|
3 layers deep is all you need to go.
1st layer: Entry points
2nd layer: logic
3rd layer: detailed code
|
|
|
|
|
Ah, yes. The DoEverythingAndThisAndAlsoThatAndOhThisOtherThingAsWell() function. 
|
|
|
|
|
Chris Losinger wrote: come on over to my world, where i'm looking at C functions that sometimes hit 1000 lines.
That's nothing! My record is over 1700 comment lines before the first line of code.
|
|
|
|
|
|
I am actually beginning to hate C++ as well, but the reasons you mention have nothing to do with the language. Try reading some typical Java "enterprise" code and finding a method that actually does something.
|
|
|
|
|
From your post, it would seem that you hate C++ because of how someone else writes Java "enterprise" code.
I suppose that makes some sense somewhere somehow to someone.
| Ravings en masse^ |
|---|
| "The difference between genius and stupidity is that genius has its limits." - Albert Einstein | | "If you are searching for perfection in others, then you seek disappointment. If you are seek perfection in yourself, then you will find failure." - Balboos HaGadol Mar 2010 |
|
|
|
|
|
And what this has todo with C++? It's more a programming style than a Need in C++ 
It does not solve my Problem, but it answers my question
modified 19-Jan-21 21:04pm.
|
|
|
|
|
Sorry for mentioning something simplistic, but I think immediately about call graphs …
When I used VC++ 4 or 5 I was extremely impressed by the call graph features … back then  ,
I found them very handy, of course … I missed them in early VC#, and they resurfaced
at some point in VC# editions (in last 10 years I think ...), BR
|
|
|
|
|
In my younger days, I was very eager about call graphs myself. Well, that was in the days when you could dive into, say, an OS and understand every detail of it.
Then I started working with communication protocols, and it was still in the days when we believed that the OSI model were for real, with its service APIs and layer borders serving as firewalls both upwards and downwards, and protocol definitions independent of service APIs. But most of all: You had no control over your peer. You had to program according to service and protocol definitions, inside the black box. We used those principles as design guidelines even for non-communication software.
When you are in a firewalling black box, you shall not be concerned with how that service you are invoking fulfills it task. You shall not know the context in which the user of your service offering operates. And you shall send and receive protocol elements without regard for how your peer does his processing. For the in-layer code, OO encapsulation hides so much that we think of object methods almost as as "primitives", not some deep well of calls into calls into calls.
If you are into the telecommunication way of programming, you essentially make all your implmentation designs (and realizations) as finite state machines. Then there is very little left for the call graphs to do. I am no longer programming communication protocols, but I've fallen into the habit of FSM style and layered design; now that you mention call graphs, to me that is a long lost memory. But I do not miss it.
|
|
|
|
|
Quote: If you are into the telecommunication way of programming, you essentially make all your implementation designs (and realizations) as finite state machines. Then there is very little left for the call graphs to do.
good point indeed … I used FSM's years ago in text analysis and this reminds me somewhat vaguely about cases of equivalency between FSM and grammars … a grammar looks like a tree so in such cases a call graph would apply … Perhaps it is difficult/impossible to generalize, BR
|
|
|
|
|
I'm in love with modern C++.
|
|
|
|
|
I still really need to catch up with all the improvements. I've been swimming in C#-land for so long. Haven't done anything serious in C++ in quite a few years beyond some C++/CLI here and there.
|
|
|
|
|
Really? In love?
Are you just saying that or do programming languages really do it for you?
|
|
|
|
|
Well, Java cannot do that for me. 
|
|
|
|
|
Munchies_Matt wrote: On and on and on you search, looking for code that actually DOES something!
Why do you think that C++ program should do something?
|
|
|
|
|
You are free to move to Managed C++ if you don't like this one. 
|
|
|
|
|
Now, I hope I'm not the only one here who experienced it as well... A couple of weeks ago my Win 10 taskbar buttons became completely unresponsive: nither "Start" nor the "Search" button showed anything on screen, but the change desktop worked fine and left click showed the alternative menu as always.
