Follow-up Comment #3, bug #66438 (group groff):
At 2024-11-14T06:28:42-0500, Sven Schober wrote:
> Follow-up Comment #2, bug #66438 (group groff):
>> Unfortunately, the `\{1,\}` solution you propose is also not a POSIX
>> BRE.
>
> Hmm, okay, I did not read the POSIX spec/standard (yet?).Issue 8, which came out this year, is around 4,000 pages. Few people read every word. For $reason$ involving ISO and/or IEC and/or IEEE, the Austin Group does not freely distribute the released version, and even the drafts are behind a registration wall. https://www.opengroup.org/austin/login.html However, an HTML version is open to all. https://pubs.opengroup.org/onlinepubs/9799919799.2024edition/ Also, we try to keep groff portable to some pretty old systems. Here's some advice from our "HACKING" file in the source distribution. ---snip--- * Write to the POSIX standard for the shell and utilities where possible. Issue 4 from 1994 is old enough that no contemporary system has a good reason for not conforming. A copy of the standard is available at the Open Group's web site. https://pubs.opengroup.org/onlinepubs/009656399/toc.pdf ---end snip--- Issue 4 was split into multiple PDFs, so to get its definition of BREs, you need the separate system interface definitions (a.k.a. "base definitions", "XBD") document. https://pubs.opengroup.org/onlinepubs/009656599/toc.pdf Find the definition of a BRE starting on page 100, PDF page 120. And for completeness, here's the Issue 4 System Headers and Interfaces ("XSH") document. https://pubs.opengroup.org/onlinepubs/009656499/toc.pdf > But this whole basic vs extended RE business is really unfortunate in > my opinion. I always struggle, everytime I write an expression. As I understand it, egrep came after grep (and ed(1)), and nobody wanted to break backward compatibility with anything, so here we are. > And, to be honest, I misinterpreted the man page of my mac os/bsd(?) > `expr` as if the bounds notation `\{1,\}` _would_ be BRE. No worries. I think macOS gets most of its Unix from FreeBSD, so you might file an issue with FreeBSD, and 10-15 years after they fix it, macOS will incorporate the change. :-| >> I've have to respell the pattern using a doubled atom and the * >> quantifier. > > You mean, soemthing like this hint from the re_format man page? > > "Also note that ‘x+’ in modern REs is equivalent to ‘xx*’" Exactly. >> Also, 10 digits of Git hash ought to be enough for anybody. > > Haha, yes, sorry. I just double clicked on the hash and pasted it. It's fine. :) That was just my way of noting that the revision to the Summary line was deliberate. _______________________________________________________ Reply to this item at: <https://savannah.gnu.org/bugs/?66438> _______________________________________________ Message sent via Savannah https://savannah.gnu.org/
signature.asc
Description: PGP signature
