Get word from mouse position in Edit, RichEdit, IE home page, etc.

[TheBest]

Hello everyone,
First of all, I must thank you for taking your time to review this article.
I want to make the feature like below.

1. move the mouse cursor over edit, richedit or IE page. Then, suppose the mouse cursor is over a word.
2. I want to get the word.

It is just the feature that many dictionary softwares have.
Can you help me how I can do it?
I am a VC++ developer. So I very appreciate if you let me know what APIs I must use. It is great if you can make me get a good sample.

Thank you very much.
Albert :) :) :)

[Astro29]

Even though this thread is 6 years old, for the rest of you searching for getting a word under the mouse pointer of a richedit control, here is the code (C++):

std::string GetWordUnderMouse(int x, int y)
{
	POINT pt = { x, y };
	int ci = SendMessage(hEdit, EM_CHARFROMPOS, 0, (LPARAM)&pt);
	if (ci < 0)
	{
		return "";
	}		
	int lix = SendMessage(hEdit, EM_EXLINEFROMCHAR, 0, ci);
	int co = ci - SendMessage(hEdit, EM_LINEINDEX, lix, 0);
	char buffer[1024] = { 0, };
	((WORD *)buffer)[0] = 1024;
	SendMessage(hEdit, EM_GETLINE, lix, (LPARAM)buffer);
	//search for spaces between words
	int j = co;
	while (j > 0 && buffer[j] != ' ')
	{
		--j;
	}
	int k = co;
	while (k < strlen(buffer) && buffer[k] != ' ')
	{
		++k;
	}
	//build output buffer
	std::ostringstream str;
	for (int i = j + 1; i <= k; ++i)
	{
		str << buffer[i];
	}
	return str.str();
}

This code assumes you have a reference to the richedit control hEdit. X and Y are the LWORD and HIWORD order from the lParam message of WM_MOUSEMOVE

[Astro29]

Actually, after playing with it a bit more, I have a) optimized the loop, b) checked for termination characters and c) if the character under mouse is a space, return an empty string (not a word there). Here's the updated code:

std::string GetWordUnderMouse(int x, int y)
{
	POINT pt = { x, y };
	int ci = SendMessage(hEdit, EM_CHARFROMPOS, 0, (LPARAM)&pt); //current character index
	if (ci < 0)
	{
		return "";
	}		
	int lix = SendMessage(hEdit, EM_EXLINEFROMCHAR, 0, ci); //line index
	int co = ci - SendMessage(hEdit, EM_LINEINDEX, lix, 0); //current character offset/position
	char buffer[1024] = { 0, }; //buffer to hold characters (1024 should be enough for the average line)
	((WORD *)buffer)[0] = 1024; //set first byte to size of buffer as specified in MSDN
	SendMessage(hEdit, EM_GETLINE, lix, (LPARAM)buffer);
	if (buffer[co] == ' ' || buffer[co] == '\n')
	{
		return ""; //currently at a space character or end of line
	}	
	std::ostringstream str; //string stream to hold the resultant word
	int i = co;
	bool forward = false; //direction the loop is searching in
	//search for spaces between words
	while (i < strlen(buffer))
	{
		if (!forward)
		{
			//decrease pointer looking for start of line or space
			--i;			
			if (buffer[i] == ' ')
			{
				++i; //skip over this character
				forward = true;				
			}
			else if (i > 0)
			{
				continue; //not at 0 keep looping backwards
			}
			else
			{
				forward = true; //at 0 so exit this part of the loop
			}
		}
		//increment pointer looking forward for a space		
		if (buffer[i] == ' ' || buffer[i] == '\n')
		{
			break; //break out of the loop
		}
		//append current characters to output stream
		str << buffer[i];
		++i;
	}
	return str.str(); // <- and here is our word
}

Please note: this method only deals with single-byte characters, not multi-byte, so unicode will not work too well with this, however, this can be changed.

[fenrenren]

Cool! thanks for your share!