276

Re: Library for creation of graphic user interfaces

Hello, night beast, you wrote: NB> already read and did not see anything about that that QObject should delete itself. Still attempts will be? It is clear you as  the child cannot behave as the adult and to recognize errors. NB> I repeat a question. No, at first you my question answer or recognize that you were mistaken that for lifetime QObject its counter of links answers: Show was specific in what file on what line there is a counter of links QObject which the Author controls its lifetime http://rsdn.org/forum/cpp.applied/6905632.1: MTD Date: 16.09 09:43

277

Re: Library for creation of graphic user interfaces

Hello, so5team, you wrote: MTD>> the Proof will be, what several times (and even orders) there is more than time and efforts or you ? S> https://github.com/valyala/fasthttp - 972 , 27 , the first  - on October, 19th 2015. S> restinio for today - 421 , 3 , the first  - on April, 5th 2017. Also what it should show, besides, that one project began on October, 19th 2015, and another on April, 5th 2017? S> But you talk with  Truth? Well it is fine.

278

Re: Library for creation of graphic user interfaces

Hello, MTD, you wrote: NB>> I repeat a question. MTD> is not present, at first you my question answer or recognize that you were mistaken that for lifetime QObject its counter of links answers: that is acknowledgement  about lies too will not be? Apprx. clear.

279

Re: Library for creation of graphic user interfaces

Hello, night beast, you wrote: NB> that is acknowledgement  about lies too will not be? It will be finite, after my question accurately and without eludings you will answer or you recognize a wrongfulness.

280

Re: Library for creation of graphic user interfaces

Hello, MTD, you wrote: MTD>>> the Proof will be, what several times (and even orders) there is more than time and efforts or you ? S>> https://github.com/valyala/fasthttp - 972 , 27 , the first  - on October, 19th 2015. S>> restinio for today - 421 , 3 , the first  - on April, 5th 2017. MTD> Also what it should show, besides, that one project began on October, 19th 2015, and another on April, 5th 2017? Probably, absence of brains. Or abilities them to use. For even date of the first  as it is not difficult to guess (in the presence of brains, it is finite), is not so mandatory coincides with a project date started.

281

Re: Library for creation of graphic user interfaces

Hello, MTD, you wrote: NB>> that is acknowledgement  about lies too will not be? MTD> it will be finite, after my question accurately and without eludings you will answer or you recognize a wrongfulness. Already answered. Accurately and without eludings. I can repeat://these objects are all used to indicate that a QObject was deleted//plus QPointer, which keeps a separate list QAtomicPointer<QtSharedPointer::ExternalRefCountData> sharedRefcount; the counter of links in QObject. It is not pleasant/not it agree - your difficulties.

282

Re: Library for creation of graphic user interfaces

Hello, so5team, you wrote: S> it is visible, absence of brains. Or abilities them to use. Evgenie, did not expect that at you so , look a chair do not burn

283

Re: Library for creation of graphic user interfaces

Hello, night beast, you wrote: NB> already answered. Accurately and without eludings. NB> I can repeat: NB> NB>//these objects are all used to indicate that a QObject was deleted NB>//plus QPointer, which keeps a separate list NB> QAtomicPointer<QtSharedPointer::ExternalRefCountData> sharedRefcount; NB> the counter of links in QObject. It not the counter of links, it simply the flag is live or dead object, I do not know why it so named, well and lifetime it does not control, generally.

284

Re: Library for creation of graphic user interfaces

Hello, MTD, you wrote: NB>> the counter of links in QObject. MTD> It not the counter of links, it simply the flag is live or dead object, I do not know why it so named, well and lifetime it does not control, generally. The second line of a comment did not master? And named so because earlier except  it also  used. I repeat. It agree or not - your problems. Now would like to hear citations on announcement about lies.

285

Re: Library for creation of graphic user interfaces

Hello, night beast, you wrote: NB> I repeat. It agree or not - your problems. Take your time meanwhile you showed a flag, I and itself showed it to you right at the beginning, you yet did not check that here so all is simple. Now show where the counter of links in QObject which controls its lifetime?

286

Re: Library for creation of graphic user interfaces

Hello, MTD, you wrote: NB>> I repeat. It agree or not - your problems. Citations will not be? Understood. I think, on it and we finish.

287

Re: Library for creation of graphic user interfaces

Hello, night beast, you wrote: NB> I think, on it and we finish. Again could not answer and merged, as it is predicted.

288

Re: Library for creation of graphic user interfaces

Hello, MTD, you wrote: NB>> I think, on it and we finish. MTD> again could not answer and merged, as it is predicted. Yes. To teach you to use  for flags above my forces so in the tenth time to show how this field it is used perhaps I will not be. You consider that a flag - on health. So on $$$ a lot of time spent.

