51

Re: Bucket

Hello, eskimo82, you wrote: E> Kernel Linux is written on a C and it on some orders is more stable than kernel Win. About what it speaks? Emmas, on some orders is more stable?) and it is possible the link to acknowledgement of given "fact"?)

52

Re: Bucket

Hello, eskimo82, you wrote: E> Kernel Linux is written on a C and it on some orders is more stable than kernel Win. About what it speaks? I suppose, here everything, as always. A kernel of Linux of the last versions (what there number, 4?) on some orders kernels of Chikagi are more stable.

53

Re: Bucket

Hello, CreatorCray, you wrote: CC> Well here that came to the first mind that enrages me in POSIX: open (2) CC> CC> int open (const char *, int...) CC> CC> Here for varargs in it API it is necessary to tear off hands. CC> Therefrom: POSIX permissions - terrible abomination. I had in I see a positive example. Negative I and itself know as much as necessary. CC> you in course as are inside arranged "the virtual call"? Than here interrupt context is worse than the other? Expose correct  on CODE segments that it did not appear that them is necessary , the class data too should be in nonpaged. To With ++ it does not concern, it is specificity . I also meant it under gotchas. CC> And generally in fire wood the virtual functions still it is necessary to guess where . CC> all is normal there on orders easier. Static polymorphism suffices for eyes.  - yes all kernel as a matter of fact on dynamic polymorphism is constructed. Only its mechanism a little another, but it only an implementation detail - conceptually is the same virtual methods. And about "" - all depends on a specific goal. In certain cases very much even it is useful. CC> And than on yours RAII in callbacks differs from other places? The code in fire wood (and basically system) is written exception free. Exceptions are applied strictly on business and is normal is thrown and caught nothing system, type got into which storage are not present. CC> And any With ++ specificity in it is not present, all it strictly kernel-specificity. I also spoke about it. In About all code directly before eyes, in pluses much that happens "offstage", and in a kernel it is necessary to understand very well that happens, and it can have what consequences in different contexts (for the driver code can appear in a context of any process, high IRQL, in a situation when so not enough the free storage that driver parts appear preempted of , and attempts will select storage to come to an end  when iron behaves incorrectly, etc., etc.) And not to drop at all system - since unlike a user-space, to "fall" is never the decision. In With there is no latent magic. So it just  specificity With ++.

54

Re: Bucket

Hello, koandrew, you wrote: K> I had in I see a positive example. Negative I and itself know as much as necessary. Announcement was about POSIX is thought over is worse, as it it is possible to confirm with what that with specific positive examples I . It is necessary as a matter of fact all Zw/Nt/Rtl API to enumerate with comparing with POSIX counterparts. And  at me desire to scribble such opus magnum it is not watched. K> I also meant it under gotchas. So it not With ++ gotchas. On About all these knee-bends in the same way it is necessary to do. K> in About all code directly before eyes, in pluses much that happens "offstage" Here it  you seriously? The person who does not understand that happens in it With ++ kernel the code "under a cowl" it not With ++ drooped and  on a stick. K> and in a kernel it is necessary to understand very well that happens, and it can have what consequences in different contexts (for the driver code can appear in a context of any process, high IRQL, in a situation when so not enough the free storage that driver parts appear preempted of , and attempts will select storage to come to an end  when iron behaves incorrectly, etc., etc.), and not to drop at all system - since unlike a user-space, to "fall" is never the decision. It is all mandatory for  in any language.... <<RSDN@Home 1.0.0 alpha 5 rev. 0>>

55

Re: Bucket

Hello, CreatorCray, you wrote: CC> Announcement was about POSIX is thought over is worse, as it it is possible to confirm with what that with specific positive examples I . It is necessary as a matter of fact all Zw/Nt/Rtl API to enumerate with comparing with POSIX counterparts. And  at me desire to scribble such opus magnum it is not watched. Pogod-pogod. You stated that on *nix platforms traditionally With ++ is made as that on "from %% ". Here I would like to look at an example where it is made well. CC> so it not With ++ gotchas. On About all these knee-bends in the same way it is necessary to do. On About all code before eyes, also is not present any implicit  magic which can quit sideways under certain circumstances. CC> here it  you are serious? The person who does not understand that happens in it With ++ kernel the code "under a cowl" it not With ++ drooped and  on a stick. Nevertheless I in the career met such much. And even among them were, forgive My God, "leading" developers from offices with very loud names. CC> it is all mandatory for  in any language. Once again - my point that in With ++ there are the certain implicit actions generated by the compiler while in From them is not present.

