Re: pruning "redundant" properties while editing


Subject: Re: pruning "redundant" properties while editing
From: Paul Rohr (paul@abisource.com)
Date: Tue Mar 13 2001 - 10:30:42 CST


At 10:35 AM 3/13/01 -0500, Randy Kramer wrote:
>If we delete redundant properties, especially at edit time, I'd like to
>request that we make sure the undo mechansim can undo that deletion. If
>unlimited undo of this is unwieldy, I request that we at least allow
>such undo for some reasonable period.

Our current undo mechanism is designed to work exactly the way you want.
You can revert to any prior state of the document *during this editing
session* by hitting undo long enough. (Note for the finicky -- abandoned
forward branches are pruned just like they are in back/forward navigation in
web browsers, so the word "any" depends on this notion of "prior".)

In particular, any format change is represented as a swapped pair of
attributes and properties (old vs. new). The proposed change is to condense
the "new" set. It definitely should not affect the "old" set.

Note, however, that you will no longer be able to undo once you close or
reload the document. We don't have a brittle so-called "fast save" option
to persist all that undo information to disk. This is, IMNSHO, a Good
Thing.

>On more than one occasion I've done dumb things like accidentally apply
>a new style to one or a range of paragraphs by mistake. If redundant
>properties are deleted at edit time, there may be no way to undo the
>style application and return to the document condition before the
>accident. In many cases this is not a significant problem, but if the
>paragraph contains individual words in bold or italic, and the
>accidental style applied is, for example, bold, the information about
>which words were in bold or italic is gone. It can be very frustrating.

Please reread my original post. According to the algorithm I'm proposing,
the only way to get back to the state you want is via undo. I'm definitely
interested in look&feel feedback about the other detailed behaviors when you
go forward *without* hitting undo.

Paul



This archive was generated by hypermail 2b25 : Tue Mar 13 2001 - 10:53:29 CST