The trouble began after a day installing / uninstalling stuff. The only way to "fix" it, after following the usual suggested ones, was to restore a System Restore Point, effing off a couple of useful programs I installed and some executables I had in my Documents folder.
My Windows 10 does not have either Cortana nor Windows Store or apps support (it's the LTSB version) so it's not one of them. To rest more easily I installed ClassicShell, so that the Start menu manager is something different from the standard but I'm not entirely sure it will prevent the problem forever.
Well, that's it, just ranting and seeking if someone had the problem before and found what triggered it (I was deep in the middle of urgent vital work so I didn't have time to put on my spleunking gear and explore the system's innards).
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
|
|
|
|
|
Last week I had the unfortunate experience to find a problem (windows installer failing to find a resource(s) after allowing the April update. Like you, I had urgent work to do, so I figured I would just use System Restore to put it back to the way it was before the update and deal with it later.
Not sure if I made a mistake as I could have sworn the date for the restore point was the day of the update, but I started noticing right away that something was very wrong. After a couple of applications refused to start, I checked my Programs and Features and was shocked to find on a half-dozen programs listed there.  Weird that 90% of my programs (including VS6/2008/2010, SSMS) work just fine (well almost) without showing up in that list. Even the 3 versions of SQL Server are all working fine. (again almost...jobs using the agent have started failing, but I can work around that for now)
I've got a new box ready to take over if I keep finding stuff that doesn't work. Unfortunately the new box has a crappy sound card (no 5.1) so I'll either need to swap that out or just go get a new one...actually thinking about an external one this time.
But no, I've not had any issues with the task bar at all. 
"Go forth into the source" - Neal Morse
|
|
|
|
|
I hammered shut both Windows Update and Windows Defender (the latter really need to be hammered via registry or it will pop up again) so it's not an update... which on the LTSB are still rare.
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
|
|
|
|
|
I had something like that once, quite a while ago, after updating my graphics driver. But it didn't persist beyond a reboot.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
As an aside to @Codesavage message on google store TOS,
"ms buys github" ... what happens when ms decides:
- items on github violate their TOS, or
- usage of something from github could lead the user into violating ms TOS, or even
- if something on github promotes/advantages a competitors product over a ms product.
(allowing the last could be considered violation of responsibility to their shareholders)
sure, yes, ms have said they will play nice and are supposedly 'pro open, pro linux and others' - but that is nearly always only if:
it's running in a ms-windows [limited] [linux] environment (posb not native capable), or, even if it makes they underlying os irrelevant open source products become heavily reliant on underlying ms products / infrastructure - need ms proprietary dev tools and/or for release/deploy proprietary ms libs (IOW not ever native).
so, is ms really 'open source' friendly? or are they just leading developers into an 'open source' world that ms can one day easily shut down and charge money to turn it back on?
looking deeper into the way they are infiltrating open source it's nearly always to [deceptively] inject ms widgets and reliance on ms into everything that comes out.
heck, will even git itself remain accessible to open source git tools, or are they going to 'extend' proprietize that too?
"ms buys github" - I really hope people jump, but more likely laziness will prevail.
... and who knows what's next? buy CP and SO to 'even further support the community' [roll it into msdn and] monetize that too?
This internet thing is amazing! Letting people use it: worst idea ever!
|
|
|
|
|
As we did in the old times. Either we use Sourceforge or we host it ourselves.
Besides, Git as a tool sucks.
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
|
|
|
|
|
Lopatir wrote: are they just leading developers into an 'open source' world that ms can one day easily shut down and charge money to turn it back on?
As github has always been a closed environment (i.e., it was funded by VC who had all the power), that could have happened at any time. I don't really see Microsoft taking a dump in the sandbox at any time.
This space for rent
|
|
|
|
|
|
Marc Clifton wrote: Host your own!
The great thing about github is the code review features, and CI hooks (autotests etc). We are 30 devs in my company, for us it is a lot more than a just a repo.
... such stuff as dreams are made on
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
