<div dir="ltr"><div>Well, except as you probably saw in the diffs, that destX := destX + kern, was omitted in space...<br></div>Since it no more is when we remove space, then composition space must be changed too...<br></div>
<div class="gmail_extra"><br><br><div class="gmail_quote">2013/10/5 Nicolas Cellier <span dir="ltr"><<a href="mailto:nicolas.cellier.aka.nice@gmail.com" target="_blank">nicolas.cellier.aka.nice@gmail.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div><div><div><div><div>I have recently promoted a uniform handling of space in DefaultStopConditions.<br>
</div>After thoughts, and after removing some cruft, I see a bit clearer: this is a wrong decision.<br>
</div>Indeed, there is no special handling required in any CharacterScanner but the CompositionScanner.<br></div>The consequence of adding this stopCondition is that we stop the inner scan loop more often that required, at each word rather than at each line for uniform text attributes.<br>
</div>This is bad for slow machine and goes against the goal of tim to have a faster and cheaper scan loop.<br>So I will retract.<br></div><br></div>
</blockquote></div><br></div>