56

Re: Bucket

Hello, eskimo82, you wrote: E> Kernel Linux is written on a C and it on some orders is more stable than kernel Win. About what it speaks? That you say lies Well or simply you do not know, about what you speak.

57

Re: Bucket

Hello, koandrew, you wrote: K> on *nix platforms traditionally With ++ it is made as that on "from %% ". K> Here I would like to look at an example where it is made well. For this purpose it will be necessary to you  in  on With ++ under  and under Windows. CC>> here it  you are serious? The person who does not understand that happens in it With ++ kernel the code "under a cowl" it not With ++ drooped and  on a stick. K> nevertheless I in the career met such much. And even among them were, forgive My God, "leading" developers from offices with very loud names. The C ++ alas, has high enough threshold of an input to write the qualitative code. And again  alas, allows to write everything to those who insufficiently qualified. As the scalpel - one it makes it brain surgery, another - bloody medley. Loudness of a name of office does not give any warranties, it as though is known for a long time. K> On About all code before eyes, also is not present any implicit  magic which can quit sideways under certain circumstances. K> Once again - my point that in With ++ there are the certain implicit actions generated by the compiler while in From them is not present. If to compare to the assembler that in With simple  the implicit actions generated by the compiler. Automatic coercion of types for example. Simply you perceive them as due and With ++  consider as magic.... <<RSDN@Home 1.0.0 alpha 5 rev. 0>>

58

Re: Bucket

Hello, koandrew, you wrote: E>> Kernel Linux is written on a C and it on some orders is more stable than kernel Win. About what it speaks? K> that you say lies Well or simply you do not know, about what you speak.  you nervous what that recently. On people you rush.... <<RSDN@Home 1.0.0 alpha 5 rev. 0>>

59

Re: Bucket

Hello, alex_public, you wrote: vsb>> Java fine language, on the order is more convenient Objective a C or Swift. _> Well if about Objective-C it is really possible to agree, comparing Java with Swift is simply ridiculous. Swift is the modern language with set of conveniences to the developer while Java wildly lagged behind in development and hardly already makes sometime up. To language it is necessary  and . At  is GC and any Eclipse 2000th years for the order , than Xcode which can compile one line minutes. Swift  waves, of course, but to program on it from it it becomes more convenient not.

60

Re: Bucket

Hello, CreatorCray, you wrote: CC> Chota you nervous what that recently. CC> on people you rush. It seems to you. It is necessary to drink less.

61

Re: Bucket

Hello, vsb, you wrote: _>> Well if about Objective-C it is really possible to agree, comparing Java with Swift is simply ridiculous. Swift is the modern language with set of conveniences to the developer while Java wildly lagged behind in development and hardly already makes sometime up. vsb> to language it is necessary  and . At  is GC and any Eclipse 2000th years for the order , than Xcode which can compile one line minutes. Swift  waves, of course, but to program on it from it it becomes more convenient not.  and who forces to program on Swift by means of XCode?))) you Can take the same AppCode from JetBrains or other popular IDE/editors.

62

Re: Bucket

Hello, AlexRK, you wrote: ARK> https://habrahabr.ru/post/345756/#comment_10590376 ARK> ARK> you sometime worked with ? Wrote under it applications/games? Saw it API or the code? All a decay. All of us die.

63

Re: Bucket

Hello, alex_public, you wrote: _> Hello, vsb, you wrote: vsb>> Java fine language, on the order is more convenient Objective a C or Swift. _> Swift is the modern language with set of conveniences to the developer while Java wildly lagged behind in development and hardly already makes sometime up. "Set" is more than three? Name 3 most important conveniences which are not present in Java. (About making up I for language will not catch you)

64

Re: Bucket

Hello, bzig, you wrote: vsb>>> Java fine language, on the order is more convenient Objective a C or Swift. _>> Swift is the modern language with set of conveniences to the developer while Java wildly lagged behind in development and hardly already makes sometime up. B> "set" is more than three? Name 3 most important conveniences which are not present in Java. (About making up I you for language will not catch) Hy, 3? Yes there I think it will be easy 123 to name if permanently to use it. I Swift do not use (only understood with it at the moment of an output for the sake of the general education and the understanding of that does not suffice in language used by me), but on storage I can  is explicit more 3-yoh points: - an automatic output of types (let and var) - support RAII ( and a construction defer) - an overload of operators - interpolation of lines - (extensions extensions for any types (including built in int, etc.) - Optional values of any types (analog nullable from C#) with special support of the organization of chains (as a matter of fact a monad maybe) and exception handlings (a construction try?) - the convenient built in containers (arrays, sets, dictionaries) with normal initialization ("var dict = [" k ": 3,14]"), iteration ("for (k, v) in dict") and cutoffs ("dict [3...]") - support of high-grade properties (with conveniently writeable operations get and set) for structures and classes - concrete parameters of functions - high-grade tuples, with automatic disclosure in variables and reset from functions - various types (including tuples) and expressions for comparing in switch - ranges (type "1. 5"), including infinite (type "3..." ) - A trifle, but convenient... - the majority of types (including containers, lines, listings, structures) are types-values that besides convenience to the programmer also increases high-speed performance of the code (escape the analysis simply is not necessary) and reduces storage consumption - possibility of operation with bare pointers and absence GC gives the chance a spelling system and  the code.

