#R11 Xref (entmod)??
5 messages in this thread
Well, I'd like to try to convince you, anyway. Sure, not every user or
discipline will have the need to change the properties of specific xref'd
entities, but some of us do. It's common enough for us to run into
situations where we have entities in certian drawings which can have lables
like "proposed-existing" or worse used to describe them. It won't happen in
new construction, usually, but can easily happen when improvement work is
being done on a parking lot, road, or highway. The need to show one
particular stretch of curb & gutter as a solid linetype in one construction
detail drawing (for purposes of clarity, or other reasons), even though it
really only one part of an overall design otherwise, is difficult to deal
with now. I wouldn't want to use the capability on a large-scale basis, but
the ability to change the appearance (or even presence of) a few selected
entities in an xref'd drawing would definately be ideal for such cases. It
modifies the appearance in only that one target drawing (and perhaps even
only one viewport), and the alternatives are not pretty. Another way may be
the ability to simply control visiblity of specific xref'd entities, and
then to draw on top of it with the correct color & linetype, but having to
duplicate those entities fragments the design database somewhat, and may
have other drawbacks. The only alternatives available now are to return to
the source file, add a new layer, and move the entities to that layer. And
then you still have the VISRETAIN problem to deal with (gosh, I hope that
get's soved in R12!), and possibly other problems related to VPLAYER on/off
settings for new viewports (?).
I really don't see it as such a major extention, but rather a needed
enhancement. Nor do I see it as defeating the purposes of xref's. In fact, I
think my example shows how it could make them much more flexible and
usefull. Just my $.02*inflation=???
Back to your side of the court…
Peter:
I would say it comes down to drafting convention – I put all of those
things onto separate layers, and therefore a layer-specific rather than
entity-specific approach solves these problems for me. I understand what
you're saying and attempting to do – the solution *I* want is different
because my use is different.
Larry
I think both conventions are valid.
When it's all done with layers, one's layering standard can become
convoluted. If I've got several groups of entities that need to have
different appearances in other drawings, then I am forced to have one
additional layer for each group. This can really make a layering standard
somewhat confusing, and just feels to klutzy to me. Perhaps a better
extention would be to move to something like the nested layers that were
spoken of here before?
What I like about entity specific properties on xref'd data, is that it is
so specific to that one drawing, and doesn't affect anything in other parts
of the drawing, or other sheets for that matter. Also, if you have to deal
with something like the UK layering system, imagine the headaches that might
cause….
Perhaps it is worth waiting for something like ObjectCad, though. It might
be that much more natural in such an environment.
I should also stretch, if it isn't clear enough, that I'm not talking about
large scale individual entity changes; I think that would be entirely
inappropriate, and would suggest to me that my layering scheme was not deep
enough in it's level of detail.
Just some food for thought…
Peter:
>> When it's all done with layers, one's layering standard can become
convoluted. If I've got several groups of entities that need to have
different appearances in other drawings, then I am forced to have one
additional layer for each group. <<
Well, if they need to appear different from each other within one drawing,
I'd say they are different entity types! Face it – a sprinkler line needs
to be shown solid in some plans, perhaps dashed in another, and hidden in
yet others. However, that sprinkler line should *not* need to be shown as
part dashed and part hidden in one drawing, unless there is an issue such
as demolition/new work/existing to remain, in which case each of these
should be placed on their own layer!
Nested layers – yes, I agree. However, over a year ago I developed my own
quasi- nested layering, and now I don't really need an internal nesting
scheme. The wildcard characters make this fairly easy, with a little
foresight and planning, followed by strict adherence to the standard.
Speaking of the UK standard… I used that as the basis of *my* standard.
The "food" you send is interesting, and please keep in my mind that I'm
only speaking for myself – pass the salt? <g>
Larry
Peter,
Having that kind of access to the XREF is rather self-defeating. You may as
well just inport the drawing as a block. You should have control over the
layers though as far as turning them on or off, freezing, thawing, color
and linetype through the layer command.