Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Tue, 7 Jan 2014 08:06:45 +0100 Martin Schreiber mse00...@gmail.com wrote: Pattern recognition... I don't care about it when I *use* a compiler... Maybe there is a misunderstanding, I mean pattern recognition by reading the code by a human. :-) Yes, there has been. I don't know if you read the MSElang Wiki, it is here: https://gitorious.org/mseide-msegui/pages/Mselang Yes I did read it. The design goals of MSElang: 1. Simple, 2. Readable. MSElang programs should be read like a letter. These two points are priorities for me too. This is why: 1. (Simple) I want automatic type conversion, but in a clean context. 2. (Readable like a letter). This is another point. I took inspiration from spoken language, which has little punctuation. Characters like $%?^| should be avoided if possible, contrary to C. For a human, layout is important, so I make it coincide to the compiler. 3.Easy to learn ... suitable for pupils Not bad. But very difficult. I have used several languages, and every time I switched, I crashed on pitfalls. Oh, stupid things, but every language has them. There are two types of pupils: those who don't know programming, and those who know. For the formers, any reasonable language can do, but perhaps little formalism is better (perhaps). For those alreay in programming, may be your idea (a strict and polished pascal) could be better at first, but after the first steps productivity would slow down. Don't take it wrong, but after reading your wiki, I had the impression that you make a strong point of formalism (nothing wrong with this), but I don't see the...: 4.Powerful ... pointers and pointer arithmetic. Object orientated high level programming. There are two different concepts of poweful. One, old style: while ((ch=getchar()) != EOF) putchar(ch); ...and assembler is even more powerful, for this. The other style of power is something like: send stream1 every customer.name without pointers, cycles, arithmetics, tests. Of course, nothing forbid to implement pointers in a powerful language: a programmer can decide to use them or not. I don't want to make the language a fascinating intellectual puzzle there is enough challenge in finding of best algorithms and architecture of the programs one has to develop... And this is why I want a powerful but simple language! I don't want to always remember that I can't writeln(myboolean) because I must explicitly convert it to a str8. And consider that there are not only writelns; consider GTK: it wants utf8, but for some reason (speed? text files? external LCD?) the main program uses ascii, or mcs4 or whatever. Ok, your way to go still works, but I think this can no more be called power. The problem in with is if one uses a local variable which is later also added to the with'ed class or record. This is a very very strong point. In fact, when I first saw php, I was stunned by those variables always referenced with $, but later I understood. So, we are condamned to be verbose? No, with my colon trick. But even sticking to WITH, the way something else does, there is no problem: with form1 do .caption = caption .left = left ... Why the freepascal team did not change the semantic of WITH? For compatibility? Will you change it? 80% solved. For str8. Then there are utf8, utf16, mcs-4... so you will have su8(), su16(), smcs4()? If there are no implicit conversions function overloading works for function result types also. I assume the print() function uses utf-8? This is not clear to me: overloading works for function resul types also. If you can overload something else than arguments, let me know how do you do it - I am interested (of course, in making it automatic :-) ). Apart from this, I think it is no good to assume print() uses utf8, because often a program sets captions in GTK, or fields in SQL, or uses external libraries written by someone else. In fact, you say MSELang will support several types of strings, why? To interface with the world, or to burden the language with unnecessary things? I assume all those types are for interfacing with the world, so who know if external function xx uses utf8 or EBCDIC. We are criticing each other, but I think this can be useful. I have a question: does MSEIDE support automatic completion? I didn't tried the last version (because of dependency issues); in 2.8.6 a quick try was unsuccessful: if it is there, I don't see it. Regards, linuxfan -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Tue, 7 Jan 2014 12:17:03 +0500 Ivanko B ivankob4m...