65

Re: Bucket

Hello, alex_public, you wrote: _> Hy, 3? Yes there I think it will be easy 123 to name if permanently to use it. I Swift do not use (only understood with it at the moment of an output for the sake of the general education and the understanding of that does not suffice in language used by me), but on storage I can  is explicit more 3-yoh points: _> - an automatic output of types (let and var) _>-... Under  such it is necessary - at me now the syntactic diabetes begins.

66

Re: Bucket

Divided your list into some parts. 0) already is _> - ranges (type "1. 5"), including infinite (type "3...") - a trifle, but convenient... _> - optional values of any types (analog nullable from C#) with special support of the organization of chains (as a matter of fact a monad maybe) and exception handlings (a construction try?) _> - the convenient built in containers (arrays, sets, dictionaries) with normal initialization ("var dict = ["k": 3,14]"), iteration ("for (k, v) in dict") and cutoffs ("dict [3...] ") 1) Formally is not present, but it is implemented in project Lombok which is connected as a plug-in  and  and allows to receive transparently this and (other functionality) in Java. _> - support of high-grade properties (with conveniently writeable operations get and set) for structures and classes 3) Is in the nearest plans _> - an automatic output of types (let and var) _> - the majority of types (including containers, lines, listings, structures) are types-values that besides convenience to the programmer also increases high-speed performance of the code (escape the analysis simply is not necessary) and reduces consumption of storage 3) Are not present, are not present in plans and it is a pity that are not present _> - high-grade tuples, with automatic disclosure in variables and reset from functions 4) Is implemented for a tick, is not compatible with i18n _> - interpolation of lines 5) Like are not present, on the other hand is not clear, on a horse-radish generally it is necessary _> - concrete parameters of functions _> - various types (including tuples) and expressions for comparing in switch _> - (extensions extensions for any types (including built in int, etc.) 6) I Hope, never will be, because it is the first step to  in  _> - an overload of operators 7) Rzhaka. To accuse managed language, that it not managed. What for one more With ++? _> - support RAII ( and a construction defer) _> - possibility of operation with bare pointers and absence GC gives the chance a spelling system and  the code. As a result, only 2) and 3) it is a pity that, no. Thus only on 3) there are no sounded plans, and 2) already saw.

67

Re: Bucket

Hello, bzig, you wrote: B> 0) Already is _>> - ranges (type "1. 5"), including infinite (type "3...") - a trifle, but convenient... _>> - optional values of any types (analog nullable from C#) with special support of the organization of chains (as a matter of fact a monad maybe) and exception handlings (a construction try?) _>> - the convenient built in containers (arrays, sets, dictionaries) with normal initialization ("var dict = [" k ": 3,14]"), iteration ("for (k, v) in dict") and cutoffs ("dict [3...]") About as it is amusing. Well time you such state, certainly can show examples of the similar code on Java? Human initialization, iteration and a cutoff on HashMap. Convenient (operation ".?") chains optional, analog "try?" And .... B> 1) Formally is not present, but it is implemented in project Lombok which is connected as a plug-in  and  and allows to receive transparently this and (other functionality) in Java. B> 2) is in the nearest plans B> 3) Is not present, is not present in plans and it is a pity that is not present B> 4) It Is implemented for a tick, it is not compatible with i18n B> 5) Like is not present, on the other hand it is not clear, on a horse-radish generally it is necessary Well i.e. as I and spoke, misses at present in Java. B> 6) I Hope, never will be, because it is the first step to  in  _>> - an overload of operators Generally that in any poor (besides a type there is nobody a plug-in) a variant try to fasten: http://amelentev.github.io/java-oo/. B> 7) Rzhaka. To accuse managed language, that it not managed. What for one more With ++? If the lack lies in the foundation architecture language it from it does not cease to be a lack. B> as a result, only 2) and 3) it is a pity that, no. Thus only on 3) there are no sounded plans, and 2) already saw. As a result of Java is become numb and backward  which managed to lag behind in the development even the old man of a C ++. And comparing with the modern languages (type Swift, Rust, etc.) generally is simply ridiculous. Thus in world JVM there are modern competitive decisions (Scala for example), but Java to them the relation is no.

