|
Isn't that specifically forbidden in the Geneva Convention?
- I would love to change the world, but they won’t give me the source code.
|
|
|
|
|
Article 4 Sub-paragraph 6 Section 2 Rule 1, Or there abouts! 
Along with Antimatter and Dark Matter they've discovered the existence of Doesn't Matter which appears to have no effect on the universe whatsoever!
Rich Tennant 5th Wave
|
|
|
|
|
Regulars may remember my previous rant about log4cxx - I am still confident that it is a c@rap tool, and should be avoided due to Apache's poor maintenance.
Whatsoever, in the mean time I found out that it is in fact a purely VS-Compiler related problem.
Many many tutorials on teh internetz have suggested to avoid the compiler errors by moving certain macros out of certain classes - Which would've needed a lot of time which I currently don't have at my hands. After a looong google search, I came to a page[^] which suggested to only modify the defines in log4cxx.h a bit, and oh wonder - If you compile it with the "pointer and list definition macros for all other cases", it builds with no error. Weird, and I don't get what I did, so if anyone could care to explain I'd be happy to listen carefully.
Still, I don't accuse Microsoft for causing the error, from my POV it is Apache who did a poor job maintainig their source.
#if defined(_MSC_VER) && _MSC_VER < 1700 && !defined(LOG4CXX_STATIC) && defined(LOG4CXX)
#define LOG4CXX_PTR_DEF(T) \
template class LOG4CXX_EXPORT log4cxx::helpers::ObjectPtrT<T>; \
typedef log4cxx::helpers::ObjectPtrT<T> T##Ptr
#define LOG4CXX_LIST_DEF(N, T) \
template class LOG4CXX_EXPORT std::allocator<T>; \
template class LOG4CXX_EXPORT std::vector<T>; \
typedef std::vector<T> N
#elif defined(_MSC_VER) && _MSC_VER < 1700 && !defined(LOG4CXX_STATIC)
#define LOG4CXX_PTR_DEF(T) \
extern template class LOG4CXX_EXPORT log4cxx::helpers::ObjectPtrT<T>; \
typedef log4cxx::helpers::ObjectPtrT<T> T##Ptr
#define LOG4CXX_LIST_DEF(N, T) \
extern template class LOG4CXX_EXPORT std::allocator<T>; \
extern template class LOG4CXX_EXPORT std::vector<T>; \
typedef std::vector<T> N
#else
#define LOG4CXX_PTR_DEF(T) typedef log4cxx::helpers::ObjectPtrT<T> T##Ptr
#define LOG4CXX_LIST_DEF(N, T) typedef std::vector<T> N
#endif
I will never again mention that Dalek Dave was the poster of the One Millionth Lounge Post, nor that it was complete drivel.
|
|
|
|
|
This is the sort of thing that made my move to C#.
(Well, that and the abysmal support for C++ in the early .NET days)
cheers
Chris Maunder
|
|
|
|
|
C# is a great language as long as you stick with .Net.
As soon as it comes to platform independency I prefer Qt-QML-C++, a combination which is hard to beat, on the front end at least, and they're getting better on DB & Web Service Support.
And it's that sort of thing which keeps me busy on weekends, sadly. You gotta wait for the ::Tasks article and the GUI mockup for another few weeks.
I will never again mention that Dalek Dave was the poster of the One Millionth Lounge Post, nor that it was complete drivel.
|
|
|
|
|
I briefly touched the Qt-QML-C++ platform for about 2 days to crank out a proof of concept app on a BB10 phone (don't ask). The one thing I found most disconcerting was the mix between JS code in the QML event handlers and C++ code in methods they called; and none of the getting started docs I found from RIM had anything to say about how the functionality should be split between the two.
Assuming that something comes out of the PoC I wrote and that I'm not able to avoid having to write it, how should I be partitioning code between the two locations?
Did you ever see history portrayed as an old man with a wise brow and pulseless heart, waging all things in the balance of reason?
Is not rather the genius of history like an eternal, imploring maiden, full of fire, with a burning heart and flaming soul, humanly warm and humanly beautiful?
--Zachris Topelius
Training a telescope on one’s own belly button will only reveal lint. You like that? You go right on staring at it. I prefer looking at galaxies.
-- Sarah Hoyt
|
|
|
|
|
I'm just getting started on Qt/QML, but from what I've understood you should separate the stuff as you would in a "normal" ASP.Net page (not exactly, but the closest I example I can think of): Everything GUI-fancyness and input validation goes into QML, and the business logic belongs into the C++ part.
Note: I don't know the RIM docs, but I guess for the basic stuff you are probably better off with the Qt-Documentation from Digia [^] and the official Forums[^]. Especially the forum is recommended, they got some really bright minds out there.
I will never again mention that Dalek Dave was the poster of the One Millionth Lounge Post, nor that it was complete drivel.
|
|
|
|
|
 Well, I guess it would have to do with what the compiler supports. I don't do native dev though, so don't quote me on this.