289

Re: Library for creation of graphic user interfaces

Hello, night beast, you wrote: NB> so in the tenth time to show how this field it is used perhaps I will not be. So you also did not show anything. NB> so on $$$ a lot of time spent. Bla-bla-bla. What for began? Again to merge?

290

Re: Library for creation of graphic user interfaces

Hello, MTD, you wrote: NB>> so in the tenth time to show how this field it is used perhaps I will not be. MTD> so you also did not show anything. NB>> so on $$$ a lot of time spent. MTD> Bla-bla-bla. What for began? Again to merge? Misters, fasten with ! On Ligovke Father Frost beat! https://youtu.be/YQT-jJGpf8c? t=262 the World, friendship, a chewing gum.

291

Re: Library for creation of graphic user interfaces

Hello, alex_public, you wrote: _> Also what, slowly works? I did not try that on such volumes, but the code obvious (two lines - the iterator from filesystem and check on a mask from regexp) and on idea there is no place to brake essentially. Slowly. WinAPI to 1000 times it appeared by search in a mask faster. For does not sort out all successively. But I do not criticize filesystem, here as with sockets - if high-speed performance we do separately under each platform is necessary. If is not present - that boost asio or analogs. Personally from filesystem the class path suffices me. Here it really convenient.

292

Re: Library for creation of graphic user interfaces

Hello, SaZ, you wrote: _>> Also what, slowly works? I did not try that on such volumes, but the code obvious (two lines - the iterator from filesystem and check on a mask from regexp) and on idea there is no place to brake essentially. SaZ> slowly. WinAPI to 1000 times it appeared by search in a mask faster. For does not sort out all successively. But I do not criticize filesystem, here as with sockets - if high-speed performance we do separately under each platform is necessary. If is not present - that boost asio or analogs. Personally from filesystem the class path suffices me. Here it really convenient. Launched now for the sake of interest the above described code on Boost (3 lines) for search *.h files in a folder of source codes Qt (207937 files and folders). Also received result in 42790 files for 2065 milliseconds. Standard  the Explorer searches noticeably there was no high-speed performance of a variant from Boost longer (truth it thus also shows the file list) so I do not know for what such tasks to you. Well and by the way, in WinAPI there is no direct analog of the given code since there it is necessary to cause hands function of search for each subdirectory (FindFirstFile is not able to work recursively). So very much I doubt that it could be "in 1000 times faster" - similar on someone's imaginations...)

293

Re: Library for creation of graphic user interfaces

Hello, alex_public, you wrote: _> Hello, SaZ, you wrote: _>>> Also what, slowly works? I did not try that on such volumes, but the code obvious (two lines - the iterator from filesystem and check on a mask from regexp) and on idea there is no place to brake essentially. SaZ>> slowly. WinAPI to 1000 times it appeared by search in a mask faster. For does not sort out all successively. But I do not criticize filesystem, here as with sockets - if high-speed performance we do separately under each platform is necessary. If is not present - that boost asio or analogs. Personally from filesystem the class path suffices me. Here it really convenient. _> launched now for the sake of interest the above described code on Boost (3 lines) for search *.h files in a folder of source codes Qt (207937 files and folders). Also received result in 42790 files for 2065 milliseconds. Standard  the Explorer searches noticeably there was no high-speed performance of a variant from Boost longer (truth it thus also shows the file list) so I do not know for what such tasks to you. Well and by the way, in WinAPI there is no direct analog of the given code since there it is necessary to cause hands function of search for each subdirectory (FindFirstFile is not able to work recursively). So very much I doubt that it could be "in 1000 times faster" - similar on someone's imaginations...) Probably something exchanged. I tried std:: experimental:: filesystem from any msvs. For a long time.

294

Re: Library for creation of graphic user interfaces

Hello, MTD, you wrote: MTD> Hello, ollv, you wrote: O>> it would be desirable to note that bicycles like  it is fast  as superfluous. MTD> since 2008 as I started to write on Qt it I hear, still I about With ++ heard the same, only earlier. All are died off by them, all die off, here that's all. So)) presence we tell so in  it does not cancel unproductive thought. Whether Mahlo who that speaks. Here what for them ? And you know, what now by the hands  is implemented - in a half-kick?  decltype and ., facilitate life repeatedly. But Error that you orient on , it is necessary to look more deeply - MFC, Cbuilder and other other other, it practically dies off with so we tell popular means  . The sense in that the world on a place is not necessary, namely the world With ++ on a place is not necessary, and that allowed to do preprocessor  with  heaps , will be implementable in pluses (however already, personally I now on the project quite successfully  , Russian Orthodox Church and other... The abstract serialization of that  in any backend substitute JSON XML/BIN all joins in   a type) O>> , O>> personally , me - long selected a word - it is opposite MTD> Clearly, judgement of fanatics especially valuably, thanks. And here fanaticism, exceptional practice look - I here faced that the recursion of a template is not deeper 500, and in  it not to repair -  .) )  I simply do not love,  it is a lot of  hemorrhoids, however I do not eliminate that is simple resources on creation normal from ideologically \filosovskoj point of view  does not suffice, and  to use bicycles  . Though  it is more  at me with objective a C and JAVA. But in any way with pluses. The blessing in everyones  (, ,  std) to it  to a course is not present. And   causes in me much more pleasant esthetic sensations, than selectors.

