Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
>>>>>> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
>
| Lars> I won't with -1 or +1 either :-), you need to check first span
| Lars> either way.
>
| start is initialized to 0, which is the start of the first span.
>
| I changed -1 to +1 in the patch.
>
| | No idea. I left it as it was. I can assert on it if you want.
>
| Lars> Nah...
>
| OK.
>
| | Lars> string str(1, par_.getChar(pos);
| | Why is it better?
>
| Lars> do the work on initialization... only that.
>
| I used push_back. Is that OK with you?
the you should just leave it as is (+=). a push back buys nothing.
| | I am not sure I like it... Doesn't the compiler give warnings
| | on that?
>
| Lars> Why?
>
| Actually, I do not like it because both branches of the if() will be
| compiled and in many cases it will not work (unavailable function).
| Moreover, it is not clear by reading the code that one part is _never_
| executed.
But the compiler will know and do the right thing.
>
| Lars> I can live with
>
| Lars> #if USE_LYX_FONTCACHE
>
| Lars> as well.
>
| I used #if defined(USE_LYX_FONTCACHE). Is that OK? I prefer to make it
| clear that USE_LYX_FONTCACHE is a boolean value and not an int.
And that is done with #if USE_LYX_FNTCACHE
with defined(...) you only say this this is some macro...
And you cannot seet USE_LYX_FONTCACHE to 0 then, because that would
trigger...
--
Lgb
