Few do's, don'ts and gotchas
rtLib can be used to succesfully resolve most of the problems connected to rich text manipulation. Still there are some considerations and lessons:
1) use rtLibSession.recycleAll in terminate statement of your code
2) as the amount of rich text increases rtLib slows down. If you need to process large volume of text or tables and you may not entrust the task to background agent, make sure to check the response time. The reading speed of tables have been significantly improved starting with version 1.1b, but it may still be slow for some applications
3) do not use Delete statement to delete user objects of rtLib in order to release memory. Though experience varies, the effect may be the same as when encountering an error - the unload of an agent takes much longer than normally.
Some other notes:
As a rule building of objects from scratch takes much less time than reading from rich text item. The processing speed depends little on amount of manipulation and is determined primarily by the number of internal records of the rich text item. Thus processing of item containing only images works much quicker than that of the same size if it contains text or tables.
Apart from direct use in manipulation of Rich Text, rtLib may be used as HTML generator. In this case it works extremely quickly as it has not need to read/write rich text.
When generating large simple tables always try to take advantage of "quick" tables.