The name of the function is complicated, so functionList should return complicated as the name.

You can see it in action here:https://ideone.com/SmDmCF

It doesn’t do anything useful since it is an example.

Here is another example:

float ( *GetPointer( int a ) )[10]
{
( void )a ;
return NULL ;
}

This is a function named GetPointer, taking an int and returning a pointer to an array of 10 float.
You can see it working here: https://ideone.com/Sxn8WQ

I intend to learn C/C++ language, someday. But, I presently know very few, about these languages !

However, in your previous post, didn’t you forget the name of the function ? I mean something like :

void D( *complicated( int a ,...., instead of void( *complicated( int a ,....

In case of a negative answer, can you tell me what name the functionList should return, in that specific case ?

Regards,

guy038

It is hard to test for false positives. Thus far I haven’t seen any.

I have found an example that is a valid C function but it isn’t detected by your regex:

void ( *complicated( int a , void( *f )( int ) ) ) ( int )
{
void( *b )( int ) = NULL ;
return b ;
}

This is a function taking an int and a pointer to a function taking an int and returning nothing, returning a pointer to a function taking an int and returning nothing.

This is not a common occurrence and actually writing such code is obnoxious. Nevertheless, can you make this work? :-)

The most important thing is that my regex version doesn’t break anything else !!

As you’re certainly more acquainted with C/C++ programs than me, it shouldn’t be very difficult, to you, to detect any additional problem :-)

Cheers,

guy038

Thank you for your reply. Your version does seem to work.

Maybe this change could be added to future versions of Notepad++? Implementers should consider that C++ can have the same function definition thus that regex should be changed as well.

Ricardo, you’re right ! the problem comes from the expression [^\)\(]* that means any sequence, even empty, of characters, different from round brackets

To match the special case of bri235445 void A( void( *call )( void ) ), I changed that previous class character into [\S ]*

Remember that the syntax \s is equivalent to the syntax [\t\n\x0B\f\r\x20\xA0]

=> The syntax \S, with the uppercase letter s, means [^\t\n\x0B\f\r\x20\xA0]

So, my regex [\S ]* matches any sequence, even empty, of a NON blank character OR a space

Then, knowing that :

• The syntax \w does match the underscore _

• The syntax [\s] can be simplified in \s

=> The mainExpr regex of the C parser, with my modification, could be simplified in :

^[\t ]*((static|const|virtual)\s+)?[\w:]+(\s+\w+)?(\s+|\*\s+|\s+\*|\s+\*\s+)(\w+\s*::)?(?!(if|while|for))\w+\s*\([\S ]*\)(\s*const\s*)?\s*\{
   

Thus, bri235445 :

• Do a backup copy of the functionList.xml file

• Open the functionList.xml file, in N++

• Change, between the double quotes, the mainExpr regex, of the c_function parser, with the regex just above

• Save the modifications. Then close and restart N++

=> You should notice that you get the functions A and B in the Function List panel, with your text, in C language, that follows :

void B( void )
{

}

void A( void( *call )( void ) )
{

}

I tested with a portable v6.7.9.2 version and everything works fine :-)

Best Regards,

guy038

P.S. :

Ricardo, your modification (.*) was nearly good :-)

But it’s very important to remember that the Function List behaves, by default, as the DOT stands for ANY character, even the End of Line characters !

So, to restrict the matches of a dot to standard characters, only, the right syntax should be (?-s).*. I just suppress the round brackets, surrounding .*, as we don’t need any back-reference \1

It passes the test, too. However, I think that the syntax [\S ]*, being more restrictive, should be preferred

	<function
	    mainExpr="^[\t ]*((static|const|virtual)[\s]+)?[\w]+([\s]+[\w]+)?([\s]+|\*[\s]+|[\s]+\*|[\s]+\*[\s]+)([\w_]+[\s]*::)?(?!(if|while|for))[\w_]+[\s]*\([^\)\(]*\)([\s]*const[\s]*)?[\n\s]*\{"

Imo, the bad part is this: \([^\)\(]*\)
It means:

\( matches the character ( literally

[^\)\(]* match a single character not present below (* Between zero and unlimited times)
    \) matches the character ) literally
    \( matches the character ( literally

\) matches the character ) literally

But… I tried modifying with \((.*)\) so it matches anything inside parens, but it does not work in Np++…

I also would like to know a solution.