68

Re: Bucket

69

Re: Bucket

_> Human initialization, iteration and a cutoff on HashMap. Stream and methods generating it. _> chains optional, Convenient (operation ".?"), analog "try?"  class Optional.of (...).map (...) .... B>> 1) Formally is not present, but it is implemented in project Lombok which is connected as a plug-in  and  and allows to receive transparently this and (other functionality) in Java. B>> 2) is in the nearest plans _> Well i.e. as I and spoke, misses at present in Java. You still spoke as will not be B>> 4) Is implemented for a tick, is not compatible with i18n B>> 5) Like are not present, on the other hand it is not clear, on a horse-radish generally it is necessary _> Well i.e. as I and spoke, misses at present in Java. Whether Mahlo what . That misses Java that and is good that does not pull in a mouth all successively, and seven times measure and once cut. _> if the lack lies in the foundation architecture language it from it does not cease to be a lack. Java became language number 1 just exactly because did not go on the way With ++ and all these advantages  feet fast methods for all tastes. Whom it sharply does not suffice, stamp back in With ++ and leave Java alone. B>> As a result, only 2) and 3) it is a pity that, no. Thus only on 3) there are no sounded plans, and 2) already saw. _> as a result of Java is become numb and backward  which managed to lag behind in the development even the old man of a C ++. Java became language number 1 just exactly because did not go on the way With ++ and all this programming in punctuation language. _> and comparing with the modern languages (type Swift, Rust, etc.) generally is simply ridiculous. Swift, all right, occupies the niche by means of Epla. And here Rust let at first something achieves. _> thus in world JVM there are modern competitive decisions (Scala for example), but Java to them the relation is no. That approximately shows that in the world of Java you are guided approximately in any way. I hope, in Swift, Raste and With ++ is more, differently all this discussion of sense had no.

70

Re: Bucket

Hello, bzig, you wrote: _>> Human initialization, iteration and a cutoff on HashMap. B> Stream and methods generating it. _>> chains optional, Convenient (operation ".?"), analog "try?" B>  class Optional.of (...).map (...).... Well so specific code samples will be?) At me small  were still in initial the message... B>>> 1) Formally is not present, but it is implemented in project Lombok which is connected as a plug-in  and  and allows to receive transparently this and (other functionality) in Java. B>>> 2) is in the nearest plans _>> Well i.e. as I and spoke, misses at present in Java. B> You still spoke as will not be Are not present, I said that already never catches up also with it and will be. Because even if in Java sometime implement described in the given subject  by this time in normally developing languages appears still in 10 times more  of which in Java now even cannot think. B>>> 4) it is implemented for a tick, it is not compatible with i18n B>>> 5) Like is not present, on the other hand it is not clear, on a horse-radish generally it is necessary _>> Well i.e. as I and spoke, misses at present in Java. Whether B> Mahlo what . . That misses Java that and is good that does not pull in a mouth all successively, and seven times measure and once cut. , well if you do not understand than for example high-grade pattern matching differs from poor switch' from Java and it carries what huge pluses to the programmer that I can do with it... By the way in real world Java (instead of in your personal, invented), it quite understand and try to consider variants http://cr.openjdk.java.net/~briangoetz/ … match.html though quits ridiculously enough. You can compare, as it looks in comparing with for a long time already operating analog from Swift'a https://appventure.me/2015/08/20/swift- … n-detail/. _>> If the lack lies in the foundation architecture language it from it does not cease to be a lack. B> Java became language number 1 just exactly because did not go on the way With ++ and all these advantages  feet fast methods for all tastes. Whom it sharply does not suffice, stamp back in With ++ and leave Java alone. And it closes for Java the whole huge sectors of software development. B>>> As a result, only 2) and 3) it is a pity that, no. Thus only on 3) there are no sounded plans, and 2) already saw. _>> as a result of Java is become numb and backward  which managed to lag behind in the development even the old man of a C ++. B> Java became language number 1 just exactly because did not go on the way With ++ and all this programming in punctuation language. You probably did not understand, speech not about syntax or something similar, and about dynamics of development of language. Compare itself, for example when in languages C#, Cs ++, Java have been added a lambda-function (in the modern languages of type Swift, Rust, etc. they it is natural to eat from a birth). _>> Thus in world JVM there are modern competitive decisions (Scala for example), but Java to them the relation is no. B> that approximately shows that in the world of Java you are guided approximately in any way. I hope, in Swift, Raste and With ++ is more, differently all this discussion of sense had no. You can want to compare Scala and Java by possibilities?

