26

Re: Re: the lame logic

Hello, Marty, you wrote: __>> is such    that private members need to be designated with underlining - to or after. I saw the person who quite seriously told that it to it does the code more clearly - it is visible that private, and that there is no __>> a truth, in most  stages the suffix m for members of a class is normally used still and the for transferred parameters M> the Suffix "_m" is used at Microsoft for members of classes with any visibility. The same WTL, and MFC like too in-in, I also speak - most  generally, by the way, in  still usage   very strange looks. Truth, not always, but it looks stranger https://chromium.googlesource.com/exper … c_bezier.h double GetX1 () const; - , pascal double range_min () const {return range_min _;} - oh, no, Si

27

Re: Re: the lame logic

Hello, andrey.desman, you wrote: AD> Unique strangeness, it what at them the order not as in CSS (top, right, bottom, left) And whence, by the way, this order in CSS? Any it unobvious and not intuitive.

28

Re: Re: the lame logic

Hello, __ kot2, you wrote: __> underlinings in members of a class -  well are not worse m_

29

Re: Re: the lame logic

Hello, Ops, you wrote: __>> underlinings in members of a class -  Ops> well are not worse m_ For that matter that m_ better than simply underlining.... <<RSDN@Home 1.1.4 stable SR1 rev. 568>>

30

Re: Re: the lame logic

Hello, CreatorCray, you wrote: __>>> underlinings in members of a class -  Ops>> well are not worse m_ CC> For that matter that m_ better than simply underlining. All is talks about shit sorts. No prefixes do the code more readably. They appear when the person opens for itself the new concept - , interfaces or private members and starts to push everywhere first

31

Re: Re: the lame logic

Hello, __ kot2, you wrote: __> all is talks about shit sorts. No prefixes do the code more readably.  does if to use reasonably. In this case the prefix allows to understand in what variable in the given place we climb: a member of a class, local or, , global. It is a unique case when the prefix is comprehensible. As soon as would make the good circuit for syntax highlighting which this question qualitatively solved without code transformation in  is an application too withers a natural way.... <<RSDN@Home 1.1.4 stable SR1 rev. 568>>

32

Re: Re: the lame logic

Hello, CreatorCray, you wrote: CC> In this case the prefix allows to understand in what variable in the given place we climb: a member of a class, local or, , global. Taking into account that the code most part works with local variables, it was meaningful to do a prefix  not for them, and that all code in prefixes turns out. But nevertheless understand that it is a perversion because without classes such was not. Here also decided to enter underlinings for the private. CC> as soon as would make the good circuit for syntax highlighting which this question qualitatively solved without code transformation in  is an application too withers a natural way. At with ++ there is one fine property - it is possible  to return any as const and it will be only for reading certainly. I do not see any difference - to cause p.x () or to read directly p.x from const Point &p;  then  p.x () it is necessary? The unique case - is any sample code which always causes p.x () which for some  is calculated. But I do not think that it for this reason is made. To the people who have come from -C# it is the concept it is given hardly, but all the same it is necessary to use language tools, instead of to write in style of another generally, actually to   and  - one of the best projects on With ++. There is such gloom that directly blood from eyes leaks also anything - people use. But if to carp, I mine  on Point.h would look so: constexpr Point (): x _ (0), y _ (0) {} -  constexpr. What does it give? constexpr Point (int x, int y): x _ (x), y _ (y) {} - what for two designers when it is enough of one with  in parameters? constexpr int x () const {return x _;} - to remove  these methods, to make terms x and y public constexpr int y () const {return y _;} void set_x (int x) {x_ = x;} - A narcotism! void set_y (int y) {y_ = y;} void SetPoint (int x, int y) {- a heavy narcotism! What for the method copying implementation of the designer is necessary? Why not p = Point (10, 20), what for this generally is necessary set? x_ = x; y_ = y;} why   in With ++ the project? void SetToMin (const Point& other); - not clear hogwash. Why it is a method of class Point? std:: string ToString () const; - why it is a method of a class instead of  f-ija?

33

Re: Re: the lame logic

Hello, Ops, you wrote: AD>> Unique strangeness, it what at them the order not as in CSS (top, right, bottom, left) Ops> And whence, by the way, this order in CSS? Any it unobvious and not intuitive. .. Can walked on hours since the midnight/midday.

