- Algorithms in C++ - 8 Updates
- Why do I post about Jesus? - 2 Updates
- "C++ - Unicode Encoding Conversions with STL Strings and Win32 APIs" - 1 Update
| woodbrian77@gmail.com: Sep 19 05:43PM -0700 On Monday, September 19, 2016 at 4:18:44 PM UTC-5, Mr Flibble wrote: > }; > } instance; > int main() {} That's a mess compared to a "Hello, World" that uses main(). Brian Ebenezer Enterprises - In G-d we trust. http://webEbenezer.net |
| David Brown <david.brown@hesbynett.no>: Sep 20 08:44AM +0200 >> } instance; >> int main() {} > That's a mess compared to a "Hello, World" that uses main(). I don't think Mr. Flibble was suggesting it was a good idea to replace main() with the constructor of a static object - merely that it is /possible/ to do so. In particular, such constructors let you do useful work before main() is called, and it is possible that an issue in such constructors means that main() is /never/ called. And in a freestanding environment, the toolchain can happily have no main() function (the "main" function can have any name), and you may have many methods for calling code before a "main" function is run. |
| "Öö Tiib" <ootiib@hot.ee>: Sep 20 04:45AM -0700 On Tuesday, 20 September 2016 09:44:40 UTC+3, David Brown wrote: > /possible/ to do so. In particular, such constructors let you do useful > work before main() is called, and it is possible that an issue in such > constructors means that main() is /never/ called. Neither woodbrian nor Flibble wrote if it is good or bad idea. Flibble seemingly wrote that it is possible and woodbrian seemingly wrote that it has messy style. IMHO it is an idea to avoid. Calling order of such static object (and static data member object) constructors is defined only within single compilation unit that defines those. In project that matters there are lot of compilation units. So when the statics are defined in several compilation units and the initialization order matters then things will become quite hairy. The reasons how that order matters may pop up in late stages of development and the "best" idioms of dealing with such fiasco (like Schwarz counter) look quite confusing and will result with pointless code generated. Additionally such statics are source of synchronization issues and later there can be destruction order fiasco as well. ;) So I typically try to avoid the statics. When these are unavoidable then I try to make these with maximally lightweight and independent construction. When more sophisticated initialization is needed then I try to dodge it by adding lazy or explicit additional initialization steps that will run after 'main' is called. > And in a freestanding environment, the toolchain can happily have no > main() function (the "main" function can have any name), and you may > have many methods for calling code before a "main" function is run. C++ standard has seemingly no good policy for choosing what belongs into freestanding library and what does not. It seems to require more than needed; also there are no mechanism of indicating optional features. As result what is "freestanding" feels incomplete and inconsistent. In practice it is some non-conforming slice of C++ and so it is hard to build any arguments upon it. |
| David Brown <david.brown@hesbynett.no>: Sep 20 03:08PM +0200 On 20/09/16 13:45, Öö Tiib wrote: > compilation units. So when the statics are defined in several > compilation units and the initialization order matters then > things will become quite hairy. Yes indeed. It would be nice if there had been more control over this, but I understand where the limitation comes from. (It would also have been nice if it were possible to declare function-static objects, but have them constructed/initialised before main() rather than when first used.) Even compiler extensions (such as gcc's init_priority attribute) leave something to be desired - it uses numbers, and has no way to say "initialise this object before that one" or "initialise that object before this one". It would all be much easier if the compiler could handle this better. Perhaps compilers could spot common idioms (such as returning a reference to a static object) and turn them into carefully ordered pre-main initialisation. |
| "Öö Tiib" <ootiib@hot.ee>: Sep 20 07:37AM -0700 On Tuesday, 20 September 2016 16:09:14 UTC+3, David Brown wrote: > something to be desired - it uses numbers, and has no way to say > "initialise this object before that one" or "initialise that object > before this one". Also it AFAIK does not affect initialization order between different shared libraries. It is comical how so lot of things use modules written in C++ while C++ lacks a concept of modules whatsoever. > Perhaps compilers could spot common idioms (such as returning a > reference to a static object) and turn them into carefully ordered > pre-main initialisation. You probably did mean linker here. I would be happy even with diagnostic that whatever order it mechanically picked results with usage of not yet constructed object. |