@gmail.com wrote: in pascal you can't write functions with variable number of arguments AFAIK, C uses va_list lfor that. And C uses the right-2-left stact order for procedures to handle va_list otherwise Pascal uses the left-2-right order making va_list handling problematic.Then writeln might be implemented as a consequence of single write-s for each argument No no, freepascal and delphi CAN do that. They have array of consts (or something like that) which you construct with [xx, yy, zz]. The net result is the same as C, and probably more clean. The problem is that you pass printf() or format() a string and a handful of arguments: printf(My name is %s and my age is %d, [myname, myage]); and there is no check on arguments, and no check on correspondance between arguments and positional specifiers %blabla in the string. In the example above, the first %s asks for a string, and the second %d for an integer. If you mistakely pass wrong arguments, or in wrong order myname and myage, results can be catastrophic! I tried to use the external function format() in linux libc from pascal, and I didn't succeed. The problem was that, contrary to what was stated in the manual, open array of consts [...] was not passed to libc in a working way. I have no problems with delphi, instead; I often use format() in windows, and every time I do I feel the deprecating look of Martin behind my shoulders... :-) Regards, linuxfan -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Tuesday 07 January 2014 10:07:25 Martin Schreiber wrote: There is code completion for classes. There is no Lazarus-style code completion. That is the first thing I switch off in any IDE. ;-) What works is code navigation by Ctrl+LeftClick, procedure parameter listing by Shift+Ctrl+Space after the ( and header-implementation switch by Ctrl+Up, Ctrl+Down. I forgot, there is a sophisticated code template functionality too. I never use it, that thing I made especially for Graeme. :-) Martin -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On 01/06/2014 01:00 PM, Martin Schreiber wrote: with l:lab1 do l.left:= 10; l.top:= 20; end; I do like the quoted with as same eliminates the ambiguity that is inheritant to with (hiding part of the namespace of outer with, the class and globals). To me with l:=lab1 would be more logical as it works as if creating a variable of the appropriate type and assigning that value to it. -Michael -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On 01/07/2014 08:17 AM, Ivanko B wrote: in pascal you can't write functions with variable number of arguments AFAIK, C uses va_list lfor that. Yep. I use it a lot. (Important for C: there is a corresponding concept and syntax in Macros that supports this. ) In Pascal we have the concept of open arrays with differently typed variables for this, including the possibility to define constants of this type on the fly in function calls. This is at least equivalent. -Michael -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
Am 07.01.2014 12:24, schrieb Michael Schnell: On 01/06/2014 01:00 PM, Martin Schreiber wrote: with l:lab1 do l.left:= 10; l.top:= 20; end; I do like the quoted with as same eliminates the ambiguity that is inheritant to with (hiding part of the namespace of outer with, the class and globals). To me with l:=lab1 would be more logical as it works as if creating a variable of the appropriate type and assigning that value to it. The reason why : instead of :=: := is an assignment. with is fetching the address of a data structure which is the first step of an assignment. The first step of := is :. ;-) Martin -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On 01/07/2014 12:07 PM, Martin Schreiber wrote: The first step of := is :. ;-) Nice !:-) :-) :-) -Michael -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Mon, 6 Jan 2014 02:14:43 +0500
Ivanko B ivankob4m...@gmail.com wrote:
lab1.: create; left=10; top=20
It like the TCL, Python etc way. Mainly designed for GUI-progarmming
w/o any GUI-designer thus with smaller executables.
I think you missed the point. Real pascal programs
are filled with such things: create an object and set
a few properties, even without any GUI involved.
Some real pascal fragments, from lazarus sources:
lhelpcore.pas:
fServer := TSimpleIPCServer.Create(nil);
fServer.ServerID := fServerName;
fServer.Global := True;
fServer.Active := True;
reglazdaemon.pp:
AProject.AddPackageDependency('FCL');
AProject.AddPackageDependency('LCL');
AProject.AddPackageDependency('LazDaemon');
AProject.Title:='Daemon application';
AProject.LazCompilerOptions.Win32GraphicApp:=False;
AProject.ProjectInfoFile:='project1.lpi';
dbexport/...frmmain.pp:
DBFData.Close;
DBFData.TableName:=AFileName;
DBFData.Open;
printers/...winprinters.inc:
PDev:=TPrinterDevice.Create;
PDev.Name :=PPRINTER_INFO_2(InfoPrt)^.pPrinterName;
PDev.Driver:=PPRINTER_INFO_2(InfoPrt)^.pDriverName;
PDev.Port :=PPRINTER_INFO_2(InfoPrt)^.pPortName;
In all these fragments a WITH clause could have helped, but
I can imagine why it was not used. Besides, pascal is the
only language that implements with, and many people
criticize it. I don't think so, I like with, but I have
found another paradigm that in many cases is better.
This colon repetition is not limited to with. See
this other example, a set of declarations:
procedure DoSelectDir(Sender: TObject);
procedure DoSaveLayout(Sender: TObject);
procedure ReadConfig; virtual;
procedure WriteConfig; virtual;
which could be written as:
procedure
DoSelectDir(Sender: TObject);
DoSaveLayout(Sender: TObject);
ReadConfig; virtual;
WriteConfig; virtual;
end
In reality, with the different syntax I think of,
this set of declaration would be written like this:
procedure
DoSelectDir of TObject Sender
DoSaveLayout of TObject Sender
virtual
ReadConfig
WriteConfig
end
end
By reducing repetitions, the syntax pushes
toward a more clean and readable source.