34

Re: Re: the lame logic

Hello, __ kot2, you wrote: CC>> In this case the prefix allows to understand in what variable in the given place we climb: a member of a class, local or, , global. __> taking into account that the code most part works with local variables, it was meaningful to do a prefix  not for them, and that all code in prefixes turns out. Well and it turns out. All local (function parameters - local) without a prefix, any  with m _, statics s _, global g_. More than any prefixes it is not necessary. __> here also decided to enter underlinings for the private.  it as that . What difference private or ? At them different containers, just and it is important to see it. __> at with ++ there is one fine property - it is possible  to return any as const and it will be only for reading certainly. __> I do not see any difference - to cause p.x () or to read directly p.x from const Point &p; Because no means always you have exceptional const an object reference and then in that place it is possible in p.x also to write that can cause at all that effect which was expected. The hole in the interface through which turns out it is possible  to spoil all. So is not present. __> to the people who have come from -C# it is the concept it is given hardly, but all the same it is necessary to use language tools, instead of to write in style of another in  is syntactic sugar , is called property. It is automatically created getter + setter which can be redefined. __> constexpr Point (): x _ (0), y _ (0) {} -  constexpr. What does it give? To me by the way too it is not clear. __> constexpr Point (int x, int y): x _ (x), y _ (y) {} - what for two designers when it is enough of one with  in parameters? Disputably, well it is fine. Here the question of stylistics, uniformity of interfaces and purity of the code in more difficult cases than this is faster. __> constexpr int x () const {return x _;} - to remove  these methods, to make terms x and y public __> constexpr int y () const {return y _;} Categorically is not present. Public fields in a class - one of signs . You want public fields - define struct. __> void set_x (int x) {x_ = x;} - a narcotism! __> void set_y (int y) {y_ = y;} No. __> void SetPoint (int x, int y) {- a heavy narcotism! What for the method copying implementation of the designer is necessary? Why not p = Point (10, 20), what for this generally is necessary set? __> x_ = x; __> y_ = y; __>} For a simple class with POD types yes, differences any, for heavier the difference starts to appear. If there is a task to adhere to uniform stylistics of interfaces for base classes which this explicitly concerns it is better to do uniformly - the compiler then all  all the same. __> why   in with ++ the project? And what to the compiler a difference? As how to name it is defined accepted in a command coding standard. __> void SetToMin (const Point& other); - not clear hogwash. Why it is a method of class Point? And why is not present? What does it do inside? __> std:: string ToString () const; - why it is a method of a class instead of  f-ija? And why it should be static function? It  it is inconvenient. If , it generally could be operator std:: string () const; Generally the branch is time in  for transferring.... <<RSDN@Home 1.1.4 stable SR1 rev. 568>>

35

Re: Re: the lame logic

Hello, __ kot2, you wrote: __> __> constexpr Point (): x _ (0), y _ (0) {} -  constexpr. What does it give? __> possibility to use in others constexpr?

36

Re: Re: the lame logic

Hello, CreatorCray, you wrote: CC>>> In this case the prefix allows to understand in what variable in the given place we climb: a member of a class, local or, , global. __>> taking into account that the code most part works with local variables, it was meaningful to do a prefix  not for them, and that all code in prefixes turns out. CC> well and it turns out. All local (function parameters - local) without a prefix, any  with m _, statics s _, global g_. More than any prefixes it is not necessary. For  there is a prefix:: about statically s_ the first time hear. To write m_ it how approximately to write _ before each noun - conveniently only to those who got used __>> I do not see any difference - to cause p.x () or to read directly p.x from const Point &p; CC> Because no means always you have exceptional const an object reference and then in that place it is possible in p.x also to write that can cause at all that effect which was expected. The hole in the interface through which turns out it is possible  to spoil all. CC> so is not present. And so the hole is the interface in a type set. That holes not was Point should be  __>> to the people who have come from -C# it is the concept it is given hardly, but all the same it is necessary to use language tools, instead of to write in style of another CC> in  is syntactic sugar , is called property. It is automatically created getter + setter which can be redefined. A hogwash it in comparison with const __>> constexpr Point (int x, int y): x _ (x), y _ (y) {} - what for two designers when it is enough of one with  in parameters? CC> it is disputable, well it is fine. Here the question of stylistics, uniformity of interfaces and purity of the code in more difficult cases than this is faster. It is possible  the political prohibition  values, but in this case pertinently in my opinion all the same to have  CC> Public fields in a class - one of signs . You want public fields - define struct. Certainly Point should be struct. There should not be in it nothing private __>> void set_x (int x) {x_ = x;} - a narcotism! __>> void set_y (int y) {y_ = y;} CC> Is not present. What is not present? A hole in the interface __>> why   in With ++ the project? CC> and what to the compiler a difference? It  to the compiler is not present a difference because it the register does not distinguish, and people suffer CC> As how to name is defined accepted in a command coding standard. __>> void SetToMin (const Point& other); - not clear hogwash. Why it is a method of class Point? CC> And why is not present? What does it do inside? And .. And what for such .. Methods in Point? __>> std:: string ToString () const; - why it is a method of a class instead of  f-ija? CC> And why it there should be a static function? It  it is inconvenient. CC> if , it generally could be operator std:: string () const; well, put tostring in a class directly could only dzhavist-sisharpnik. There should not be it a member of a class on mind

