Follow-up Comment #5, bug #67408 (group groff): [comment #4 comment #4:] > My used script was a mess.
That's not all that was. > The culprit is the ".if !..." Yes; your finding is duplicative of my comment #3. > These redefinitions of delimiters are against productivity, against > evolution. This is a meaningless assertion. > They also reduce the freedom of choice. This, too, is nonsense. > They also break the mission statement of groff. You make no argument how, but I'm not interested in hearing from you about it, so you need not elucidate your thinking now. > They should all be removed. That's not a correct conclusion. 1. For "compatibility mode" to fulfill its promise, it needs to handle delimiters in the context-sensitive manner--which is not in any way described by CSTR #54--that Ossanna and Kernighan _troff_s manifested. 2. For delimiters to work using the "input level" semantics innovated by GNU _troff_, they need to be consistent and (relatively) context-free, not varying by the type of expression being evaluated (numeric, "string", or formatted output comparison expression). > GNU groff should not be made to be both old and new. _groff_'s compatibility mode is a 35+ year old feature and I have no appetite to remove it. You can feel free to do so in your fork. > [The Unix philosophy: "Small is beautiful", "Make each program do one > thing well", and so on] Take this argument to Kernighan. > The compatibility mode is abused. I see no evidence of that. There is, however, demand for it. https://www.tuhs.org/pipermail/tuhs/2025-September/032467.html > It was introduced, when new requests were introduced with longer names > than two characters. Yes. > Old troff allowed skipping a space between a request and its argument. Yes. GNU _troff_, and possibly _sqtroff_ before it, changed the syntax of the language. Your comments on this ticket are not helpful. Please refrain from making more. _______________________________________________________ Reply to this item at: <https://savannah.gnu.org/bugs/?67408> _______________________________________________ Message sent via Savannah https://savannah.gnu.org/
signature.asc
Description: PGP signature