| Mr Flibble <flibbleREMOVETHISBIT@i42.co.uk>: Sep 20 06:34PM +0100 On 20/09/2016 12:45, Öö Tiib wrote: > compilation units. So when the statics are defined in several > compilation units and the initialization order matters then > things will become quite hairy. [snip] But if you only have one such static object in your program then there is no problem: you could use Meyers' Singleton for other objects. /Flibble |
| David Brown <david.brown@hesbynett.no>: Sep 20 07:35PM +0200 On 20/09/16 16:37, Öö Tiib wrote: > Also it AFAIK does not affect initialization order between different > shared libraries. It is comical how so lot of things use modules > written in C++ while C++ lacks a concept of modules whatsoever. One day, we will have proper modules in C++. I believe both clang and MSVC have experimental module systems (I was going to say "module concepts", but that would add to confusion with another overdue feature of C++). But until one of these, or a combination, is picked as the standard, it is not possible to work with them for anything more than experimentation. > You probably did mean linker here. I would be happy even with > diagnostic that whatever order it mechanically picked results with > usage of not yet constructed object. I really meant "toolchain", as I suspect this would need support from both the compiler and the linker. Certainly a useful diagnostic that things are not as you want them to be is much better than silently wrong code! |
| "A. Bolmarcich" <aggedor@bitbucket.net>: Sep 20 06:54PM > red. >> However, Wirth rewrote it in Modula-2 or Oberon (I don't remember which of >> these two languages), and that version is available. About ten years after "Algorithms + Data Structures = Programs" was published, a new edition was published with the title "Algorithms & Data Structures". According the to Preface of the new edition: "The major change which prevades the entire text concerns the programming language used to express the algorithms. Pascal has been replaced by Modula-2." --- news://freenews.netfront.net/ - complaints: news@netfront.net --- |
| Juha Nieminen <nospam@thanks.invalid>: Sep 20 06:03AM > What will Hell feel like? When will you get into your thick skull that your words have no magic powers, even though in your delusions you think they have? You quoting scripture will not magically influence people. They are just empty words. Why do you keep making these posts? Because it feeds your own self-righteousness, you hypocrite. --- news://freenews.netfront.net/ - complaints: news@netfront.net --- |
| "Rick C. Hodgin" <rick.c.hodgin@gmail.com>: Sep 20 07:30AM -0700 On Tuesday, September 20, 2016 at 2:03:35 AM UTC-4, Juha Nieminen wrote: > > What will Hell feel like? > When will you get into your thick skull that your words have no magic > powers, I have no power whatsoever. I cannot even teach people those things I am espousing. It comes from within as God changes the person so that they themselves will then know that what I am espousing is the truth. It is not me convincing anyone, but it is me speaking the truth so that those whom God has changed will know within themselves, as by the very imparting of the knowledge of God and of the truth of His Son Jesus Christ, that they are a sinner, that they need to be forgiven, and that only Jesus Christ can forgive them. I speak the truth to you and everyone else. It is not my job to make you believe it. It is only my job to speak it accurately. > even though in your delusions you think they have? You quoting > scripture will not magically influence people. They are just empty > words. Consider your position alongside the teachings of Jesus Christ: http://biblehub.com/kjv/john/6.htm 63 It is the spirit that quickeneth; the flesh profiteth nothing: the words that I speak unto you, they are spirit, and they are life. 64 But there are some of you that believe not. For Jesus knew from the beginning who they were that believed not, and who should betray him. 65 And he said, Therefore said I unto you, that no man can come unto me, except it were given unto him of my Father. It is not me, Juha. I point you not to me, not to a church, not to a denomination, but to Jesus Christ Himself. I point you to truth, and I ask you to investigate it ALL for yourself. You do not need me or anyone else to teach you these things, but they are all before you if you will only pursue them. God has given us the Bible, and protected it, so that we can have a solid foundation upon which to return, so that His Holy Spirit aligns with the word He has given us, because Satan can also appear as an angel of light and leads many people falsely who think they are following after God, but are not. But by verifying our actions with the counsel of many, and holding those actions to the guidance of scripture and the counsel of many, we can be assured that we are moving as we should. > Why do you keep making these posts? Because it feeds your own > self-righteousness, you hypocrite. Because I have a burden for the