Follow-up Comment #3, bug #68651 (group groff): I have a new stage to append.
7. Drop recognition of numeric parameters to the `\O` escape sequence that are made unnecessary by implementing the earlier stages. For now, I'm pretty confident that means ditching the troublesome, esoteric, and syntactically bizarre `\O5`--I have no idea why the escape sequence parameter is read as a `symbol`, meaning it has the syntax of a GNU _troff_ identifier. Other escape sequences, like GNU-style special character escape sequences `\[e aa]`, or parameterized string interpolations like `\*[str foo bar]`, for example, don't work that way. Okay, I fib. I have **an** idea why the escape sequence parsing was done this way. It's because the formatter's internals are painfully underdocumented. The internal parser API is also not 100% internally consistent; see bug #68586. Eliminating this syntactical wart from the GNU _troff_ language would be icing on the cake of delivering this redesign. [https://www.gnu.org/software/groff/manual/groff.html.node/Suppressing-Output.html `\O2`, `\O3`, and `\O4` were apparently also implemented solely for _grohtml_ support]. I'm not itching nearly as badly to get rid of those, though simplifying their behavior might be warranted. Having "stackable" output suppression _might_ be something worth preserving. By the time I get to this stage, maybe someone will have thought of a good non-HTML application for them, and can spare them the axe. If not...my guillotine will be ready--slish chop! _______________________________________________________ Reply to this item at: <https://savannah.gnu.org/bugs/?68651> _______________________________________________ Message sent via Savannah https://savannah.gnu.org/
signature.asc
Description: PGP signature