71

Re: Bucket

_> Well so specific code samples will be?) at me small  were still in initial the message... import static java.util.stream. IntStream.range; import static java.util.stream. IntStream.iterate; range (1,5).map (...) iterate (3, i-> i ++).map (...) import static java.util. Optional.ofNullable; ofNullable (...).map (...).map (...).orElse (...) hashMap.entries ().stream ().map ((k, v)->...) _> Is not present, I said that already never catches up also with it and will be. Because even if in Java sometime implement described in the given subject  by this time in normally developing languages appears still in 10 times more  of which in Java now even cannot think. Value-types add this or next year, and there will be almost ideal language. And when add tuples - becomes ideal. B>>>> 4) it is implemented for a tick, it is not compatible with i18n B>>>> 5) Like is not present, on the other hand it is not clear, on a horse-radish generally it is necessary _>>> Well i.e. as I and spoke, misses at present in Java. Whether B>> Mahlo what . That misses Java that and is good that does not pull in a mouth all successively, and seven times measure and once cut. _> Hy, well if you do not understand than for example high-grade pattern matching differs from poor switch' from Java and it carries what huge pluses to the programmer that I can do with it... Carries Nothing. I on the Rock, a pattern  programmed a thing quite good, but its usefulness is strongly restricted and in 99 % projects that with it that without it - differences are not present. _>>> If the lack lies in the foundation architecture language it from it does not cease to be a lack. B>> Java became language number 1 just exactly because did not go on the way With ++ and all these advantages  feet fast methods for all tastes. Whom it sharply does not suffice, stamp back in With ++ and leave Java alone. _> and it closes for Java the whole huge sectors of software development. General-purpose  is not necessary. _> you probably did not understand, speech not about syntax or something similar, and about dynamics of development of language. Compare itself, for example when in languages C#, Cs ++, Java have been added a lambda-function (in the modern languages of type Swift, Rust, etc. they it is natural to eat from a birth). Well, compared. Swift and Rast  a tuna. Also what? C# - 2007 Cs ++ - 2011 Java SE 8 (March 18, 2014) Swift reached the 1.0 milestone on September 9, 2014 Rust 1.0, the first stable release, was released on May 15, 2015 _> Can you want to compare Scala and Java by possibilities? The rock went on the way With ++ and for this reason it appeared to nobody is necessary. Java and Kotlin already never allow to it to type popularity, it is more than, is now. , most likely, too was late, but, probably, it can be untwisted on sheaf Android Studio + Android.

72

Re: Bucket

73

Re: Bucket

Hello, alex_public, you wrote: _> Hello, bzig, you wrote: _>>> Well so specific code samples will be?) import static java.util. Optional.empty; import static java.util. Optional.of; import java.util. Map; import java.util. Optional; import java.util.function. Function; import java.util.function. Supplier; import java.util.stream. IntStream; public class Test {public static <T> Optional <T> _try (Supplier <Optional <T>> f) {try {return f.get ();} catch (Exception ex) {return empty ();}} public static void main (String [] args) {//1 int [] ints = {1,3,5,7}; IntStream.of (ints).skip (1).limit (2).forEach (System.out:: println);//2 Map <String, Double> m = Map.of ("k1", 3.14, "k2", 2.72);//Java9, the rest is Java8 m.forEach ((k, v)-> System.out.println (k + ":" +v));//3 Function <Integer, Integer> f = i-> {if (i <0) throw new IllegalArgumentException (); else return i;}; Function <Integer, Optional <Integer>> g = i-> i <40? empty (): of (i-40); Function <Integer, Integer> h = i-> i / 2; System.out.println (_try (()-> of (42).map (f).flatMap (g).map (h))); System.out.println (_try (()-> of (2).map (f).flatMap (g).map (h))); System.out.println (_try (()-> of (-42).map (f).flatMap (g).map (h)));}} Autput 3 5 k1: 3.14 k2: 2.72 Optional [1] Optional.empty Optional.empty _> It without an automatic output of types?))) will be in 2 months http://openjdk.java.net/projects/jdk/10/ _> Well i.e. itself you see that Java the lagging most behind language. It seemed to me that the lagging most behind language from five is Rast. But at you to see, any calendar in which 2015 goes before 2014. _> "Scala it is necessary to nobody" is certainly interesting judgement...) )) Where I such told? The rock has the tiny share and any more does not increase it by one percent. And here reduction is inevitable.

74

Re: Bucket

75

Re: Bucket