Regards,
linuxfan
--
Rapidly troubleshoot problems before they affect your business. Most IT
organizations don't have a clear picture of how application performance
affects their revenue. With AppDynamics, you get 100% visibility into your
Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk
___
mseide-msegui-talk mailing list
mseide-msegui-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Sunday 05 January 2014 21:26:00 Linuxfan wrote: Hi folks, after 30 years of programming with different languages (basic, pascal, C, asm, tcl, php, bash...) I come to the conclusion that none of them is satisfactory for me. Sounds familiar. :) Although for me Delphi 7 is not so much away for my needs. I would like a language that needs few strokes on the keyboard, with little punctuation, but very precise, expressive, powerful and smart, without hidden features or pitfalls. Yup, mostly the intention of MSElang too, I like to add readability. ... To better explain, I will make two examples on how heretic I am. Let's talk about verbosity: var lab1: Tlabel; lab1 := Tlabel.create; lab1.left := 10; lab1.top := 20; ... Here, lab1 is cited 3 times, not counting the declaration. A WITH clause would help little: lab1 := TLabel.create; with lab1 do begin left := 10; top := 20 end Now lab1 is cited 2 times, but with, begin end have come in. What about of, instead: lab1.: create; left=10; top=20 The trick is the colon. It means repeat from the start of the line, until the physical line end. What I do not like in this approach is the dependency of semantic by the layout. For MSElang it is a no-go because MSElang is layout independent with some minor exceptions (//-comments for example). In MSEgui with-statements are used often. I know that most Delphi experts preach that with is evil, I don't think so. MSElang has the mandatory alias name in with statements in order to avoid overlooked name clashes. In MSElang the example would be written as lab1:= tlabel.create(); with l:lab1 do l.left:= 10; l.top:= 20; end; The example makes not much sense because writing lab1.left:= 10; lab1.right:= 10; is faster. In MSEgui initializing properties of a classinstance is not used often because the values are streamed and set in object inspector of MSEide. This is a task for the lexer/parser; the compiler doesn't see the colon. In this example the compiler, reading from the parser, reads lab1, ., then create, then ;. A complete statement. A note about this: lab1.create is interpreted as lab1 := Tlabel.create (what else could be this statement: a constructor invocation without assignment?). This also is against the philosophy of MSElang. Things what happen should be visible in code and they should look always the same. If there is an assignment, write it as assignment. Then the compiler goes on for a new statement and, because the parser still has a colon pending, the compiler reads lab1.left=10;. Then it reads lab1.top=20. Then the physical line ends, so the parser frees the colon. (You also see = instead of :=). This idea goes even further because the colon can be implicit: when the compiler expects more tokens, but they are not there, an implicit colon is assumed: Against the MSElang design principles too, things should always look the same. In MSElang the ';' for example becomes a mandatory statement terminator. It helps to separate the statements while reading the code. In procedure headers the ';' has been replaced by ',' so in header definition and procedure call parameter separators look similar. var // after a var keyword, shouldn't be an identifier? // but there is not: so the compiler tells the parser // assume an implied colon int x // here the compiler reads var int x int y // and here var int y end // this end closes the (implicit) colon This paradigm is simpler and more consistent than the normal pascal way. A pascal compiler, while parsing a var block, must always check that the next token is not a keyword (procedure, function, type, absolute, external...). I don't understand the motivation. To solve a fascinating intellectual puzzle? This is another thing I like to avoid in MSElang. Terminating 'var' by 'end' is a possibility I have to think about. I assume same applies to 'const' and 'type'? At a certain point, freepascal had to forbid declaring object variables after its methods, just because of that ambiguity. I don't understand, please explain. You may ask why the type specifier stays before the identifier, as in C?. Because I think it is better, but the explanation is long. I don't think it is better. When one searches a variable definition in code normally one want to find the variable name to check the type not the opposite. So scanning the name at the left edge of the editor window is easier than to follow the gap between type and name. Second heretic idea. One of the best things of pascal is its strong typing system. But it is not strong enough, especially when some kind of automatic type conversion is desired. Let assume we write a procedure print to output human readable data. We want to print messages, numbers, and perhaps
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
In MSElang the example would be written as
lab1:= tlabel.create();
with l:lab1 do
l.left:= 10;
l.top:= 20;
end;
==
You probaly don't have TCL experience :
button .btn1 -caption Exit -command {exit}
pack .btn1 -padx 5 -pady 5
that's single line per whole definition - reduces definition length
(an important kind of readability) very much.
