[next-master] Question: can we improve the system of dealing with unencoded open/close punctuation?

[arrowtype]

Currently, the script seems to check for unencoded open/close punctuation, but only if it specifically has a name using the .uc suffix. However, there are plenty of potential names that could fall outside of that. For one, .case might be another logical suffix for case-specific punctuation, but then there also might be any other potential reasonable suffixes on punctuation alts.

MM2SpaceCenter/MM2SpaceCenter.roboFontExt/lib/MM2SpaceCenter.py

Lines 508 to 522 in 3b04a29

	openCloseUnencodedPairs = {
	"parenleft.uc": "parenright.uc",
	"bracketleft.uc": "bracketright.uc",
	"braceleft.uc": "braceright.uc",
	"exclamdown.uc": "exclam.uc",
	"questiondown.uc": "question.uc",
	"guilsinglleft.uc": "guilsinglright.uc",
	"guillemotleft.uc": "guillemotright.uc",
	"guilsinglright.uc": "guilsinglleft.uc",
	"guillemotright.uc": "guillemotleft.uc",
	"slash": "backslash", #should be encoded but adding here because those aren't working for some reason
	"backslash": "slash", #should be encoded but adding here because those aren't working for some reason
	}

I’m making a note of this as something to potentially look into after #26.

[benkiel]

Maybe you don't know about this? robotools/defcon#391 Also, you can use pseudo unicode: split at . see if the first thing has a unicode, use that.

[benkiel]

To be clear, I'd use something to make a pair list of open/close: BIDI may be good there, then make a mapping file by splitting the suffixes to map to the unicode encoded version, then you can just do a lookup to get the right open/close

[cjdunn]

@benkiel that's a great suggestion! I‘m not going to be working on this for a bit, but @arrowtype this seems like it would be helpful if you're going to keep working on this feature. Thank you both!

[arrowtype]

@benkiel thanks so much for pointing this out!

you can use pseudo unicode: split at . see if the first thing has a unicode, use that.

I think that Wei’s suffix handling feature does essentially this, so it might just be a matter of adapting/extending that to work for open/close punctuation, as well.

And then, using BIDI would probably be a big improvement over our current, simplistic way of just listing a bunch of potential open/close punctuation (which is almost certainly not as comprehensive as BIDI).

[arrowtype]

As a note: I did check whether .case punctuation is handled with the current MM2SC version (0.3.0), and it is not.