demand all the options, and they couldn't fit all the resident parts into RAM. So they sold that newspaper a 2-CPU solution: When a function in CPU 1 needed to call some code that couldn't fit it, it raised an interrupt signal to CPU 2, the interrupt ID identifying the required function, and went to sleep while CPU 2 was working. Return from the function, in CPU 2, generated an interrupt back to CPU 1, interrupt ID being the result status, so that CPU 1 would wake up and continue its work...
I heard about this around 1980, at a time when the PDP-8s had been replaced by PDP-11s serveral years ago, so I guess the story dates to around 1970 ("Computer Techniques A/S", later renamed "Comtec AS", was founded in 1967).
|
|
|
|
|
Oh, I'm more than old enough to know how it worked. I'm just appalled at the notion of doing it in FORTRAN. I've written my own self-modifying code a couple of times, although not in the last 25 years or so. I can see where .NET could let you write self-generating code, but that's a kettle of different piscus, as it were.
Software Zen: delete this;
|
|
|
|
|
I'm just glad I never had the opportunity to do it in COBOL
Sent from my Amstrad PC 1640
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
I have two claims to fame in my career: I've never written an ECO(*) (Engineering Change Order), and I've never learned COBOL.
(*) A document used within my company that describes how to build a product. It's incredibly complicated, and for software it's completely over-engineered.
Software Zen: delete this;
|
|
|
|
|
Most oldtimers know the Ken Thompson Turing Award lecture "Reflections on trusting trust" (and the youngers really should be introduced to it!) For a few years after this paper was published (1984), various writers expanded on the idea.
One of the articles in my basement archive discusses how this compiler trojan could be implemented in the backend, code generating part, of a compiler suite such as gcc. It would then apply to all the compilers using that backend - Cobol is mentioned as an example, even though the malicious code never existed in Cobol form. This article discusses how the trojan could avoid detection, e.g. by propagating only to executables of a certain minimum size so that the size of the trojan code would be a tiny little fraction. It also discusses how a parse tree could be recognized as a compiler being compiled, and how the trojan could added at the parse tree level before code generation, so that it would spread even to the early (language dependent) compiler stages, not just the back end.
This would not exactly be self modifying code, but it illustrates that even Cobol certainly isn't safe for malware.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