295

Re: Library for creation of graphic user interfaces

I look at your posts and I see that it is all  from area "if ifs and ans were pots and pans, yes mushrooms" would grow in a mouth. It as though theoretically type yes - on pluses and a reflection, both , and orm, etc., and here practically everyone faced it solves - "yes well it !#" and a beret C#/Java and rejoices lives. Because, that is - reads off scale on inconvenience and-or complexity in usage. When it is impossible  from pluses - that do  - even it in one thousand times more conveniently, than various modern variants  reflections to pluses. Qt very elegantly solves a problem. That there moc works - I do not see problems, there it is not necessary to climb. And to worry because of it is all the same what to worry that except the compiler there the preprocessor and  works still. Qt in the essence generally very elegant library. As a matter of fact an example as it is necessary to approach to development of libraries. As contrast it is possible to take a vector of development of modern with ++ - explicitly contracted not there: instead of  uniform architecture started to mold there all in style boost - "time works, and so descends, and  that it all ill-matched,  and without uniform architecture/purpose". And that just about there will be a reflection (and units) - so I will already not remind at all how many years it "here here" lasts - and except  the result any is not present. And it is far not the fact that in a normal type for practical usage it is in time appears at our life.

296

Re: Library for creation of graphic user interfaces

Hello, push, you wrote: P> I Look at your posts and I see that it is all  from area "if ifs and ans were pots and pans, yes mushrooms" would grow in a mouth. It as though theoretically type yes - on pluses and a reflection, both , and orm, etc., and here practically everyone faced it solves - "yes well it !#" and a beret C#/Java and rejoices lives. Because, that is - reads off scale on inconvenience and-or complexity in usage. It is not necessary to project the (and others who too did not master language) sensation on all around.) )) P> When it is impossible  from pluses - that do  - even it in one thousand times more conveniently, than various modern variants  reflections to pluses. Reflections in the modern C ++ are not present basically (and in my opinion it is a principal lack of language which should be in a priority at Committee) so any crutches replacing this functionality, a priori are variations on a subject . Simply it can be implemented both exterior tools, and means of the language (metaprogramming possibilities in the modern C ++ quite allow it). However anyway this all crutches which will be thrown out on a dustbin after acceptance normal static  (on the sample of language D) in the language standard (strongly I hope that already in a C ++ 20 we see it). P> Qt very elegantly solves a problem. That there moc works - I do not see problems, there it is not necessary to climb. And to worry because of it is all the same what to worry that except the compiler there the preprocessor and  works still. Qt anything there elegantly does not solve, at least because from it the solution of a problem qualitative  GUI for which reflections any there (and accordingly moc'), etc. are not required first of all wait. And those areas in which the reflection is useful (serialization type, ORM, etc.) Have in the C world ++ set of the implementations, on a head above a variant from Qt. P> Qt in the essence generally very elegant library. As a matter of fact an example as it is necessary to approach to development of libraries. As contrast it is possible to take a vector of development of modern with ++ - explicitly contracted not there: instead of  uniform architecture started to mold there all in style boost - "time works, and so descends, and  that it all ill-matched,  and without uniform architecture/purpose". It just Qt lagged behind development of the modern C ++, remained somewhere in 90-s'. Though gradually they try to move ahead more close to modern lines in programming, but very slowly... P> And that just about there will be a reflection (and units) - so I will already not remind at all how many years it "here here" lasts - and except  the result any is not present. And it is far not the fact that in a normal type for practical usage it is in time appears at our life. It not so. The C ++ 11 was a major landmark in the C world ++ not only the innovations in language, but also that strict enough rules of its further development - new versions with quite unambiguous mechanisms of arguing of innovations, etc. Yes has been entered each 3 years, in a C ++ 17 the important had not time to accept (to agree more precisely between all sides) a lot of. But the main thing that quite operating implementations of all these necessary things have been already presented to Committee so now at arguing of upcoming version of the standard just with them and begin. And any more from acquaintance, and are engaged in finishing to a state satisfying all sides. Therefore I on 100 % am assured that in the following standard will be , coroutines and ranges (as STL extension on basis ). And with a high probability units, , transactions, operation with a network (the component boost.asio in standard library).