BFC of mirrored parts?

[Magnus Forsberg]

In the list of Non-BFC'd parts there is a lot of parts that only need a correct BFC statement.
Parts like physical coloured shortcuts or mirrored versions of a design (left/right).

Two of the Technic Panels Farings, 32189 and 32535, have been fixed, but we also need to add a BFC statement to the mirrored parts, 32188 and 32534. All that is needed is to add " 0 BFC CERTIFY CCW " to them, (and change from Part to Unofficial_Part).

Many files was caught in Rolands libfix update, and are now at the PT, but all the mirrored parts are still left behind in the official library and needs to be sent to Admin for recycling.

Is there a way to find them and/or correct them automagically?

[Steffen]

I think those should simply be cycled through the PT as usual.

[Magnus Forsberg]

Could it be that mirrored files don't need to be BFC'd?
I didn't check the mentioned files, 32188.dat and 32534.dat, until today,
and to my surprice they look OK even without a correct bfc line like "0 BFC CERTIFY CCW"

They seam to 'inherit' the properties of the parent part.
Is it enough that (all) the 'sub'files are made BFC, and that means that we don't have to make all files BFC?

[Max Martin Richter]

Basically I would say, that you are right. That's the reason why you can see some older parts in a Blue/Green mix, while checking the BFC status visually.

But I would really prefer to have a BFC statement in ALL parts. Otherwise we would have a new exception in our rules.

/Max

[Travis Cobbs]

If a file is determined to be a part, its "parent BFC" state is forced to be treated as if it were "CERTIFY", so if the part itself is certified, it doesn't matter whether its parent is certified. This is covered in the third paragraph of the Language Extension Functionality section of the BFC spec. The actual examples given there don't include uncertified mirrored parent parts, but the general wording of that paragraph does make it clear that BFC should be enabled in this case.

[Magnus Forsberg]

And in plain English?
All files must have a BFC statement, even mirrored parts, part aliases, physical colour shortcuts, and so on. Right?

Could someone help us find those files?
Find all files, without a BFC statement, only containing a single type 1 line.

[Chris Dee]

It's not just those with a single type 1 line, but any file that containing only type 1 lines can have an arbitrary BFC statement inserted.

[Travis Cobbs]

In plain English, the mirror files should probably contain a BFC statement for completeness, but adding such a statement should have no impact on the results with BFC-compliant renderers.

[Travis Cobbs]

This is not true. It's perfectly reasonable to have a file that only contains type 1 references to primitives, and no local geometry. Such a file would potentially need INVERTNEXT statements.

[Steffen]

Yes.
Some tools might or might not add additional logic to infer BFC status for certain files.
We still should add BFC CERTIFY statements where missing.