Count string in string

[7erra 629]

Hello everyone,

 

I wanted to count the linebreaks "<br/>" inside of another string but couldn't find a way. Example: "My first line<br/>My second line<br/>Third line<br/>you guessed it, fifth line<br/>I lied that was the fourth." should return 4. With toString [10] it would be easier since you could use

_lineCount = {_x == 10} count (toArray _string);

but not with parts that are longer than one character.

[code34 248]

hi

 

could lead to good idea :D

 

https://limpet.net/mbrubeck/2014/08/11/toy-layout-engine-2.html

[Muzzleflash 111]

How about this (untested):

 

CountNonOverlappingOccurrences = {
  params ["_array","_pattern"];
  private _patternSize = count _pattern;
  if (_patternSize == 0) exitWith {0};
  private _result = 0;
  // Let _i denote the start of potential match in _array
  for "_i" from 0 to (count _array - patternSize) do {
    scopeName "main";
    // Let _j denote the number of characters matched so far.
    private _j = 0;
    while ((_array select (_i + _j)) isEqualTo (_pattern select j)) do {
      _j = _j + 1;
      if (_j == _patternSize) then {
        //Got a match. Move next search to after current match
        _i = _i + _j;
        _result = _result + 1;
        breakTo "main";
      };
    };
  };
  _result
};

private _lineBreaks = [toArray _string, toArray "<br/>"] call CountNonOverlappingOccurrences;

 

[Lucullus 71]

 

Another way:

_str = "My first line<br/>My second line<br/>Third line<br/>you guessed it, fifth line<br/>I lied that was the fourth.";
_lines = count (_str splitString "/");

hint str (_lines); // 5

 

[das attorney 858]

I would do:

 

_manipulate = "My first line<br/>My second line<br/>Third line<br/>you guessed it, fifth line<br/>I lied that was the fourth.";
_manipulate = _manipulate splitString "<>";
_manipulate = _manipulate select {_x find "br/" == -1};
diag_log _manipulate

 

Should return this:

 

["My first line","My second line","Third line","you guessed it, fifth line","I lied that was the fourth."]

Then you can do all the endl stuff with forEach to use it in a tooltip or whatever.

[code34 248]

wow nice :) could lead to encapsulte that in parser object

[Muzzleflash 111]

The only thing to be careful of with  splitString is that it will split on character in the delimiter. So if you split on "<>" you are fine with what @das attorney suggests, since you text will likely not contain those symbols. But if it did, you would be in trouble.

[7erra 629]

1 hour ago, Muzzleflash said:

The only thing to be careful of with  splitString is that it will split on character in the delimiter.

That is the reason why I avoided using splitString but @das attorney's method is really simple and fast.

 

Here are the test results:

My own solution, worst of all:

Spoiler

_text = "Line<br/>";
for "_i" from 1 to 100 do {
	_text = _text +format ["Line %1<br/>",_i];
};
_tCount = count _text;
_duration = _tCount/20;
_lineHeight = 0.0607143;
_br = ["<br/>","<br />"];
_10 = toString [10];
_editString = _text;
{
	_char = count _x;
	_no = _editString find _x;
	private _loop = 0;
	while {-1 != _editString find _x} do {
		_no = _editString find _x;
		_splitStr = _editString splitString "";
		_splitStr deleteRange [(_no +1), _char -1];
		_splitStr set [_no, _10];
		_editString = _splitStr joinString "";
	};
} forEach _br;
_lineCount = {_x == 10} count (toArray _editString);
systemChat str _lineCount;

_deltaY = _lineCount;
[
	_text,
	-1,
	0 * safezoneH + safezoneY,
	_duration,
	1,
	_deltaY,
	789
] spawn BIS_fnc_dynamicText;

/*
	Result:
	13.6622 ms

	Cycles:
	74/10000
*/

 

 

@Muzzleflash's solution:

Spoiler

_text = "Line<br/>";
for "_i" from 1 to 100 do {
	_text = _text +format ["Line %1<br/>",_i];
};
// _text will later on be the input text
_tCount = count _text;
_duration = _tCount/20;
_lineHeight = 0.0607143;
_br = ["<br/>","<br />"];

_textArray = toArray _text;
_lineCount = 0;
{
	_brArray = toArray _x;
	private _patternSize = count _brArray;
	if (_patternSize == 0) exitWith {0};
	// Let _i denote the start of potential match in _textArray
	for "_i" from 0 to (count _textArray - _patternSize) do {
	scopeName "main";
	// Let _j denote the number of characters matched so far.
	private _j = 0;
	while {(_textArray select (_i + _j)) isEqualTo (_brArray select _j)} do {
		_j = _j + 1;
		if (_j == _patternSize) then {
			//Got a match. Move next search to after current match
			_i = _i + _j;
			_lineCount = _lineCount + 1;
			breakTo "main";
		};
	};
	};
} forEach _br;

systemChat str _lineCount;
_deltaY = _lineCount;
[
	_text,
	-1,
	0 * safezoneH + safezoneY,
	_duration,
	1,
	_deltaY,
	789
] spawn BIS_fnc_dynamicText;

/*
	Result:
	5.85965 ms

	Cycles:
	171/10000
*/

I edited some script errors and removed the call necessity.

 

And the best of them all:

@das attorney's script:

_text = "Line<br/>";
for "_i" from 1 to 100 do {
	_text = _text +format ["Line %1<br/>",_i];
};
_tCount = count _text;
_duration = _tCount/20;
_lineHeight = 0.0607143;
_br = ["<br/>","<br />"];

_manipulate = _text;
_manipulate = _manipulate splitString "<>";
_lineCount = {_x find "br/" != -1 OR _x find "br /" != -1} count _manipulate;
systemChat str _lineCount;

_deltaY = _lineCount;
[
	_text,
	-1,
	0 * safezoneH + safezoneY,
	_duration,
	1,
	_deltaY,
	789
] spawn BIS_fnc_dynamicText;

/*
	Result:
	0.499251 ms

	Cycles:
	2003/10000
*/

The splitString won't be a problem since im only counting "br/" and "br /".

 

Thanks everyone :)