What do you get when you cross a joke with a rhetorical question?
|
|
|
|
|
From what I found out so far it has something to do with the restrictions on template declaration & definition which has changed since VS 2008.
I will never again mention that Dalek Dave was the poster of the One Millionth Lounge Post, nor that it was complete drivel.
|
|
|
|
|
Ah. I see.
What do you get when you cross a joke with a rhetorical question?
|
|
|
|
|
Marco Bertschi wrote: Apache who did a poor job maintainig their source
I'm sure Apache will be delighted if you step up as the next maintainer of Log4cxx.
Not every Apache project gets much in the way of funding, and I think it's unreasonable to expect the original author to maintain it for free forever ... let's just be grateful that he contributed it in the first place ...
Personally I would give Boost.Log[^] a try - it's still rather new and shiny, and AFAIK Andrey is still maintaining the library ...
|
|
|
|
|
Espen Harlinn wrote: I'm sure Apache will be delighted if you step up as the next maintainer of Log4cxx.
I'm not even sure if it should be maintained - There is Boost.Log, which we don't use because we'd have to test it first for it's reliability [Medical Equipment, Regulatory Guidelines are strong] but also log4cpp and log4cplus, all of the log4X libraries being ports of the original log4j library.
I think Apache should either maintain it, or have an official statement saying that log4cxx was last adjusted to build with VC 6, and not compatible with VS 2008, 2010 and newer.
I will never again mention that Dalek Dave was the poster of the One Millionth Lounge Post, nor that it was complete drivel.
|
|
|
|
|
Marco Bertschi wrote:
I'm not even sure if it should be maintained
Cannot say I disagree.
Marco Bertschi wrote: There is Boost.Log
We're currently using Boost.Log - and it does the job.
But if you're looking for something that's both actively maintained, and used to develop software with rather stringent requirements, you could have a look at ACE[^]. The who is using ACE [^] page should give you an indication about the requirements that ACE meets.
|
|
|
|
|
Espen Harlinn wrote: But if you're looking for something that's both actively maintained, and used to develop software with rather stringent requirements, you could have a look at ACE[^]. The who is using ACE [^] page should give you an indication about the requirements that ACE meets.
It's not a problem with Boost itself - Even for ACE, we'd need to do an Assessment to prove that it works the way we expect it to work. That's a company guideline to prevent lawsuits (just in case), and because we (the project) currently don't have the time or the budget to do the assessment we decided to stick with log4cxx, which is a rather nasty piece of software but was already assessed by other projects.
I will never again mention that Dalek Dave was the poster of the One Millionth Lounge Post, nor that it was complete drivel.
|
|
|
|
|
At the top of the page, above the advert next to Bob.
It now says 10.47 million instead...  
|
|
|
|
|
I actually debated long and hard with myself about that and the exact number seemed...overly precise given its size.
I guess, though, when has "overly precise" ever been a bad thing in computer science?
OK, I'll add it back.
cheers
Chris Maunder
|
|
|
|
|
Pedant Alert...
How can you be overly precise.
One is either precise or imprecise.
There is no try.
---------------------------------
Obscurum per obscurius.
Ad astra per alas porci.
Quidquid latine dictum sit, altum videtur .
|
|
|
|
|
Actually there are degrees of precision. Sure, in accounting 2 decimal places is enough, but I would wager others would consider that insufficient.
cheers
Chris Maunder
|
|
|
|
|
In physics an order of magnitude is close enough, alas that is not the case in accountancy.
Penny perfect or it is wrong!
If a balance sheet doesn't balance then there is a late night pending 
Been there far too often.
---------------------------------
Obscurum per obscurius.
Ad astra per alas porci.
Quidquid latine dictum sit, altum videtur .
|
|
|
|
|
I am taking an accounting class and had that happen. I quickly found an off-by-one error I had made (893.83 instead of 893.84). 
Accounting is interesting, although I don't know why Ivy Tech requires it for a computer programming degree. (They won't this fall, though, as the entire info tech department is being put into their own building and a complete restructuring of the classes needed is being finalized, and they removed Accounting from the requirements (it still will be available as an elective, though)).
What do you get when you cross a joke with a rhetorical question?
|
|
|
|
|
Brisingr Aerowing wrote: I am taking an accounting class
Sorry for your luck. 
The report of my death was an exaggeration - Mark Twain
Simply Elegant Designs JimmyRopes Designs
I'm on-line therefore I am.
JimmyRopes
|
|
|
|
|
Tell That To Turbo Tax, One Year They Said I HAd An Extra Dollar unaccounted For because They Wouldn't AlLow Me To Input Pennies.
You'll never get very far if all you do is follow instructions.
|
|
|
|
|
No, Accuracy Is Boolean, And If Inaccurate, Then Precision Comes Into Play.
You'll never get very far if all you do is follow instructions.
|
|
|
|
|
...and if you are wrong, you are only off by a bit.
Never moon a werewolf.
- Harvey
|
|
|
|
|
Yes, but based on whether your bits are arranged big-endian or little-endian that bit could be very significant!
- I would love to change the world, but they won’t give me the source code.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
