XProfan

- Page 4 -

Frank
Abbing

to that first remaining these function yet undokumentiert consist. it exists no technischer reason, tappt im dunkeln To Remove.
and to that second becomes it in the next XProfan-Version (The möglicherweise XProfan 2008 is) a substitute give, the not only bequemer, separate too explicit faster is (cue: statisches left) and the DEF as API-Integration very quick forget can.

there will Dietmars Herzklappe sure quick again calm.

I sags again, I find it aale, if unnützer Balast time abgeschüttelt becomes. Rolands engineering such things in the code consist To let but from the guide To Remove stops I for a skilful train.

means so correctly. understand do I But nevertheless not, be because it's about Größenreduzierung the Runtime. i see it therefore not because Roland without further seinem Precompiler procure can old Commands in z.B. callByName(alterbefehl,...) transfiguring To let. Sozusagen a function callByName The old functions auffängt without the a Befehls-ID uses becomes.

RGH

@iF:

If I z.B. 20 create.....-functions through 1 function create(...... supplant have I 19 IDs gespart and can tappt im dunkeln on another place use. But this is only a Page the Medallie.
The second is the Übersichtlichkeit and/or Logik the Language and so Perhaps too The leichtere Erlernbarkeit withal zunehmender Komplexität. (here's sure too some Geschmacksache.) so have I these Containerfunktionen z.B. with ZIM (of/ one american - occupied/canadian data base-4GL) kennen and value learnt: any Settings get z.B. with of/ one function with numerous Subfunktionen tuned. If I means a attitude benötigte, knew I immediate, whom function To use was and where I in the Doku nachzuschlagen having and in the Source is immediate discernible, that it itself circa a attitude deals, because there a Setfunktion standing. (It's all right Yes not only ums write the Source code, separate too therefore, it later yet To understand.) as me then The IDs in Profan auszugehen drohten, fell me these engineering one, what about me setting tappt im dunkeln in Profan one (ex 7.0). (and the Erweiterbarkeit this Containerfunktionen through SUBPROC set the whole The crown on. the there were self in ZIM not!

)
to Übersichtlichkeit heard too, that I do not for a function several command/functions vein alternative Schreibweisen having would like. One example: first of all gift it whom Befehk MESSAGEBOX. around the squeezed Button To detect, was a Systemvariable abzufragen. the erwies itself later as unpraktisch and it watts The function MESSAGEBOX() introduced, The The squeezed Button zurückgab. in the welt the Ausmistens watts the commands now abgeschafft (1 ID watts spare). so but the middle-aged Source code the middle-aged Hasen still function (and Dietmars Herzklappe geschont becomes ;) ) GetControlParas (B the integrierte PreCompiler to the Compilieren/interpret at reading the Source code whom MESSAGEBOX-commands into MESSAGEBOX-Function to. Neueinsteiger must with the antiquated Messagebox-commands but not any more belästigt go, therefore missing it in the introduction Help. (in the attachment the antiquated command diving it but still on, so too the Neueinsteiger old Source code understand can.)
On these point try I, The Language moreover to develop, without The compatibility with Füßen To zurück.

Greeting
Roland

Intel Duo E8400 3,0 GHz / 4 GB RAM / 1000 GB HDD - ATI Radeon HD 4770 512 MB - Windows 7 Home Premium 32Bit - XProfan X4

Nico
Madysa

and therefore has yet none of us the Profan-program given.

Nico Madysa

- Page 5 -

Frank
Abbing

Blödsinn, it vanish over ands over again time renomierte people in the Versenkung. Z.B. Thorsten Haggert or Heiko Reddingius. want Yes time know, where The abgewandert are...

Nico
Madysa

I meant too sooner The, The here very active are, z.B... you!

Nico Madysa

Frank
Abbing

Reines Profan would me zuwenig, there be I sincere. In Symbiose with Assembler does it but fully aale. Have presently none urge to Veränderung.

Sebastian
König

Quote from the Helpfile of XProfan 10:

Importent Info: In künftigen XProfan-versions becomes Def only yet for Definition of Konstanten akin go.

in the serious:
to that first remaining these function yet undokumentiert consist. it exists no technischer reason, tappt im dunkeln To Remove.
and to that second becomes it in the next XProfan-Version (The möglicherweise XProfan 2008 is) a substitute give, the not only bequemer, separate too explicit faster is (cue: statisches left) and the DEF as API-Integration very quick forget can.

Hi,

i want my mustard too time add

:

The Entwöhnung the User of überholten Features through Streichen the Erwähnung in the Help stops I for a very good lane. and the static left as Substitute for API-DEFs sound really interestingly - i'm very tensely!

though would like here for plädieren, next to the DEFinition of Konstanten too furthermore The Possibility the Definition of singlelined functions in the Help documents To let. only ought to one the kid Perhaps a others names give: of my opinion to would macro very zutreffend.

Begründung: next to the Possibility, often recurring bills with only of/ one row (The Variante with of/ one colon-separated singlelined procedure is already More Tipparbeit...) elegant abzuküren, there too a handfesten Geschwindigkeitsvorteil the DEF-Variante! Folgender code shows by me a profit of ca. 25% in the macro-Variante (16172 ms To 21500 ms, XProfan 10.0a, compiled):
Compile Mark Separation

DEF MulDivMacro(3) (&(1) * &(2) / &(3))

proc MulDivFn

    parameters x&,y&,z&
    return x& * y& / z&

endproc

DEF &Count 500000
Cls
Randomize
declare start&,stop&
declare tmp&
print Rechne...
start& = &GetTickCount

whileloop &Count

    tmp& = MulDivMacro(Rnd(5000),Rnd(6000),Rnd(7000)+1)

wend

stop& = &GetTickCount
print Zeit mit Makro: ,int(stop& - start&),ms
print Rechne...
start& = &GetTickCount

whileloop &Count

    tmp& = MulDivFn(Rnd(5000),Rnd(6000),Rnd(7000)+1)

wend

stop& = &GetTickCount
print Zeit mit Funktion: ,int(stop& - start&),ms
waitkey
end

MfG

Sebastian

Windows XP, XProfan/Profan² 4.5 bis 11
Profan2Cpp-Homepage: [...]

Alte Profan²-Seite: [...]

09/08/07 ▲

Sebastian
König

Hmm... there quite no opinions to (my Posting directly over this, my I)?

Roland?

Windows XP, XProfan/Profan² 4.5 bis 11
Profan2Cpp-Homepage: [...]

Alte Profan²-Seite: [...]

09/09/07 ▲

09/10/07 ▲