Work with filestrings read by BaseFile

[Helper]

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 01/12/2005 at 06:03, xxxxxxxx wrote:

User Information:
Cinema 4D Version:   7 XL 
Platform:   Windows  ;   
Language(s) :   C.O.F.F.E.E  ;

---------
Hello out there,
I am quite new on coding C.O.F.F.E.E. To start my lessions on filescripting, I am coding a menuplugin to import OBJ-files.
Unfortunately C.O.F.F.E.E, thus plattform independent, is horrible slow (I think because it is an interpreter, although compiling did not make it faster).
Here is the way the file is read:
I read the file into a long string with BaseFile->Readstring. Then I am trying to splitt this string linewise and access data in the line.
Therefore I am searching CRLF-signs in te string with the strstr-function. I got some tipps from other users, that this function is very slow. (Hello Maxon: Is it like this and if, why?)
Does anybody have a tuning-trick to speed up string-operations like these in C.O.F.F.E.E., perhaps using other functions than strstr/strchr?
Or does anybody have a tipp for me, how reading a file could be done quicker?
As I am not very familiar with VC++ and because it would no longer be plattform-idependent I would prefer not to recode in C++-SDK (even if it runs much faster then).
Thanks for help everybody,
COFFEJUNKIE
Ps: Sorry for my English, I did not speak or write it for longer now...

[Helper]

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 10/12/2005 at 04:57, xxxxxxxx wrote:

I don't know about the performance, but this is a function that I have used to extract a line in C.O.F.F.E.E.:

    
    
    // Remove a line from data and return it  
    // Must be called by reference: getline(&data)  
    getline(data)  
    {  
      var pos = strstr(data, "\n");  
      var line = strmid(data, 0, pos-1);  
      data = strmid(data, pos+1, sizeof(data));  
      return line;  
    }

This is an example of how it's used:

    
    
    main()  
    {  
      // Open file for reading  
      var f = new(BaseFile);  
      f->Open(file);
    
    
    
    
      // Read the whole file  
      var data = f->ReadString(f->GetLength());  
         
      // Read the number of objects  
      var max_index = evaluate(getline(&data));  
        
      // Loop through objects  
      var i;  
      for (i = 0; i < max_index; ++i)  
      {  
        var obj = getline(&data);  
        if (i == index)  
        {  
          // Output the name of obj number 'index'  
          name = obj;  
        }  
      }  
      ...  
    }

[Helper]

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 16/12/2005 at 06:48, xxxxxxxx wrote:

Hello Mikael,
 
first of all, thanks for your answer. There is only one question I have and do not understand in your code:
Your code to remove the extracted line from the datastring is
data = strmid(data, pos+1, sizeof(data));

but I think you have to reduce sizeof(data) by the amount of removed characters like
data = strmid(data, pos+1, sizeof(data)-pos);
or is this wrong? And if so, why (perhaps because of possible unicode?)
Best greetings,
COFFEJUNKIE

[Helper]

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 17/12/2005 at 10:27, xxxxxxxx wrote:

That's just me being lazy. The docs said that strmid ignored the size going beyond the length of the string, so I didn't bother thinking of what to subtract (pos, pos+1 or pos-1). Your version is better, at least if you've chosen the right amount to subtract...