--
Rapidly troubleshoot problems before they affect your business. Most IT
organizations don't have a clear picture of how application performance
affects their revenue. With AppDynamics, you get 100% visibility into your
Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk
___
mseide-msegui-talk mailing list
mseide-msegui-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Mon, 6 Jan 2014 19:41:22 +0500
Ivanko B ivankob4m...@gmail.com wrote:
In MSElang the example would be written as
lab1:= tlabel.create();
with l:lab1 do
l.left:= 10;
l.top:= 20;
end;
==
You probaly don't have TCL experience :
button .btn1 -caption Exit -command {exit}
pack .btn1 -padx 5 -pady 5
that's single line per whole definition - reduces definition length
(an important kind of readability) very much.
I don't know to whom this comment is directed, because nobody
criticized tcl until now. But to be very precise, these are TWO lines
of code, because without pack the button is useless; while the pascal
source specify geometry with left/top.
Anyway, I find tcl is a very respectable language, and its power goes
far beyond many people think. In designing my language I was inspired
by optional arguments like -caption, -command and so on. Especially
the last one -command, is concise and pwerful (the power of Tcl!).
Unfortunately it has several defect (but all is debatable...).
Cheers,
linuxfan
--
Rapidly troubleshoot problems before they affect your business. Most IT
organizations don't have a clear picture of how application performance
affects their revenue. With AppDynamics, you get 100% visibility into your
Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk
___
mseide-msegui-talk mailing list
mseide-msegui-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
I don't think it is better... Ouch. Another design principle of MSElang is... [...] === Quite opposite preferentions of two long-time exptrienced programmers. Linuxfan hates typing, Martin (anb me too) hates ambiguity irregularity :) Less typing means more ambiguity and vie versa. -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Monday 06 January 2014 17:27:32 Ivanko B wrote: I don't think it is better... Ouch. Another design principle of MSElang is... [...] === Quite opposite preferentions of two long-time exptrienced programmers. It seems so. :-) Martin -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
print counter+12 // counter is an integer === Is the current writeln(counter,12) worse ? Why ? But writeln definetely needs variations with C-style *printf %-formatting: writeln(%s%d,counter,12); -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Mon, 6 Jan 2014 21:27:32 +0500 Ivanko B ivankob4m...@gmail.com wrote: I don't think it is better... Ouch. Another design principle of MSElang is... [...] === Quite opposite preferentions of two long-time exptrienced programmers. Linuxfan hates typing, Martin (anb me too) hates ambiguity irregularity :) Less typing means more ambiguity and vie versa. I don't want to speak too much - I will slow down I promise - but less typing is not necessarily more ambiguity, I think. It depends. There is no ambiguity when There is exactly one way to interpret each program statement syntactically. That's all. For sure, to fill a language with keywords and punctuation does not render it less ambiguous; it helps in catching typing errors, probably. C compilers tend to throw generic errors (parse error at near...) in wrong places even if the language itself is not ambiguous. Pascal instead is very precise in this, and I appreciate it. print counter+12 // counter is an integer Is the current writeln(counter,12) worse ? Why? No no, writeln has nothing wrong. What is wrong is that in pascal you can't write functions with variable number of arguments, so write/writeln do violate pascal rules. But writeln definetely needs variations with C-style *printf %-formatting: writeln(%s%d,counter,12); This trick is very handy, especially for international programs, and you can do it in pascal too using function(..., [...]) (in theory; in practice I tried without success). It also has a big problem, and there is no solution: you can pass it whatever you want, even without corrispondence to the various %xx. When you run a program and you see strange messages like Error: %s\n, well, you know they used this trick and something went wrong! Regards, linuxfan -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Mon, 6 Jan 2014 18:55:42 +0100 Martin Schreiber mse00...