Allow special chars as word seperators

[ghost]

... numbers are also "words" ...

[calleluks]

Hey @DBOTW, thanks for the PR!

The implementation looks good to me.

I haven't been able to find a standard definition of the ALTWERASE mode. The GNU libc documentation mentions also counting underscores as word characters and it seems like the handling built into linux also does this: https://github.com/torvalds/linux/blob/2acf097f16abba684012cca670a61d94178bd1ab/drivers/tty/n_tty.c#L998

Do you know more about ALTWERASE, @mptre?

[ghost]

For me an underscore seperates words (like a "visible blank"), I don't care about ALTWERASE ... ;-)

[mptre]

The OpenBSD kernel also treats underscore as a word character. I would
propose us doing something as follows:

#define	ISWORD(c)	(isalnum((c)) || (c) == '_')

The macro is not necessary if the line with the conditional doesn't get
excessively long.

[mptre]

The manual must be updated as well:

diff --git pick.1 pick.1
index 0d5104e..efc1c84 100644
--- pick.1
+++ pick.1
@@ -74,8 +74,10 @@ Delete one character to the left of the cursor in the search query input field.
 .It Ic Delete | Ctrl-D
 Delete the character under the cursor in the search query input field.
 .It Ic Ctrl-W
-Delete to the beginning of the closest word to the left in the search query
-input field.
+Delete the word to the left of the cursor.
+A word is recognized as a sequence of characters for which either
+.Xr isalnum 3
+is true or the character is an underscore.
 .It Ic Ctrl-U
 Delete to the beginning of the line in the search query input field.
 .It Ic Ctrl-K

[mptre]

@DBOTW are you up for the task of finishing this PR with respect to the
feedback above? Otherwise, no worries; I will go ahead and fix it.

[ghost]

Of course @mptre ... done ...

[calleluks]

Thanks, @DBOTW! This looks good to me.

[mptre]

This have to wait until v1.8.0 is out. I would also like to make
some final adjustment.

[mptre]

Closing in favor of #234.