>>>>> "Dov" == Dov Feldstern <[EMAIL PROTECTED]> writes:
Dov> Jean-Marc Lasgouttes wrote: >>>>>>> "Dov" == Dov Feldstern <[EMAIL PROTECTED]> writes: >> Dov> The cause is what Elazar explained: the piece of code which he Dov> patched is the one moving the cursor to the end of inset, by Dov> subtracting the inset's width. This subtraction *should* take Dov> place inside the inset, because inside it we want the cursor Dov> position to be calculated relative to *left* edge of the inset; >> Actually, I do not understand the original code from Martin. I do >> not understand why the fact that the character at cursor is an >> inset changes anything. Dov> This I also don't fully understand. But this code is necessary Dov> (see below). I mean, what's necessary is to know when you're Dov> *inside* an inset. I would tend to think that _some_ of the callers to cursorX need to apply Martin's test by themselves, but that the code does not belong in the method itself. However, I am just waving hands, I do not have time to look at the code long enough. JMarc