@gmail.com wrote: On Monday 06 January 2014 16:34:31 Linuxfan wrote: I think that program sources are text files and always will be. ... I don't agree, sorry. Don't be sorry. It is apparent that we have different concepts, stop. The purpose is to let a the sub or method call always look the same because of pattern recognition. ...and later... A comma in MSElang is a separator, a semicolon is a terminator. And the same look of definition and call also because of pattern recognition. Pattern recognition... I don't care about it when I *use* a compiler; moreover, I think the compiler should work much more than today's compiler do. It is true that fast compilation is desirable, also. But computers are getting more and more powerful. The problem in with is if one uses a local variable which is later also added to the with'ed class or record. This is a point where we share the same idea! WITH *can* lead to problems, but if the programmer is careful, no problem. Even in my idea they always look the same, if you know the language. It always should look the same independent if one knows the language well or not. ;-) Hmmm... procedure DoSelectDir of TObject Sender DoSaveLayout of TObject Sender virtual ReadConfig WriteConfig end end This is for me the worst possible indention style, sorry. Actually the mail program has messed up the indentation, but I understand. Every person has its own tastes. Error: Fields cannot appear after a method or property definition, ... In MSElang modifiers are probably not separated by ;. And this is a good point. Why should it if it is not necessary? I don't think it is a too big burden to write f1:= 1.0 + f2 * 2.0; It can be, when not necessary. Who wants to make things clear, can always do, like you. But those that do not want? ... converting thenewtype to str8 ... and integer to str8 ... You always write s() because of function overloading. There are no ambiguities because MSElang has no implicit type conversions. 80% solved. For str8. Then there are utf8, utf16, mcs-4... so you will have su8(), su16(), smcs4()? The expectations are 10 times faster compilation times than FPC 2.6.2. MSELang is not enough modern. I don't care. It is a tool mainly for me. Well, what to say. I hope we both succeed in doing our language. With affect, I hope that one day there will be a debate in internet MSELAng vs Java vs Python vs C# vs Simple! (Simple is the name I chose for my language). Have a nice evening, regards, linuxfan -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
On Monday 06 January 2014 20:41:48 Linuxfan wrote: On Mon, 6 Jan 2014 18:55:42 +0100 The purpose is to let a the sub or method call always look the same because of pattern recognition. ...and later... A comma in MSElang is a separator, a semicolon is a terminator. And the same look of definition and call also because of pattern recognition. Pattern recognition... I don't care about it when I *use* a compiler; moreover, I think the compiler should work much more than today's compiler do. It is true that fast compilation is desirable, also. But computers are getting more and more powerful. Maybe there is a misunderstanding, I mean pattern recognition by reading the code by a human. :-) I don't know if you read the MSElang Wiki, it is here: https://gitorious.org/mseide-msegui/pages/Mselang The design goals of MSElang: The design goals Ultimate goal is to build the most productive universal programming environment, usable from programming of microcontrollers to MSEgui projects. I like pragmatic solutions but hate compromises. 1.Simple. Reduce the language concepts one has to learn to the minimum, make them as orthogonal as possible. 2.Readable. MSElang programs should be read like a letter. 3.Easy to learn. Because of 1. and 2. it should be suitable for pupils as first programming language. 4.Powerful Allow to go down to the bare metal, it has pointers and pointer arithmetic. Object orientated high level programming. 5.Fast running State of the art optimizations. 6.Fast compiling While defining the language keep in mind how it can be implemented efficiently. I don't want to make the language a fascinating intellectual puzzle there is enough challenge in finding of best algorithms and architecture of the programs one has to develop... The problem in with is if one uses a local variable which is later also added to the with'ed class or record. This is a point where we share the same idea! WITH *can* lead to problems, but if the programmer is careful, no problem. How with Pascal with can one be sure that a *newly* added field to a record will not hide a already used variable in an *existing* with statement? [...] ... converting thenewtype to str8 ... and integer to str8 ... You always write s() because of function overloading. There are no ambiguities because MSElang has no implicit type conversions. 80% solved. For str8. Then there are utf8, utf16, mcs-4... so you will have su8(), su16(), smcs4()? If there are no implicit conversions function overloading works for function result types also. I assume the print() function uses utf-8? Martin -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
in pascal you can't write functions with variable number of arguments AFAIK, C uses va_list lfor that. And C uses the right-2-left stact order for procedures to handle va_list otherwise Pascal uses the left-2-right order making va_list handling problematic.Then writeln might be implemented as a consequence of single write-s for each argument -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] MSELang heretic ideas
lab1.: create; left=10; top=20 It like the TCL, Python etc way. Mainly designed for GUI-progarmming w/o any GUI-designer thus with smaller executables. -- Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831iu=/4140/ostg.clktrk ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