37

Re: Re: the lame logic

Hello, __ kot2, you wrote: __> that is not present? The hole in the interface It not a hole, is the certain contract. The hole it when any can walk there as to itself in a kitchen garden, for example to take the address from public field and to transfer devils , whence to it will write. __> it  to the compiler is not present a difference because it the register does not distinguish, and people suffer those strange mazo-ascetics who always types each letter hands Suffer only. __> and .. And what for such .. Methods in Point? Work strictly with Point where it still to be? __> well put tostring in a class directly could only dzhavist-sisharpnik. There should not be it a member of a class on mind What or the rational explanation will be?... <<RSDN@Home 1.1.4 stable SR1 rev. 568>>

38

Re: Re: the lame logic

Hello, __ kot2, you wrote: M>> But what for in members of a class? I too sometimes use them, when arguments of the designer/method on sense coincide with members of a class. Then I use underlinings for arguments as - in one place, and for members of a class them everywhere it is necessary to drag arguments. Yes, I know that the compiler (in case of the designer, at least) itself understands, but  eyes all the same inconveniently, therefore and I add __> is such    that private members need to be designated with underlining - to or after. I saw the person who quite seriously told that it to it does the code more clearly - it is visible that private, and that there is no __> a truth, in most  stages the suffix m for members of a class and the for the transferred parameters You is normally used still so is dashing all scarified (on my previous operation it was accepted mClassMember, on  - _classMember; Personally to me esthetically the first is more pleasant), what even I do not know, how you normally distinguish parameter from a member of a class.

39

Re: Re: the lame logic

Hello, __ kot2, you wrote: __> well and implementation of a class too the disputable. I personally would leave x and y public, than to fence gettery-setters to them Setters and  are necessary for normal redefinition in descendants. For example, in a setter it is possible to pull still the notification message someone about parameter change. Though in case of any simple things like Point/Rect I, of course, would do by their structures with all public fields, certainly.

40

Re: Re: the lame logic

Hello, Dair, you wrote: D> You so is dashing scarified all (on my previous operation it was accepted mClassMember, on  - _classMember; personally to me esthetically the first is more pleasant), what even I do not know, how you normally distinguish parameter from a member of a class. I do not know, how __ kot2, personally I - on names. Names such I try to give, that it was clear. At rare coincidence - I add underlining to parameters

41

Re: Re: the lame logic

Hello, __ kot2, you wrote: __> Hello, Marty, you wrote: __>>> is such    that private members need to be designated with underlining - to or after. I saw the person who quite seriously told that it to it does the code more clearly - it is visible that private, and that there is no __>>> a truth, in most  stages the suffix m for members of a class is normally used still and the for transferred parameters M>> the Suffix "_m" is used at Microsoft for members of classes with any visibility. The same WTL, and MFC like too __> in-in, I also speak - most  __> generally, by the way, in  still usage   very strange looks. Truth, not always, but it looks stranger In CBOSS requirements to code design were  a class Title is mandatory the letter designating its type begins with ' a C ', structures with ' S',  with "m _", in the beginning of a title of a variable: n - for int, b for bool, sz for an ASCIIZ-line, ' a ' for an array, "g _" for global variables, "gc _" for global constants m_bTerminating, m_nCount, m_szName, etc.

42

Re: Re: the lame logic

__> Hello, Kodt, you wrote:>> Basic library of geometry, all is simple and clear, a pastoral and idyll... __> underlinings in members of a class - . The person with two underlinings in  told

43

Re: Re: the lame logic

Hello, Kodt, you wrote: constexpr Insets (int top, int left, int bottom, int right) Bypass on a circle counter-clockwise? Normally pass on hour - left, top, right, bottom.

