OK, fixed cosmetic issues. Yes, that case is intentional. When the break line is in the middle of a word, we normally scan backwards to try to find the beginning of that word, and then we break there. But if the beginning of the word is in fact at the beginning of the current break line, then that means we have a super-long word that spans the entire viewport, and we must "give up" and break _after_ the word instead, beyond the preferred break offset (requiring horizontal scrolling in the editor). This case is exercised many times when AdjustBreakOffsetToWordTest is run (I just checked to be sure).
[ Full content available at: https://github.com/apache/incubator-netbeans/pull/598 ] This message was relayed via gitbox.apache.org for [email protected]