lost souls of this world. Hell is a place of ongoing fiery torment that will never end. It is a containment system for those who are not operating in truth, but are operating in in sin, in falseness. We are all eternal beings, and we will endure on to eternity. My purpose in writing all of this is so you will continue to shine and thrive in eternity. That you will prosper greatly ongoingly. I do not want the end of you to come in the fiery place of torment. I want you to have the full future God has carved out for you, if you will receive it. In a sentence: I have a love for you (and others) because God's love is real, and has changed me, and now lives and dwells within me. It is that love God has for His creation which resides in me. And it is from within that love that I move, because His burden is unto you and your eternal soul, and this has now also become my burden because He has redeemed me, and His Holy Spirit dwells within me. Jesus said: http://biblehub.com/kjv/revelation/3-20.htm 20 Behold, I stand at the door, and knock: if any man hear my voice, and open the door, I will come in to him, and will sup with him, and he with me. It's very difficult to ignore a knocking on the door from within a house or apartment. Jesus is constantly knocking upon your life, your mind, your soul, your thoughts. He is there in the world, in the beauty, in the love, in the things done in truth, out of kindness and generosity, and He is there in the words I speak, and those others like me speak when we point you to Him. He stands at the door and knocks, and will receive everyone who opens the door and receives Him. He is God Almighty, with all possessions of the entire universe under His control and command. And the Bible teaches us that we will be married to Him in eternity, such that His possessions will also be our possessions, as it is here on Earth after the manner of marriage, where all that's hers becomes his, and all that's his becomes hers, the Earthly pattern of the eternal marriage to Jesus Christ by the collective church. It has been God's plan from the beginning. And it will endure forever. It is a plan well executed. And all who will be saved will be with Him, possessing also the entire universe as their own. And all who have rejected truth, rejected Him, will be contained in the place of fiery torment, where they can no longer interact with the rest of God's creation, for they in their sin and falseness are as a cancer unto truth, tearing it down continually, replacing solid foundations with diseased weakness. ----- I posted in the comp.lang.c group just now (slightly redacted): You have value. Your worth is measured as being well above that of finest purest gold. I want you to continue on after you leave this world, shining and thriving forever. I want your future to be secure that you may prosper greatly. Jesus makes it possible. It is my prayer for you, and for everyone in this group. ----- It is my ongoing prayer for you, and for many others. All the people I encounter in my life actually. Even my "enemies." God's love living inside of me makes it possible. It is not a Rick-thing, but it is the life God has given me through Jesus Christ welling up unto this ongoing outpouring of love and prayer from within the hope I now have assurance of through Him. It is ALL God. It is not me. I too am just a recipient. But what He has given me is available to each of you. All you have to do is seek the truth, and you too will discover what I have found. And my prayer is that you will do even more for His Kingdom than I have done, that you will outshine me so that you might even encourage me to do more. Best regards, Rick C. Hodgin |
| Tim Rentsch <txr@alumni.caltech.edu>: Sep 19 06:23PM -0700 > Well, after checking around it seems to me that it's likely I was > wrong about 16-bit `wchar_t` being invalid for Unicode. > So, punch me. I have no complaints with any of your comments. You were conveying a result that was true to the best of your understanding. Indeed it may very well have proven to be true. I certainly am not an expert in the C++ standard, so I asked about it, that's all. No one should be faulted for not knowing all the myriad implications of the C++ standard, which is a daunting document, and I didn't mean to imply anything along those lines. > But I don't know that, and it's certainly not my private opinion: I > got it from others, that I thought were experts in the area: a ?fact? > that I was pretty sure of. Yes, that was my impression, and I appreciate that you did. My question was meant only as a question. > <url: > http://stackoverflow.com/questions/39548465/is-16-bit-wchar-t-formally-valid-for-representing-full-unicode> > Hopefully some convincing answers will be forthcoming there. Excellent! I will take a look. |
| You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page. To unsubscribe from this group and stop receiving emails from it send an email to comp.lang.c+++unsubscribe@googlegroups.com. |
No comments:
Post a Comment