Numbering bug in "Runtime Semantics: IsWordChar Abstract Operation"

[jorendorff]

The table should be inside step 3. Since we don't figure that out, step 4 is treated as a new list, and gets the number 1.

[allenwb]

in rev24 I made the last item explicitly a new list that starts at 4. Maybe that will help.

[jorendorff]

in rev24 I made the last item explicitly a new list that starts at 4

Thank you!

This shouldn't affect the script either way, if the appearance is unchanged.

The script tries to rely on how things look, not on the, er, theoretical structure of the Word document. So it should generate the same markup whether the last item is a new list or not.

(Once upon a time the script did depend on Word's notion of where lists begin and end. That didn't work very well. The generated markup was really bad, and getting all the margins and spacing to work out to hide that fact was beyond me.)