44

Re: Re: the lame logic

Hello, Maniacal, you wrote: M> In CBOSS requirements to code design were  aka Hungarian type notation M> a class Title is mandatory the letter designating its type begins with ' a C ', structures with ' S',  with "m _", in the beginning of a title of a variable: n - for int, b for bool, sz for an ASCIIZ-line, ' a ' for an array, "g _" for global variables, "gc _" for global constants M> m_bTerminating, m_nCount, m_szName, etc. It  is convenient and is practical! Though, for someone Archaically old horse can also - a furrow does not spoil!!!

45

Re: Re: the lame logic

Hello, Dair, you wrote: D> Hello, ononim, you wrote: D>>> So it insets, at them well is fine still x/y (instead of left/top and what to write instead of right/bottom?), but width/height at them precisely is not present. D>>> Nikolay, not so understood O>> the Main trick in that that ordinate go in advance abscissas, sorry May a service jacket. D> and, now that's something like it, really, a jamb. In CSS so. Possibly developers of chrome with CSS face also type record padding: 10px 5px; it it is clear, accordingly and the designer so will be more clear to be read. Though with 4-argumentnym the designer the order a bit wrong, it is visible its author CSS badly learned.

46

Re: Re: the lame logic

Hello, Ops, you wrote: AD>> Unique strangeness, it what at them the order not as in CSS (top, right, bottom, left) Ops> And whence, by the way, this order in CSS? Any it unobvious and not intuitive. Clockwise. 0 (12) hours, 3 hours, 6 hours, 9 hours. I do not think that any order will be more obvious.

47

Re: Re: the lame logic

Hello, Kodt, you wrote: And here suddenly constexpr Insets (int top, int left, int bottom, int right): top _ (top), left _ (left), bottom _ (bottom), right _ (right) {}... Dear experts! How you think, why I paid to this attention? Listing of indents counter-clockwise. Interesting, and they really support the negative indents?

48

Re: Re: the lame logic

Hello, Marty, you wrote: M> the Suffix "_m" is used at Microsoft for members of classes with any visibility. The same WTL, and MFC like too Only not a suffix, and a prefix, and not "_m", and "m _"

49

Re: Re: the lame logic

Hello, CreatorCray, you wrote: __>> that is not present? The hole in interface CC> It not a hole, is the certain contract. The hole it when any can walk there as to itself in a kitchen garden, for example to take the address from public field and to transfer devils , whence to it will write. It is thin enough edge, here it is possible to consider long. But that I see -  invented  in style C#-Java, and in With ++ instead of  for access control it is used const, and for changes  -  and controlers.  generally arose because of necessity   in classes. To change  through set () - the most simple and reliable method to receive  a system state, therefore it is necessary to watch closely it very much, and it is better generally  __>>  to the compiler there is no difference because it the register does not distinguish, and people suffer CC> those strange mazo-ascetics who always types each letter hands Suffer only. That is it is more convenient to push a point and then a mouse ? I about necessity to push  what for each time for a name dial-up in pascal why was so is made? Because at that time you could hammer on  and type the lower register - as to you more conveniently. Permanently to push  for a call of each method it just and there is a masochism __>> and .. And what for such .. Methods in Point? CC> Work strictly with Point where it still to be? That is let  the person adds the not clear methods in Point if to it so it is pleasant? So on  iterations there someone  generally Point:: draw () with  and opengl __>> well put tostring in a class directly could only dzhavist-sisharpnik. There should not be it a member of a class on mind CC> What or the rational explanation will be? Any who looked that such stl which is included into the standard With ++ should understand the concept of sharing of algorithms and the data. Point - the data. There should not be any tostring, draw, setmin, fromstring and by the way that designer with vast comments about compatibility with macos too on mind should not be. It is not necessary to litter a class of the data. Represent, if in vector each of committee would add any the f-ii, personally to it in its project the convenient? Too that-thread there about the designer for macos, an akoj-thread shuffle_step (), or set_min ()?

50

Re: Re: the lame logic

Hello, Dair, you wrote: D> You so is dashing scarified all (on my previous operation it was accepted mClassMember, on  - _classMember; personally to me esthetically the first is more pleasant), what even I do not know, how you normally distinguish parameter from a member of a class. If not to write data handling long  as terms of classes of the data (the same tostring () at point) no problems are present