PureBasic Forums - English

maybe you explain what is your goal.
Your code looks complicated and it is not the best for debugging.
I'm sure there are more clear solutions for your problem.

(In case, please forgive my dumbness :D) What "to statements"? Here your examples always give 0.0 :(
Top




; this 2 lines in the Macro return different values - but should be the same
Macro IIFn(_Expression_, _TruePart_, _FalsePart_)
Bool(Not _Expression_) * _FalsePart_ + Bool(_Expression ...

EnableExplicit

Macro IIFn(_Expression_, _TruePart_, _FalsePart_)
; Bool(Not _Expression_) * _FalsePart_ + Bool(_Expression_) * _TruePart_
Bool(_Expression_) * _TruePart_ + Bool(Not _Expression_) * _FalsePart_
EndMacro

Debug IIFn(1=0, 11.1, 10) ; should be 10, gives 0.0!

Define res.d ...

Macro IIFn(_Expression_, _TruePart_, _FalsePart_)
Bool(_Expression_)*_TruePart_ + Bool(Not(_Expression_))*_FalsePart_
EndMacro
Debug IIFn(1=0, 11.1, 10) ; should be 10, gives 0.0!

seems to be a PB-Bug!
If I change the to statements it works

EnableExplicit

Macro IIFn(_Expression_, _TruePart ...

here a collection of different IIF Macros



; IIF : similar to the VB IIf Function
; use it : IIf(varReturn, A>1000, "large", "small")
Macro IIF(varReturn, _Expression_, valTrue, valFalse)
If Bool(_Expression_)
varReturn = valTrue
Else
varReturn = valFalse
EndIf
EndMacro

;IIFe ...

k but, I still get zero when I reduce down to search.s = #CRLF$
(again, I am using PB6.20)

The problem seems to be the String you read from the file.

The Ryzen processor somehow limits the load on the processor. I remember that on Intel, an incorrect loop could load the processor to 100%, so I always had a processor sensor in the tray. While experimenting with the CLCL program, I noticed that the processor was only loaded by 8%. This was strange ...

Hi Randy, your code is a good example for don't do that!

Never try to toggle ON/OFF with the binary not! Always use Bool Not.
This is a big differernce.
The binary Not toggles all bits.
The bool Not toggles between #False and #True.

I modified your code to show the Problem.
If I set first t ...

I downloaded source of V 1.01, Compiling with PB6.21!

There are some issues with the SourceProject File.
Once selected it is nearly not possible to change. I have to Quit first, then it change but not analyze.

Try to analize the PB-IDE Sourcecode. From https://github.com/fantaisie-software ...

Huh? There's AI for that!

Try to convert VB6 or Fortran to PureBasic with AI!
This will change your mind!

here I show some advanced String parsing techniques with univeral CharPointer and the use of my Purebasic Extention Module PX.

- Skip Spaces and Tabs
- Find end of an Expression or Word and get the length
- Check next character

Please download the additional PX module here
https://github.com ...

Basically you need to write a compiler, since you don't just parse the syntax, you also need to split it into code sections. I think it's better to use character by character reading for the syntax parser, so you can easily modify the parser behavior during the reading process.
You need to do what ...

Here the Link to the Project Oberon Book form Niklaus Wirth, Jürgen Gutknecht of ETH Zuerich.

https://people.inf.ethz.ch/wirth/ProjectOberon1992.pdf

The compiler description with Code start at page 266 (12. The Compiler)

My question is: what do you think is the most efficient approach for this kind of parsing?

Regex – fast for simple patterns but can become messy with more complex syntax.
Lexer + Parser – reading character by character, generating tokens, and then interpreting them via syntax analysis.
AST ...

here is my Version of SplitStringArray and SplitStringList inspired by mk-soft version with DQuote.
I post it here, because I used a different way of implementation.
It supports Multicharacter Separator with DQuote at a very high speed.

I used a few special tricks to get this done.
- Prototype the ...